14:00         4390.     Real Time MR Thermometry for Monitoring Focused Ultrasound in the Liver

Andrew B. Holbrook^1,2, Juan M. Santos^3,4, Elena Kaye^1,3, Viole Rieke¹, Kim Butts Pauly¹

¹Radiology, Stanford University, Stanford, CA, USA; ²Bioengineering, Stanford University, Stanford, CA, USA; ³Electrical Engineering, Stanford University, Stanford, CA, USA; ⁴HeartVista, Inc., Los Altos, CA, USA

We have integrated a high speed, high resolution pulse sequence with referenceless MR thermometry to produce temperature images of high intensity focused ultrasound ablations in real time for therapies in the liver. The sequence was tested in a moving phantom. Maximum temperatures were compared to an identical sonication in a stationary phantom, and the system’s real time reconstruction was also tested. The moving phantom temperature data showed good agreement with the stationary phantom, as did its shape. Finally, in-vivo tests in a patient abdomen were performed to verify that minimal “temperature” rises from phase noise and blood flow occurred.

14:30         4391.     Localized FFT Phase-Correction Algorithm for Improved Real-Time PRF Shift Thermometry

R. Jason Stafford¹, Axel J. Krafft², Roger Jason McNichols³, Ashok Gowda³, Michael Bock², John D. Hazle¹

¹Imaging Physics, The University of Texas M. D. Anderson Cancer Center, Houston, TX, USA; ²Division of Medical Physics in Radiology, German Cancer Research Center (DKFZ), Heidelberg, Germany; ³BioTex, Inc., Houston, TX

MR-guidance of thermal therapy using the proton resonance frequency technique is susceptible to error from background variations in the magnetic field, particularly those caused by motion. Several techniques for suppressing these background errors so that useful thermometry can be performed have been presented in recent years. This work is concerned with a technique that uses iterative Fourier extrapolation into the treatment zone in order to obtain an estimate of the background and remove it.

15:00         4392.     A Fast Thermal Imaging Sequence for Focused Ultrasound Ablation of the Liver

Yuval Zur¹

¹GE Healthcare, Tirat Carmel, Israel

We developed a restricted FOV gradient echo EPI sequence for focused ultrasound ablation of the liver during free respiration. The scan time is ~50 msec with spatial resolution of 1.8 by 1.8 mm. A low SAR b1 insensitive broad band RF pulse is employed to suppress signal out of the FOV. The images are filtered along the time axis and SNR improves by ~2.9. The filtering enables us to acquire up to 3 slices simultaneously, or improve spatial resolution by using the UNFOLD technique.

15:30         4393.     MR-Guided Focused Ultrasound Ablation of the Rat Liver

Randy L. King¹, Viola Rieke², Kim Butts Pauly²

¹Bioengineering, Stanford University, Stanford, CA, USA; ²Radiology, Stanford University, Stanford, CA, USA

>HIFU is being developed for the minimally invasive treatment of primary and metastatic liver cancer. We examined the feasibility of the rat model for the study of HIFU treatment of the liver. Our results show that it is possible to localize thermal lesions to the liver of the rat by sonicating through the ribs with no skin burns. We also demonstrate the ability of MR to thermally monitor and localize the lesions resulting from such a treatment.

13:30         4394.     A Novel Application for MR Thermometry: Post Mortem Interval Estimation in Forensic Medicine

Sara Maria Sprinkhuizen¹, Hedwig A. Tromp², Chris J. Bakker¹, Jessica D. Workum¹, Hendrik de Leeuw¹, Max A. Viergever¹, Lambertus Wilhelmus Bartels¹

¹Image Sciences Institute, University Medical Center, Utrecht, Netherlands; ²Dept. Pathology/Toxicology, Netherlands Forensic Institute, The Hague, Netherlands

The exact time point of death (the post-mortem interval, PMI) is of great importance in the legal and criminological field. Currently, the nomogram of Henssge is used worldwide for PMI estimation. The rectal temperature used as input for the nomogram is a single point measurement, spatially as well as temporally. It is hypothesized that absolute, spatiotemporal temperature data acquired using MRS techniques improves the outcome of the PMI estimation. The goal of this work was to assess the feasibility of MRS based MRT to map absolute temperatures in post-mortem situations.

14:00         4395.     Fast Temperature Measurement Using a 2DRF Pulse Enables Both Reduced-FOV Imaging and Fat Suppression

Chang-Sheng Mei^1,2, Jing Yuan¹, Nathan J. McDannold¹, Lawrence P. Panych¹

¹Department of Radiology, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA; ²Department of Physics, Boston College, Chestnut Hill, MA, USA

During thermal ablation, the region of heating can be much smaller than the image FOV, resulting in waste of scan time to cover regions away from the heated zone. In addition, temperature estimation is problematic when the PRF method is used in tissues that contain both water and fat. We propose a technique that uses a 2DRF pulse to excite a reduced volume and to suppress fat. Focused ultrasound heating experiments were conducted where the reduced-FOV was selected to contain the focal spot. The observed temporal resolution increased by 2.3 fold without temperature error due to the presence of fat.

14:30         4396.     Online Phase-Collection Based Correction of MR Thermometry for Breast HIFU Ablation: Evaluation Under Various Respiratory Motion Susceptibility Conditions in Phantom.

Matthieu Lepetit-Coiffe¹, Baudouin Denis de Senneville¹, Philippe Lourenco de Oliveira¹, Charles Mougenot¹, Bruno Quesson¹, Jean Palussiere¹, Chrit Moonen¹

¹Laboratoire IMF CNRS UMR 5231 / Universite Bordeaux 2, Bordeaux, France, Metropolitan

An online phase-atlas based correction was evaluated regarding temperature accuracy improvement under various frequency and amplitude of respiratory motion conditions simulated thanks to a home made phantom. Fixed acoustic power and duration of High Intensity Focused Ultrasound (HIFU) was delivered during real time MR temperature acquisition in the respiratory mimicking phantom; Temperature maps were corrected with the proposed correction and results compared to temperature maps acquired under same HIFU delivery conditions without any perturbations.

15:30         4397.     Initial Evaluation of a Fast Chemical Shift Thermometry Technique in the Breast

Hua Ai¹, Brian A. Taylor¹, Andrew M. Elliott¹, Huong LePetross², John D. Hazle¹, Roger Jason Stafford¹

¹Imaging Physics, The University of Texas MD Anderson Cancer Center, Houston, TX, USA; ²Diagnostic Imaging, The University of Texas MD Anderson Cancer Center, Houston, TX, USA

A fast chemical shift imaging technique for real-time temperature imaging in the breast was investigated. The uncertainty in temperature was evaluated by measuring the proton resonant frequency shift between lipid and water at various points across the breast. Temperature uncertainty over 20s was about 3.0~3.6°C. Calculated frequency difference is 13.8% percent lower than the expected quadrature sum of two frequencies, indicating that temperature uncertainty was consistently reduced by subtraction of lipid peak while the improvement was largely limited by low SNR. Moderate compression of breast is needed to minimize motion and hence temporal uncertainty in temperature.

13:30         4398.     Increased Applicability of a Manipulator to Assist MR Guided Microwave Ablation of Liver Tumors and Clinical Experiences of 14 Cases

Shigehiro Morikawa¹, Toshiro Inubushi¹, Shigeyuki Naka², Koichiro Murakami², Yoshimasa Kurumi², Tohru Tani², Hasnine Akter Haque³, Junichi Tokuda⁴, Nobuhiko Hata⁴

¹Biomedical MR Science Center, Shiga University of Medical Science, Ohtsu, Shiga, Japan; ²Department of Surgery, Shiga University of Medical Science, Ohtsu, Shiga, Japan; ³GE Yokogawa Medical System, Hino, Tokyo, Japan; ⁴Brigham and Women's Hospital, Boston, MA, USA

We developed a motorized MR-compatible manipulator with synergetic remote-center-of-motion control to assist MR guided microwave ablation of liver tumors and started clinical studies using it. To increase the applicability of this manipulator, several types of adaptors of the handpiece were developed. As a result, lateral approach to tumors was enabled and liver tumors in various locations could be punctured under the assistance of this manipulator. We have already experienced 14 clinical cases with liver tumors. All cases were successfully treated without any complications. Nowadays, this manipulator is routinely used in this procedure.

14:00         4399.     Automatic Passive Tracking of an Manually Steerable Instrument Holder for MR-Guided Interventions Applied in LITT

Axel Joachim Krafft¹, Florian Maier¹, Patrik Zamecnik, André de Oliveira^1,2, Jürgen Walter Jenne^3,4, Roger Jason Stafford⁵, Kamran Ahrar⁶, Axel Winkel⁷, Wolfhard Semmler¹, Michael Bock¹

¹Medical Physics in Radiology, German Cancer Research Center (DKFZ), Heidelberg, Germany; ²Siemens AG, Erlangen, Germany; ³Clinical Cooperation Unit Radiation Oncology, German Cancer Research Center (DKFZ), Heidelberg, Germany; ⁴Mediri GmbH, Heidelberg, Germany; ⁵Imaging Physics, The University of Texas M.D. Anderson Cancer Center, Houston, TX, USA; ⁶Interventional Radiology, The University of Texas M.D. Anderson Cancer Center, Houston, TX, USA; ⁷Invivo Germany GmbH, Schwerin, Germany

The increasing complexity of MR-guided interventions demands high SNR and short image acquisition times. Therefore, a rising number of such procedures is performed in closed-bore MR scanners where patient access is severely restricted. Consequently, safe, accurate, and continuous instrument monitoring and guidance is a mandatory pre-requisite. A combination of an automatic passive tracking technique and a manually steerable instrument holder is proposed to meet the requirements of percutaneous interventions. The setup was tested during LITT as an advanced minimally invasive technique. Our experiments demonstrate the potential of this approach which is suitable for a wide range of interventional applications.

14:30         4400.     A Technique for Improving 2D X-Ray to 3D MRI Registration in Fluoroscopy-Guided Orthopedic Interventions

Martijn John van der Bom¹, Josien Petronella Pluim¹, Matthew Gounis², Lambertus Wilbert Bartels¹

¹Image Sciences Institute, University Medical Center Utrecht, Utrecht, Netherlands; ²Radiology, University of Massachusetts Medical School, Worcester, MA, USA

MRI can provide 3D insight and soft tissue contrast for real-time intra-procedural navigation. However, MRI to X-ray registration is technically challenging due to insufficient similar landmarks caused by differences in physical imaging principles. We have investigated a novel technique for improved 3D MR to 2D X-ray registration. A simulated CT is generated from multispectral MRI data using a k-nearest neighbor classification method. This simulated CT than provides sufficient landmarks for MRI to X-ray registration. Classification was based on MR sequences commonly used in clinical practice. Validation in ex vivo experiments shows that this method is accurate and reliable.

15:00         4401.     Validation of a Vasculature-Based Co-Registration Technique of X-Ray and MR Images for the Guidance of Cardiovascular Interventions

Perry Radau¹, Andriy Smatukha², Peter Jones¹, Regis Vaillant³, Alexander J. Dick⁴, Graham A. Wright¹

¹Imaging Research, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada; ²Applied Science Laboratory, General Electric Healthcare, Toronto, ON, Canada; ³Applied Science Laboratory, General Electric Healthcare, Buc, France; ⁴Cardiology, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada

Combined X-ray and MRI (XMR) guidance of interventions has been shown to offer considerable benefits by providing soft-tissue and functional information and real-time visualization of devices. This study presents and validates a software solution, FluoroFusion, which does not require hardware calibrations, optical tracking, or external markers. Phantom images were successfully fused with 3D error of 1.7 ± 0.7 mm and range 0.1-3.7 mm, from 14 trials utilizing two X-ray views each. In a study with patient data the accuracy was similar based on qualitative visual assessment. Its use requires minimal workflow adjustment or operator time. Images from any stand-alone MRI and X-ray systems could be used with Fluorofusion.

13:30         4402.     Automatic Slice Positioning of the Interventional Guidewire Using Passive Paramagnetic Markers

Sunil Patil¹, Oliver Bieri¹, Permi Jhooti¹, Klaus Scheffler¹

¹Division of Radiological Physics, University of Basel Hospital, Basel, Switzerland

In this work, a novel method for passive real-time tracking of paramagnetic markers is proposed, termed Projection Reconstruction Imaging using echo-Dephasing (PRIDE). PRIDE is based on the acquisition of projections along all three physical (X, Y, and Z) axes, enabling a real-time positional update of the slice to the detected guidewire location. PRIDE exploits paramagnetic marker-related gradients along each projection to give rise to peak signal corresponding to marker position. Our initial in-vitro and in-vivo results indicate that PRIDE can be successfully used for real-time slice positioning.

14:00         4403.     Multiparametric Monitoring of Chemical Ablatons Using a Rapid Chemical Shift Imaging Technique

Brian A. Taylor¹, Andrew M. Elliott¹, John D. Hazle¹, Roger Jason Stafford¹

¹Imaging Physics, The University of Texas MD Anderson Cancer Center, Houston, TX, USA

The potential for monitoring chemical ablations with a high spatiotemporal resolution CSI technique is demonstrated ex vivo. Results demonstrated the ability to dynamically detect the chemical shift of the methyl group in ethanol and separate it from the water component. Additionally, dynamic T1-weighted images and quantitative T2* maps provided additional information which correlated with the agent location as observed on post-injection T1-weighted images. If a reference standard of known concentration is kept in the image, the amplitude of each chemical shift could possibly provide a quantitative means to monitoring the progress of therapy and better relate to outcome.

14:30         4404.     A Reliable Real Time Liver Motion Tracking for Application in Magnetic Resonance Guided High Intensity Focused Ultrasound Therapy

Dattesh D. Shanbhag¹, Rekha Tranquebar¹, James C. Ross²

¹Imaging Technologies, GE Global Research, Bangalore, Karnataka, India; ²Department of Radiology, Surgical Planning Laboratory, Boston, MA, USA

Bias error analysis for liver feature tracking using an automated thin-plate spline algorithm is presented. The mean errors observed with automated algorithm were 1.8 mm and 2.9 mm with a maximum bias of 3.5 mm and 5mm, along the S-I and R-L respectively. The analysis shows that automated tracking could reliably replace the tedious manual tracking of features during the MRgFUS therapy procedure.

15:00         4405.     Compressed Sensing for Highly Accelerated 3D Visualization of 19F-Catheters

Carsten Oliver Schirra¹, David Brunner², Jochen Keupp³, Reza Razavi¹, Tobias Schaeffter¹, Sebastian Kozerke^1,2

¹Division of Imaging Sciences, King's College London, London, UK; ²Institute for Biomedical Engineering, Univeristy and ETH Zurich, Zurich, Switzerland; ³Philips Research Europe, Hamburg, Germany

Short response times and RF safety are essential in MR-guided interventions and numerous techniques have been proposed to address these issues.

14:00         4406.     The Influence of Background Gradients in Multi Gradient-Echo MR Thermometry

Sara Maria Sprinkhuizen¹, Chris J. Bakker¹, Lambertus Wilhelmus Bartels¹

¹Image Sciences Institute, University Medical Center, Utrecht, Netherlands

In multi gradient-echo (mGE) based MR thermometry (MRT), multiple gradient-echo images are acquired. The mGE signal contains spectral information of the substances involved, which can be post-processed into temperature data. Our study shows the impact of background gradients on the mGE based MRT technique. Such gradients shift the acquired echoes, and therefore alter the spectral properties of the mGE signal. This leads to errors in the temperature data, regardless of the post-processing technique used. A correction method is presented which largely adjusts for this effect.

14:30         4407.     Temperature Error Reduction During MRI Guided HIFU Treatment

Xiaodong ZHOU^1,2, Qiang HE^1,2, Qiang ZHANG², Cheng NI^1,2

¹College of life science and technology, Tongji University, Shang hai, China; ²Siemens Mindit Magnetic Resonance Ltd., Shenzhen, Guangdong, China

We propose a method to remove ultrasound applicator susceptibility temperature error based on PRF-shift phase difference method during MRgHIFU. With this method reference image is not necessary to acquire repetitively when applicator moves position. Therefore tissue cooling down time between reference imaging can be saved. MRgHIFU treatment is simplified and total treatment time is much reduced.

15:00         4408.     Practical Thoughts on Using Repeated Bipolar Gradients for MR-ARFI

Jing Chen^1,2, Kim Butts Pauly²

¹Electrical Engineering, Stanford University, Stanford, CA, USA; ²Radiology, Stanford University, Stanford, CA, USA

MR-ARFI is a novel method for monitoring HIFU treatments. We have previously shown how to improve the accuracy and precision of MR-ARFI by using the repeated bipolar gradients to encode the displacement. In this work, two practical aspects in the in vivo application of MR-ARFI are analyzed. The bulk motion sensitivity is considered, and a list of optimized encoding width for different tissue types is provided.

15:30         4409.     Measurements of Kinetic Stability, Blood-Brain Barrier Permeability and Cytotoxicity for Two Thulium Based Contrast Agents

Daniel Coman^1,2, Margarita Gattas-Sethi¹, Hubert K.F. Trubel^3,4, Peter Herman^1,5, Fahmeed Hyder^2,6, Garry Kiefer⁷, Francesco d’Errico⁶

¹Department of Diagnostic Radiology, Yale University, New Haven, CT, USA; ²Quantitative Neuroscience with Magnetic Resonance (QNMR), Yale University, New Haven, CT, USA; ³Bayer HealthCare Research Center, Wuppertal, Germany; ⁴Department of Pediatrics, HELIOS-Klinikum Wuppertal, University Witten/Herdecke, Wuppertal, Germany; ⁵Semmelweis University, POB 448, H-1446, Budapest, Hungary; ⁶Departments of Diagnostic Radiology and Biomedical Engineering, Yale University, New Haven, CT, USA; ⁷Macrocyclics, Dallas, TX 75235, USA

In the last decade, a new non-invasive method for temperature and pH determination was developed based on temperature and pH dependencies of the ¹H chemical shifts emanating from TmDOTP^5-. More recently a similar temperature-sensitive probe was introduced, based on temperature dependence of the methyl ¹H chemical shift of TmDOTMA^-, which is pH independent. Our results show that these two agents are kinetically stable, they cross the blood-brain barrier, they clearly show the lack of any acute toxicity effects on Chinese hamster lung cells and they demonstrate that the brain’s activity is unaffected by their presence in the extracellular space.

13:30         4410.     Real-Time Non-Subtraction Thermal Ablation Monitoring In-Vivo Using RE-TOSSI

Jamal Jon Derakhshan^1,2, Sherif G. Nour^1,2, Simi Paul², Stephen R. Yutzy^1,2, Jeffrey L. Sunshine², Mark A. Griswold^1,2, Jeffrey L. Duerk^1,2

¹Biomedical Engineering, Case Western Reserve University, Cleveland, OH, USA; ²Radiology, University Hospitals of Cleveland, Cleveland, OH, USA

Thermal ablation treatment monitoring in real-time has been challenging in MRI. Some techniques require a pre-treatment image for baseline subtraction which can cause severe distortions in temperature/thermal dose maps if motion occurs between acquisition of the baseline and subsequent imaging assessment. In a study involving 26 laser ablations in the paraspinal and tongue base muscles of 11 pigs at 1.5 T, RE-TOSSI, a new SSFP based pulse sequence with 5 second temporal resolution, is shown to directly visualize growing thermal ablation lesions (as compared to post ablation contrast enhanced images) with high CNR and resolution without requiring a pre-treatment image.

14:00         4411.     Dynamics and Distributions of Temperature Changes During Pharyngeal Selective Brain Cooling by ¹H CSI

Daniel Coman^1,2, Hubert K.F. Trubel^3,4, Peter Herman^1,5, Fahmeed Hyder^2,6

¹Department of Diagnostic Radiology, Yale University, New Haven, CT, USA; ²Quantitative Neuroscience with Magnetic Resonance (QNMR), Yale University, New Haven, CT, USA; ³Bayer HealthCare Research Center, Wuppertal, Germany; ⁴Department of Pediatrics, HELIOS-Klinikum Wuppertal, University Witten/Herdecke, Wuppertal, Germany; ⁵Semmelweis University, POB 448, H-1446, Budapest, Hungary; ⁶Department of Diagnostic Radiology and Biomedical Engineering, Yale University, New Haven, CT, USA

Pharyngeal selective brain cooling (pSBC) has been suggested a more suitable alternative to whole body cooling to decrease brain temperature. Temperature distributions in rat brain can be obtained within minutes by using a new exogenous temperature-sensitive probe, TmDOTMA^-. The pharyngeal brain cooling rate constants and the recovering rate constants were calculated from the temperature variation over time. The recovering rate constants show a distribution which is slightly more dispersed than that of the cooling rate constants. Our results indicate that pSBC rate constants are tightly dependent on the local net heat contribution, measured indirectly by the recovering rate constants.

14:30         4412.     Feasibility Study of a New System for MR-Guided Scanned Focused Ultrasound Hyperthermia in Small Animals

Robert Staruch¹, Rajiv Chopra

¹University of Toronto, Toronto, ON, Canada

In order to facilitate hyperthermia-mediated drug delivery studies in small animal models, a motorized MR-guided transducer positioning system has been developed for focused ultrasound heating of rapidly scanned circular trajectories between 5 and 20mm in diameter. With this system, the temperature artifacts during simultaneous scanning and imaging were small enough to allow controlled heating, and good spatial and temporal temperature control (+/- 1 degree C) were achieved by applying multiple single-point controllers along the circular trajectory.

15:00         4413.     Computer Simulation and Temperature Measurement for MR Hyperthermia Therapy Using Coaxial-Slot Antenna

Tae Hyung Kim¹, Kee Chin Tan¹, Song I. Chun¹, Ki Sueng Choi¹, Yong Hee Han¹, Chi Woong Mun¹

¹Biomedical engneering, Inje Univ., Gimhae, Kyungnam, Korea

Hyperthemia therapy combined with Magnetic Resonance Imaging (MRI) has several advantages for the treatment of small tumor. High Frequency Ultra Sound (HIFU) method has been commonly used for heating source. HIFU method, however, has some drawbacks. For this, coaxial-slot antenna driven by 2.45GHz ¥ì-wave was proposed. The coaxial-slot antenna was designed by computer simulation. To optimize the design of antenna, reflection parameter(S11) and SAR distribution in muscle model at the center frequency, 2.45§×, were analyzed using HFSS s/w. The temperature was measured to test heating performances of hand-made coaxial-slot antenna based on the distance from the heating point and power using agarose gel phantom. The SAR distribution has similar trends on both computer simulation and temperature measurement experiment. In this study suggest that coaxial-slot antenna is a useful heating source for MR hyperthemia therapy.

13:30         4414.     Design and Preliminary Clinical Studies of an MRI-Guided Transrectal Prostate Intervention System

Axel Krieger¹, Peter Guion², Csaba Csoma¹, Iulian I. Iordachita¹, Anurag K. Singh^2,3, Aradhana Kaushal², Cynthia Menard⁴, Gabor Fichtinger^1,5, Louis L. Whitcomb¹

¹Johns Hopkins University, Baltimore, MD, USA; ²NCI - NIH-DHHS, Bethesda, MD, USA; ³Roswell Park Cancer Institute, Buffalo, USA; ⁴Princess Margaret Hospital, Toronto, Canada; ⁵Queen's University, Kingston, Canada

This paper reports on the development and preliminary clinical studies of a new system for MRI guided transrectal prostate interventions. This system employs a novel hybrid tracking method which reduces procedure time and simplifies deployment on different scanners while achieving millimeter needle placement accuracy. Transrectal MRI guided prostate interventions such as biopsies and gold marker placements inside a high-field MR scanner are reported in initial clinical trials, demonstrating accurate and fast needle targeting of the complete clinical target volume.

14:00         4415.     Comparison of Two Real Time Tracking Methods for a Robotic Assistance System

Jaane Rauschenberg¹, André De Oliveira², Florian Maier¹, Wolfhard Semmler¹, Michael Bock¹

¹Medical Physics in Radiology, German Cancer Research Center, Heidelberg, Germany; ²Siemens AG, Erlangen, Germany

Passive markers such as contrast-agent filled spheres provide a cost-effective and reliable method to localize devices in an MR system. In this work two real time methods are evaluated that track the position of a robotic assistance system with passive markers. A fully automatic localization algorithm with sub-pixel precision, a combination of a phase-only cross correlation with a center-of-mass algorithm, is compared to a direct coordinate exchange between robot and MR. Slice positions and orientations are evaluated in a phantom experiment yielding a similar angular precision of about 1.3° for the needle axis.

14:30         4416.     MRI-Guided Monitoring of Cement Injection in Vertebral Bodies

René Schilling¹, Hermann Josef Bail², Florian Wichlas², Christian Jürgen Seebauer², Jens Rump¹, Ulf Teichgräber¹

¹Department of Radiology, Charité Universitätsmedizin Berlin, Berlin, Germany; ²Center for Musculoskeletal Surgery, Charité Universitätsmedizin Berlin, Berlin, Germany

At last years’ ISMRM meeting, we presented modified PMMA bone cement for interventional use in open MRI systems. In this study, we evaluated various techniques for the safe application of cement in vertebral bodies under MRI–monitoring, using a phantom with artificial leakage ducts. Although monitoring a volume with near real-time MRI is still challenging, the experimental results suggest that a fast T1FSE (1 image/ 2.4s) and a “radial-shift” technique provide sufficient detection of leakage ducts up to 3mm in diameter. The experimental filling time (85s) was markedly shorter, than the cement injection time recommended by the manufacturer.

15:00         4417.     Nd:YAG Laser Based Laparoscopic Liver Resection in a 1.0 Tesla High Field Open MRI: Evaluation of Suitable Dynamic MR Sequences for Image Guided Surgery

Sascha Santosh Chopra¹, Guido Schumacher¹, Sven Christian Schmidt¹, Christian Seebauer², Ioannis Papanikolaou³, Ivo Van der Voort³, Florian Streitparth⁴, Jens Rump⁴, Ulf Teichgraeber⁴

¹Dept. of Surgery, Charité Campus Virchow-Clinic; University Medicine Berlin , Berlin, Germany; ²Dept. Trauma Surgery, Charité Campus Virchow-Clinic; University Medicine Berlin; ³Dept. Gastroenterology, Charité Campus Virchow-Clinic; University Medicine Berlin; ⁴Dept. of Radiology, Charité Campus Mitte; University Medicine Berlin

Aim of this study was to identify suitable interactive (dynamic) magnetic resonance (MR) sequences for real time MR guided liver dissection in a 1.0 Tesla high field open MRI Scanner. Four dynamic sequences encompassing balanced steady state free precession (bSSFP), T1W gradient echo (GRE), T2W GRE and T2W fast spin echo (FSE) were analyzed regarding the image quality, artifact susceptibility and the performance of SNR and CNR. The T2W TSE sequence (1.5 seconds/image) was considered superior due to an intraoperative SNR of 6.9 (±0.7) and CNR (vessel to parenchyma) of 5.6 (±1.7) in the interventional setting. As a proof of concept MR-guided laparoscopic liver resection was performed in two healthy house pigs using the T2W TSE sequence.

13:30         4418.     An Optical and Electromagnetic Hybrid Tracking System for MR Image Guided Interventional Procedures

Shigehiro Morikawa¹, Hasnine Akter Haque², Shigeyuki Naka³, Koichiro Murakami³, Yoshimasa Kurumi³, Tohru Tani³, Toshiro Inubushi¹

¹Biomedical MR Science Center, Shiga University of Medical Science, Ohtsu, Shiga, Japan; ²GE Yokogawa Medical System, Hino, Tokyo, Japan; ³Department of Surgery, Shiga University of Medical Science, Ohtsu, Shiga, Japan

In MR guided interventional procedures, an optical or electromagnetic tracking system has been used for image plane control. The optical tracking system precisely works in the strong magnetic fields, but line of sight must be maintained. An electromagnetic tracking system, which utilizes gradient pulses during data collection, can be used in any directions and even in covered area, but it requires continuous MR image acquisitions and its accuracy depends on the location. To take advantages of two sensor systems complementarily, we have developed a hybrid tracking system of them and enabled seamless navigation in MR guided interventional procedures.

14:00         4419.     Dual MRI-Tracking/SJM NavX Electro-Physiology Catheters for Navigation Inside an MRI: Initial Experience

Ehud Schmidt¹, Jeremy D. Dando², Robert D. Darrow³, Daniel Kacher¹, Laurence M. Epstein⁴, Ferenc Jolesz¹, Charles L. Dumoulin⁵, Vivek Y. Reddy⁶

¹Radiology, Brigham and Womens Hospital, Boston, MA, USA; ²Atrial Fibrillation Division, St. Jude Medical, Minnetonka, MA, USA; ³GE Global Research, Niskayuna, NY, USA; ⁴Cardiology, Brigham and Womens Hospital, Boston, MA, USA; ⁵Radiology, Cincinnati Children’s Hospital, Cincinnati, OH, USA; ⁶Cardiology, University of Miami, Miami, FL, USA

An MRI-compatible Electro-Physiology catheter was constructed with positional tracking elements using both MRI Tracking micro-coils and St Jude NavX non-MRI, Voltage based, tracking. The catheter was tested in a Left Atrial Phantom inside an MRI. Simultaneous low-receiver bandwidth (+16 and +32 KHz) MR Tracking at 20 fps and Navx Tracking at 12 fps was possible. High Bandwidth (+128 KHz) Real-time MR Imaging disturbed the NavX tracking, but intermittent use of imaging and NavX tracking was possible, since the Navx system returned to the same position after each MRI scan.

14:30         4420.     Tracking System for Real-Time MR-Guided Percutaneous Interventions at 1.5T

R. Jason Stafford¹, Barry Fetics², Amir Roth², Christine Lorenz³, Axel J. Krafft⁴, Michael Bock⁴, Kamran Ahrar⁵

¹Imaging Physics, The University of Texas M. D. Anderson Cancer Center, Houston, TX, USA; ²Robin Medical, Inc., Baltimore, MD, USA; ³Siemens Medical Solutions, Inc., Baltimore, MD; ⁴Division of Medical Physics in Radiology, German Cancer Research Center (DKFZ), Heidelberg, Germany; ⁵Diagnostic Radiology, The University of Texas M. D. Anderson Cancer Center, Houston, TX, USA

New short bore, wide aperture 1.5T scanners make MR-guidance of percutaneous interventions in cylindrical bore scanners with real-time imaging technically feasible. A tracking system that operates within the bore of the magnet to aid in the needle or applicator placement of these procedures is needed for increasing workflow. In this report we describe our initial experience with an FDA approved active tracking system for MRI that can be used in a cylindrical bore to help plan and target needle/applicator placement in real time.

15:00         4421.     MRPen – 3D Marker Tracking for Percutaneous Interventions

Florian Maier¹, Axel Joachim Krafft¹, André de Oliveira², Wolfhard Semmler¹, Michael Bock¹

¹Medical Physics in Radiology, German Cancer Research Center (DKFZ), Heidelberg, Germany; ²Healthcare Sector, Siemens AG, Erlangen, Germany

In percutanous MR-guided interventions passive markers are used to delineate the position or the trajectory of rigid instruments, e.g. needles. In this work, a modified passive marker and a modified passive tracking pulse sequence are proposed. Based on the 3D position and rotation of the marker an imaging slice is automatically adjusted in real time. Measurements indicate that the marker design combined with the pulse sequence allows an accurate rotation estimation and 3D tracking of rigid instruments. Thus, the choice of an optimal needle trajectory in percutaneous interventions is highly facilitated and under manual control of the operator.

13:30         4426.     Intravascular 3.0T MRI Using an MR Imaging-Guidwire

Yanfeng Meng^1,2, Zhaoqi Zhang¹, Huidong Gu³, Xing Lv⁴, Jue Zhang⁴, Jing Fang⁴, Bensheng Qiu², Xiaoming Yang²

¹Radiology, Beijing Anzhen Hospital, Capital Medical University, Beijing, China; ²Radiology, University of Washington, Seattle, WA, USA; ³MRI Research Team, MR Modality, GE Healthcare China, Beijing, China; ⁴Biomedical Engineering, Peking University, Beijing, China

Previous studies demonstrate MR imaging-guidewires (MRIG) can be used to create high-resolution vessel wall and plaque images with 1.5T MR scanners. The aim of this study was to evaluate the possibility of generating MRIG-mediated intravascular MR imaging with a 3.0T MR scanner. We first tested, in vitro, the functionality of a 0.032-in MRIG for intraluminal 3T MRI of a phantom, which demonstrated clearly the wall of the phantom tube and the MRIG itself with high SNRs. We then validated, in vivo, the feasibility of generating intravascular 3T MRI of abdominal aorta in two rabbits, which showed clearly the aortic walls when using different sequences. This study proves the capability of using the MRIG to generate high-SNR, intravascular 3.0T MRI, which may enable further MRIG-guided interventions in the 3T environment.

14:00         4427.     Tapering Improves the Forward-Looking Properties of the Interventional Loopless Antennae

Di Qian^1,2, AbdELMonem M. EL-Sharkawy¹, Paul A. Bottomley^1,2

¹Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University, School of Medicine, Baltimore, MD, USA; ²Department of Electrical and Computer Engineering, Johns Hopkins University, Baltimore, MD, USA

The interventional loopless antenna has low tip sensitivity, which is exacerbated by uniform insulation. It is shown that the antenna’s current sensitivity can be redistributed distally by tapering the insulation. Electromagnetic (EM) method-of-moment analysis, experiments in phantoms and biological specimens at 3T, demonstrate that tapering can produce ~4-fold gains in the signal-to-noise ratio 1cm from the tip. Insulation tapering enhances the tip visibility for interventional MRI (iMRI) using loopless antennae without compromising the antenna’s tiny cross-sectional profile. Numerical EM analysis is a valuable design tool for improving the performance characteristics.

14:30         4428.     High Precision Measurement of Micro-Coil Locations for Real-Time Tracking Applications

Melvyn Boon King Ooi¹, Sascha Krueger², William J. Thomas³, Srirama V. Swaminathan⁴, Truman R. Brown^1,3

¹Biomedical Engineering, Columbia University, New York, USA; ²Philips Research Europe, Hamburg, Germany; ³Radiology, Columbia University, New York, USA; ⁴Philips Medical Systems, Cleveland, OH, USA

The ability to track the positions of multiple micro RF-coil in the MR system has been the foundation of recent advances in MR-guided medical devices, as well as motion detection and correction applications. The success of these marker-based tools is dependent upon the quality of the micro-coil position calculation. The current work presents a fully developed methodology – consisting of a short tracking pulse-sequence and micro-coil hardware – capable of high measurement accuracy and reproducibility in a temporal resolution suitable for use in real-time applications. The technique’s efficacy is validated by a series of measurements in a well-defined grid-phantom.

15:00         4429.     Interactive MR Image Scan Guidance with a Capability of Motion Compensation Using Endoscout Sensor

Hasnine Akter Haque¹, Shigehiro Morikawa², Shigeyuki Naka³, Koichiro Murakami⁴

¹Japan Applied Science Laboratory, GE Yokogawa Medical Systems Inc., Hino, Tokyo, Japan; ²Biomedical MR Science Center, Shiga University of Medical Science, Ohtsu, Shiga, Japan; ³Department of Surgery, Shiga University of Medical Science, Ohtsu, Shiga, Japan; ⁴Department of Surgery,, Shiga University of Medical Science, Ohtsu, Shiga, Japan

Interactive real-time MR image guided navigation for the intervention use is a promising strategy. However, it is difficult to perform complicated surgical procedures with the guidance of Real-time MR images only. Reformatted image from the pre-acquired high-resolution interoperative image volume is commonly used for improved visibility, with an assumption that no change of patient position during surgery, which is not always possible. Here we propose a method of full automatic motion compensation technique using tiny gradient tracking sensor (Endoscout) attached to the patient.

13:30         4430.     MR Compatible Phased Array HIFU Transducer for Localized Prostate Cancer Treatment; in Vitro Validation

Lorena PETRUSCA¹, Rares SALOMIR¹, Emmanuel BLANC², Lucie BRASSET², Francois COTTON³, Jean-Yves CHAPELON¹

¹U556, Inserm, Lyon, France; ²EDAP-TMS, Lyon, France; ³RMN Unit, CHU Lyon Sud, Lyon, France

High Intensity Focused Ultrasound (HIFU) under MRI guidance represents a minimally invasive approach for localized prostate cancer treatment. In our in-vitro study, an MR-compatible multi-element endorectal HIFU device capable of focusing the beam at different depths in the tissue is being investigated. Fast MR thermometry was made in 2 orthogonal planes simultaneously for different sonication sequences chosen based on numerical simulations: elementary lesions, lines, slices and volumes. Temperature and thermal dose maps obtained showed that a homogenous lethal dose was induced in the desired shape. Dynamic focusing strategy permits to treat patients with more complex anatomy: smaller or larger prostate.

14:00         4431.     3-D Control Over Spatial Heating Using Multi-Element Ultrasound Heating Applicators and Real-Time MR Temperature Feedback

Rajiv Chopra^1,2, Kee Tang¹, Matthew Asselin¹, Mathieu Burtnyk^1,2, MIchael Bronskill^1,2

¹Imaging Research, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada; ²Medical Biophysics, University of Toronto, Toronto, Ontario, Canada

The aim of this study was to evaluate the capability to deliver a targeted 3D pattern of thermal damage in the prostate gland using a multi-element transurethral ultrasound heating applicator operating under MR temperature control. MRI-guided transurethral ultrasound therapy was performed in 10 dogs using a five element heating applicator operating at 8.3MHz. MR thermometry was acquired in five planes centred on each transducer to measure the heating pattern in the plane of rotation of each element, and to control the treatment. The results of this study indicated that multi-planar MR thermometry enabled precise heating of a 3D region in the prostate gland.

14:30         4432.     Feasibility of Cardiac-Gated 3T MRI-Guided Myocardial Ablation with High Intensity Focused Ultrasound

Aravind Swaminathan¹, Viola Rieke², Randy Lee King², John Pauly³, Kim Butts-Pauly², Michael McConnell¹

¹Medicine, Stanford University School of Medicine, Stanford, CA, USA; ²Radiology, Stanford University, Stanford, CA, USA; ³Electrical Engineering, Stanford University, Stanford, CA, USA

3T MRI guided cardiac HIFU was performed and proved feasible in an ex-vivo study on porcine myocardium. Left ventricular septal ablation lesions could be created of varying size and with cardiac gating. Ablation was successfully monitored with MR temperature mapping.

15:00         4433.     T2-Weighted Imaging of the Left Atrium Acutely After Pulmonary Vein Isolation Demonstrates Wall Thickening and Edema.

Dana C. Peters¹, Jeff Hsing¹, Kraig V. Kissinger¹, Beth Goddu¹, Reza Nezafat¹, Jason E. Taclas¹, Mark E.  Josephson¹, John V. Wylie¹, Warren J.  Manning¹

¹Medicine, Cardiovascular Division, Beth Israel Deaconess Medical Center & Harvard Medical School, Boston, MA, USA

Radio frequency (RF) ablation of the left atrium (LA) to treat atrial fibrillation has a 65-80% success rate. In order to understand the acute effects of pulmonary vein isolation (PVI) in treatment of AF, the RF ablations were imaged acutely (1 day following PVI) in five patients and two healthy subjects, who served as controls. The T2-weighted sequence to visualize edema in the LA was optimized. TE values, breath-holding vs. free-breathing, and fat-suppression methods were compared. We found evidence of thickened LA wall with increased T2. This study is an important step in anatomically asessing the completeness of PVI, in the acute stage.

13:30         4434.     Inductively Coupled Coils for Local SNR-Enhancement During MR-Guided Prostate Biopsy

Reiner Umathum¹, Axel Joachim Krafft¹, Andre de Oliveira^1,2, Michael Bock¹

¹Medical Physics in Radiology, German Cancer Research Center, Heidelberg, Germany; ²Healthcare Sector, Siemens AG, Erlangen, Germany

MR-guided prostate biopsy procedures need an improved image SNR. MRS metabolic information could further help deciding from which areas samples need to be taken. As MR data quality crucially depends on local SNR, typically endorectal coils are utilized. In commercial needle holders passive markers fill the available space so that integration of a coil becomes difficult. Inductively coupled coils without direct connection to the MR system have significantly smaller space requirements. Two coil designs were examined for their ability to improve the local SNR at or near the biopsy sample area and for possible integration into commercial biopsy systems.

14:00         4435.     Active Needle for Real-Time MRI-Guided Percutaneous Procedures

Christina Elena  Saikus¹, Kanishka Ratnakaya¹, Anthony Z. Faranesh¹, Robert J. Lederman¹, Ozgur Kocaturk¹

¹Translational Medicine Branch, Division of Intramural Research, National Heart Lung and Blood Institute, National Institutes of Health, Bethesda, MD, USA

MRI is increasingly used to guide percutaneous procedures but passive needle artifacts can often be non-specific and difficult to locate and follow under MRI. We have developed an active profiling needle that incorporates a loop antenna to provide distinct device appearance displayed on simultaneously acquired real-time images. The needle visibility and handling were evaluated during in vitro phantom imaging and in vivo real-time MRI-guided vascular access experiments in swine. Incorporating active visualization in the needle design provided clear information on needle location, insertion depth, and shaft positioning which can help safely conduct a variety of percutaneous interventional MRI procedures.

14:30         4436.     A Gd-DTPA Based Insoluble Polymer Coating Technique for Endovascular Device Visualization

Tomoka Kurita^1,2, Kagayaki Kuroda^3,4, Takeo Ohsaka¹

¹Department of Electronic Chemistry, Interdisciplinary Graduate School of Science and Engineering, Tokyo Institute of Technology, Midori-ku, Yokohama, Japan; ²R & D Center, Terumo Corporation, Ashigarakami-gun, Kanagawa, Japan; ³Department of Human and Information Sciences, School of Information Science and Technology, Tokai University, Hiratsuka, Kanagawa, Japan; ⁴Medical Device Development Center, Foundation for Biomedical Research and Innovation, Chuo-ku, Kobe, Japan

For MR-guided endovascular procedures, the passive visualization technique of guidewires and catheters with the coating of the Gd-DTPA based polymer was developed. The guidewire and urethane tube were coated using the polymer by a dipping process and they were soaked into saline for swelling. Imaging was performed using a clinical 1.5 Tesla MRI with T₁-FLAIR and FSPGR sequence. Coating devices were visible with a high signal intensity without any susceptibility artifact. The signal enhancement was maintainable and repeatable. This study shows that the MR-visible catheters and guidewires coated with the Gd-DTPA based polymer will contribute to the development of MR-guided endovascular procedures.

15:00         4437.     Optimization of Piezoelectric Motors to Enhance MR Compatiblity for Interventional Devices

Yi Y. Wang¹, Mohammed Salman Shazeeb^2,3, Christopher H. Sotak^2,4, Gregory Scott Fischer¹

¹Mechanical Engineering, Worcester Polytechnic Institute, Worcester, MA, USA; ²Biomedical Engineering, Worcester Polytechnic Institute, Worcester, MA, USA; ³Medical Physics, University of Massachusetts Medical School, Worcester, MA, USA; ⁴Radiology, University of Massachusetts Medical School, Worcester, MA, USA

MR enables closed loop image-guided surgery, but the inability to use conventional sensors and actuators in high-filed MR limits the availability of assistive technologies for interventional procedures. We have worked to optimize the Piezo LEG piezoelectric motors so that they can be successfully used in the MR environment. Through thorough shielding, grounding and filtering, we have managed to reduce average SNR degradation to under 3% for standard T1 and T2 scan protocols; this level of compatibility surpasses previously reported results. With appropriate considerations, this technology enables high precision MR-compatible interventional devices such as the neurosurgery robot we are developing.

14:00         4438.     A Fast Clinically Viable 3D T2 Mapping Technique

Ravi Teja Seethamraju¹, Vladimir Jellus², Timothy Hughes³, Mukesh G. Harisinghani^4,5, Alexander Guimaraes^4,5

¹MR R&D, Siemens Medical Solutions, Charlestown, MA, USA; ²Siemens AG., Erlangen, Germany; ³Siemens AG, Erlangen, Germany; ⁴Center for Molecular Imaging Research, Boston, MA, USA; ⁵Radiology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA

Though T2 mapping is desirable in many clinical applications it is not preferred due to the long acquisition times associated with the conventional Spin Echo (SE) sequence. Turbo or fast SE sequences though considerably faster than spin echo are not suitable for T2 mapping due to the stimulated echoes associated with them. Here we present an alternative technique based on DESS (Double echo steady state) to produce T2 maps.

14:30         4439.     Rapid T1 Mapping of Mouse Myocardium with Saturation Recovery Look-Locker Method

Wen Li^1,2, Jia Zhong^1,2, Ming Lu^1,2, Mark Griswold^1,2, Xin Yu^1,2

¹Biomedical Engineering Department, Case Western Reserve University, Cleveland, OH, USA; ²Case Center for Imaging Research, Case Western Reserve University, Cleveland, OH, USA

An ECG-triggered saturation recovery Look-Locker (SRLL) method is developed for fast cardiac T1 mapping. High temporal resolution (2.5 minutes) can be achieved without the aid of parallel imaging or EPI. In vitro validation was performed by comparing T1 measurements with that obtained from inversion-recovery method. The in vivo application of this method in delineating the kinetics of manganese-induced T1 shortening was demonstrated in mouse heart in response to intraperitoneal injection of MnCl2. These results suggest that SRLL may provide a rapid T1 mapping method for cardiac MRI using small animal models.

15:00         4440.     Improved Accuracy in T₁ Mapping and Flip Angle Correction with Random Spoiling in Radial Gradient Echo Imaging

Wei Lin¹, Hee Kwon Song¹

¹Department of Radiology, University of Pennsylvania, Philadelphia, PA, USA

Due to its high imaging speed and SNR efficiency, spoiled gradient echo imaging plays an important role in many quantitative MR methods. However, conventional RF spoiling generates non-ideal signal intensities, particularly at short repetition times and large flip angles, which could significantly reduce measurement accuracy. We propose a novel spoiling scheme, based on random gradient moments and RF phases, in conjunction with the radial acquisition scheme, to achieve ideally spoiled image intensities for a wide range of T₁, T₂, TR and flip angles. Phantom experiments demonstrate significant improvement in T₁ mapping and flip angle accuracy over conventional RF spoiling.

15:30         4441.     Fast High-Resolution T1 Mapping Using Inversion Recovery Look-Locker Echo-Planar Imaging at a Steady State: Optimization for Accuracy and Reliability

Wanyong Shin¹, Hong Gu¹, Yihong Yang¹

¹Neuroimaging Research Branch, National Institute on Drug Abuse, Baltimore, MD, USA

A fast T1 measurement sequence using inversion recovery Look-Locker echo-planar imaging at steady state (IR LL-EPI SS) is presented. Delay time for a full magnetization recovery is not required in the sequence, saving acquisition time significantly. Imaging parameters were optimized to minimize the bias from the imperfection of excitation pulses and to maximize the accuracy and reliability of T1 measurements. IR LL-EPI SS method preserves similar accuracy and reliability as conventional Look-Locker echo-planar imaging, while saving 20% acquisition time. Quantitative T1 mapping with 1×1×4 mm³ resolution and whole brain coverage (28 slices) can be collected in approximately 4 minutes.

13:30         4442.     Accuracy of T₁ and Equilibrium Magnetization Maps Using a Spoiled Gradient-Recalled Echo Sequence with Variable Flip Angles at 4.7 T, 7 T, and 9.4 T: A Gadolinium-Doped Gel Phantom Study.

Hideto Kuribayashi¹, Takuya Minowa², Yoshie Maitani², Masaki Sekino³, Hiroyuki Ohsaki³, Masayuki Ueda⁴, Yasuo Katayama⁴

¹Varian Inc., Minato-ku, Tokyo, Japan; ²Hoshi University, Shinagawa-ku, Tokyo, Japan; ³The University of Tokyo, Kashiwa, Chiba, Japan; ⁴Nippon Medical School, Bunkyo-ku, Tokyo, Japan

T₁ calculated using a spoiled gradient-recalled echo sequence with variable flip angles (SPGR-VFA) was underestimated especially at 9.4 T and with a short TR (< 10 ms). T₂* wasn't found to be the main reason of the shortened T₁ from the SPGR-VFA method at 9.4 T. Variable TR measurements were found to be useful for more accurate T₁map measurement. Magnetization equilibrium maps using the SPGR-VFA method were interestingly stable among the variable TRs.

14:00         4443.     Flip Angle Optimisation Using an In-Vivo T1 Distribution-Weighted Approach Based on the Cramer-Rao Lower Bound Theory for Accurate T1 Quantification

Keiko Miyazaki¹, David J. Collins¹, Dow-Mu Koh^1,2, Martin O. Leach¹, Matthew R. Orton¹

¹Cancer Research UK Clinical Magnetic Resonance Research Group, The Institute of Cancer Research, Sutton, Surrey, UK; ²Academic Department of Radiology, Royal Marsden Hospital, Sutton, Surrey, UK

The variable flip angle method allows accurate and efficient T1 quantification which is important in many fields of MR. Optimisation of flip angles when imaging samples with a large T1 range of values has been far from trivial and little work has been published on this subject. During the optimisation, it is important that the actual distribution of T1 values in the image sample is taken into account. We propose a novel flip angle optimisation approach, based on the Cramer-Rao lower bound (CRLB) theory, which is weighted by the probability distribution of the T1 range of interest.

14:30         4444.     Simultaneous Quantitation of T1 and T2 Using a Novel 2D Magnetization-Prepared Multi-Slice Pulse Sequence

Eric Tzguang Han¹

¹Applied Science Laboratory, GE Healthcare, Menlo Park, CA, USA

A novel 2D multi-slice pulse sequence uses a magnetization preparation scheme to impart varying degrees of T1- and T2- weighting in a single scan – enabling simultaneous quantitation of T1, T2, and M0. Using this new method, in vivo T1 and T2 maps of the head and knee were acquired. Measured T1 and T2 values were within published ranges.

15:00         4445.     T1-Mapping Using Two Flip Angle Spoiled FLASH-EPI Hybrid Sequences

Christine Preibisch^1,2, Steffen Volz², Ralf Deichmann²

¹Abteilung für Neuroradiologie, Klinkum rechts der Isar der TU München, München, Germany; ²Brain Imaging Center, Universitätsklinikum Frankfurt, Frankfurt, Germany

Problem: Several T1-mapping methods are based on dual flip angle spoiled gradient echo sequences. We propose a FLASH-EPI hybrid readout for improving the signal-to-noise ratio (SNR).

13:30         4446.     Singleshot Measurements of T₁ and Field Variation Using 2D Simultaneous Singleshot Spin-, Gradient-, and Stimulated-EPI (2D Ss-SGSTEPI)

Xianfeng Shi¹, Xin Liu¹, EunKee Jeong²

¹Department of Physics, University of Utah, Salt Lake City, UT, USA; ²Department of Radiology, University of Utah, Salt Lake City, UT, USA

A stimulated echo is formed after three 90^o RF pulses. Magnetization-preparation is accomplished immediately after the first RF pulse and the second RF pulse restores a half of prepared magnetization into the longitudinal space. Other half is remained on the transverse plane. In the conventional stimulated-echo pulse sequence, a half of the prepared magnetization is discarded. A novel imaging technique has been developed to utilize other half of the prepared magnetization at the spin-echo position and simultaneously acquire spin-, gradient-, and stimulated-EPI in a singleshot using 2D singleshot spin-/stimulated-EPI (2D ss-SGSTEPI).

14:00         4447.     Fast T1 Mapping Using a CPMG Sequence

Robert L. Greenman¹, Robert V. Mulkern²

¹Radiology, Harvard University Medical School and Beth Israel Deaconess Medical Center, Boston, MA, USA; ²Radiology, Harvard University Medical School and Boston Children's Hospital, Boston, MA, USA

Many pathologies are characterized by changes in the T1 relaxation rate of specific tissues. Most of the widely accepted methods of measurement of the tissue T1 values are time consuming. The Carr-Purcell-Meiboom-Gill (CPMG) pulse sequence is widely used for fast imaging where spin-echo contrast is required. We present a novel method for measuring T1 relaxation rates, based on the CPMG sequence. The results of T1 measurements of a phantom and human skeletal muscle suggest that this method may provide fast, accurate T1 information.

14:30         4448.     Apparent Change in the T1 of Lipids in Mixture

Houchun Harry Hu¹, Krishna S. Nayak¹

¹Ming Hsieh Department of Electrical Engineering, University of Southern California, Los Angeles, CA, USA

This work investigates potential bias in the T1 spin-lattice relaxation constant between pure fat and fat in mixture with water and lean tissue. We present results obtained at 3 Tesla from fat-water emulsions, ground pork meat, and in vivo experiments that suggest the existence of this additional T1 bias in fat. We observed the apparent T1 of fats to increase from their natural, pure-form T1 (200-350 ms at 3 Tesla) by as much as two to three-fold, when fat was in relatively homogenous mixtures characterized by fat-water signal fractions of less than 40%.

15:00         4449.     Quantification of Noise Efficiency with T₁ Corrected IDEAL Spoiled Gradient Echo Imaging

Curtis Nathan Wiens¹, Bryan Thomas Addeman², Shawn Joseph Kisch², Catherine D. Hines³, Hanzhou Yu⁴, Jean H. Brittain⁴, Scott B. Reeder^3,5, Charles A. McKenzie^1,2

¹Physics, University of Western Ontario, London, Ontario, Canada; ²Medical Biophysics, University of Western Ontario, London, Ontario, Canada; ³Biomedical Engineering, University of Wisconsin-Madison, Madison, WI, USA; ⁴Applied Science Laboratory, GE Healthcare; ⁵Radiology, University of Wisconsin-Madison, Madison, Wiconsin, USA

Both single and multi-flip angle methods have been used to minimize the T₁ bias that can cause a difference between the true fat mass fraction and the fat fraction estimated by IDEAL-SGPR. Breath-hold restrictions in abdominal imaging require that data for each flip angle have short acquisition times, and thus with lower SNR, when acquiring multiple flip angles. To examine this tradeoff, the noise efficiency of using 1-5 flip angles for fat fraction measurement was determined. Measurements in phantoms of varying water-fat fractions demonstrated that a single flip angle is the most noise efficient method to determine the fat fraction.

13:30         4450.     Accurate and Efficient Mapping of Flip Angle and T1 Using Simultaneous Actual Flip Angle - Variable Flip Angle Imaging (AFI-T1)

Samuel A. Hurley¹, Alexey A. Samsonov²

¹Medical Physics, University of Wisconsin - Madison, Madison, WI, USA; ²Radiology, University of Wisconsin - Madison, Madison, WI, USA

The knowledge of flip angle is essential for many quantitative MRI techniques such as variable flip angle (VFA) T1 mapping. We propose a new approach, which circumvents the shortcomings of Actual Flip angle Imaging (AFI) such as T1 dependence of flip angle values for short T1/long TR regimes, and simultaneously yields T1 maps with higher precision in less imaging time. The new method (AFI-T1) exploits a synergy between AFI and VFA T1 methods arising from the similarity of the steady state spoiled gradient echo (SPGR) based pulse sequences.

14:00         4451.     Skin T₁ Mapping at 1.5T, 3T, and 7T.

Joelle Karine Barral¹, Nikola Stikov¹, Erik Gudmundson², P. Stoica², Dwight George Nishimura¹

¹Electrical Engineering, Stanford University, Stanford, CA, USA; ²Information Technology, Uppsala University, Uppsala, Sweden

Increasing the field strength seems promising for high-resolution skin imaging but a legitimate concern is the increase in the spin-lattice relaxation parameter T₁. Skin T₁ maps at 1.5T, 3T, and 7T are compared and a novel non-linear least-square approach for fast and accurate T₁-estimation is proposed.

14:30         4452.     Fast Three-Point Approach for Volumetric T1 Mapping

Rahul Sarkar¹, Alan R. Moody^1,2, General Leung^1,2

¹Medical Biophysics, University of Toronto, Toronto, Ontario, Canada; ²Medical Imaging, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada

A fast volumetric technique for T1 quantitation at 3T is presented. Results are validated in a phantom model and initial results are shown in a whole brain scan requiring less than 5 minutes.

15:00         4453.     Towards Whole-Brain Quantitative T₁ Mapping at 3.0T for Imaging Hippocampal Sclerosis

Rebecca Sara Samson¹, Mark R. Symms², Mahinda Yogarajah², Niels Focke², John S. Duncan²

¹Department of Neuroinflammation, UCL Institute of Neurology, London, UK; ²Department of Clinical & Experimental Epilepsy, UCL Institute of Neurology, London, UK

This study aimed to establish the normal variation in T₁ measurements made using a 3D, whole-brain T₁ mapping method at 3.0T with a view to applying the technique in Hippocampal Sclerosis. Measured T₁ values are consistent with literature and inter- and intra-subject variation were demonstrated to be uniformly low, indicating the potential of this method for clinical applications, including imaging the hippocampus. T₁ mapping in Hippocampal Sclerosis may provide complementary information to the well-established technique of T₂ measurement.

14:00         4454.     Crushed Early Acquisition Spin Echo (CEASE): A Novel Technique for Positive Contrast and Spectroscopic Imaging of Superparamagnetic Particles

Bernard MCW Siow¹, Aiqing Chen², Gavin Clowry², Li Sun², Andrew M. Blamire¹

¹Newcastle Magnetic Resonance Centre, Newcastle University, Newcastle upon Tyne, Tyne and Wear, UK; ²Institute of Neuroscience, Newcastle University, Newcastle upon Tyne, Tyne and Wear, UK

Imaging of iron oxide particles with T2* weighted scans provides negative contrast, however, in biological samples other hypointense regions are likely to be present. Positive contrast methods generate contrast that is more specific to iron oxide particles. Here we present a novel sequence that provides positive contrast and provides spectroscopic imaging: The CEASE sequence eliminates background signal by crushing macroscopic magnetisation and rephases spins local to the iron oxide particles. We have shown that it provides positive contrast and spectroscopic imaging in a phantom containing MPIO and an ex-vivo rat brain with stem cells that have been labelled with MPIO.

14:30         4455.     Flow-Insensitve Magnetization Preparation for T2* Contrast

Christian Stehning¹, Jürgen Rahmer¹, Ulrike Blume², Tobias Schaeffter²

¹Philips Research Europe, Hamburg, Germany; ²Division of Imaging Science, King's College, London, UK

A magnetization preparation pulse sequence to obtain magnetic resonance images with strong T2* weighting is described. It overcomes the need to use sequences with long echo times (TE), which are inherently sensitive to flow artifacts.

15:00         4456.     Selective Depiction of Holmium-Loaded Microspheres (HoMS) Using Susceptibility Gradient Mapping (SGM): Initial Experience in Animal Models

Peter Roland Seevinck¹, Gopal Varma², Hendrik de Leeuw¹, Julien Senegas³, Tobias Schaeffter², Chris J.G. Bakker¹

¹Imaging Sciences Institute, Department of Radiology , University Medical Center, Utrecht, Netherlands; ²Division of Imaging Sciences, King's College London, UK; ³Philips Research Europe, Hamburg, Germany

The potential of susceptibility gradient mapping (SGM) for the selective depiction of holmium-loaded microspheres (HoMS) in the liver in ex vivo and in vivo animal models was investigated and confirmed. It was shown that positive contrast can be optimized by varying the echo time. Although SGM is not sensitive to the origin of the induced d÷ affect, directional information (i.e. 1D and 2D SG maps) can be used to exclude macroscopic field variations along the direction of choice. This allows discriminating between T2* effects induced by paramagnetic substances (local d÷) and macroscopic field inhomogeneities.

15:30         4457.     Quantification of SPIO Using T2*  and Phase Imaging

Alexia Rodríguez-Ruano¹, Marina Benito¹, Cristina Santa Marta², Gema Elvira³, Paula Montesinos¹, Juan Jose Vaquero¹, Augusto Silva³, Manuel Desco^1,4

¹Medicina y Cirugía Experimental, Hospital General Universitario Gregorio Marañón, Madrid, Spain; ²Dept. Física Matemática y Fluidos, Universidad Nacional de Educación a Distancia (UNED), Madrid, Spain; ³Cancer Genetics and Cancer Stem Cells Laboratory, Centre of Biological Research (CIB-CSIC), Madrid, Spain; ⁴Centro de investigación en red en salud mental (CIBERSAM), Madrid, Spain

Quantification in susceptibility imaging is becoming a request for many applications. Susceptibility effects have been studied mainly from two approaches, SWI and T2* relaxometry. However, most of the studies using these techniques were not particularly concerned with quantification issues. We have assessed the sensitivity and quantification ability of these techniques by comparing the measurements obtained using different types of images with the concentration of SPIO particles in the sample, using an agar phantom with different iron concentrations. Phase image contrast results in a linear relation with iron concentration similar to that obtained with T2* map, while requiring shorter acquisition time.

13:30         4458.     A Three-Field Study of Cerebral Transverse Relaxation Rates in Vivo: Implications for Brain Iron Measurements

Ana Maria Oros-Peusquens¹, Miriam Rabea Kubach¹, Maija Laurila¹, N. Jon Shah^1,2

¹Institute of Neurosciences and Biophysics, Research Centre Juelich, Juelich, Germany; ²Faculty of Medicine, Department of Neurology, RWTH Aachen University,  JARA, Aachen, Germany

Relaxation times in pure water and their field dependence is well predicted by the Blombergen-Purcell-Pound theory; this dependence for typical MRI fields is negligible. Proton relaxation times in tissue are different from those of pure water and have a pronounced field dependence possibly due to micro-structural ferritin content, for example. We report a 3-field study of T2/T2* in vivo performed with the aim of providing quantitative information about transverse relaxation in a field range of 1.5T-4T. Further, we discriminate between the linear and quadratic dependence of T2/T2* on field strength and investigate the correlation between the rate increase in regions with clear a linear dependence and their estimated iron content. The relevance of this methodology to brain iron content measurements in vivo is discussed.

14:00         4459.     Dependence of Blood R₂ Relaxation Rate on Echo Spacing Using a CPMG Sequence at 7T

Alexander Graeme Gardener¹, Susan Tracy Francis¹, Penny Anne Gowland¹

¹Sir Peter Mansfield Magnetic Resonance Centre, School of Physics and Astronomy, University of Nottingham, Nottingham, UK

The dependence of measured in-vitro blood R₂ relaxation rate on sequence parameters and sample oxygenation and haematocrit was investigated at 7T using a multi-echo CPMG sequence. Results show that as inter-echo spacing increases, measured R₂ rates increase, with low oxygenation and high haematocrit blood samples showing the largest enhancement. Pure plasma showed little dependency on inter-echo spacing as expected. Results for fitting blood R_2,0 components at 7T are presented, and compared to previous literature studies at lower field strengths.

14:30         4460.     Magnetic Field Correlation Contrast in the Human Brain at 7 Tesla

Maarten J. Versluis^1,2, Andrew G. Webb^1,2, Mark A. van Buchem^1,2, Matthias J P van Osch^1,2

¹Radiology, Leiden University Medical Center, Leiden, Netherlands; ²C.J. Gorter Center for high field MRI, Leiden University Medical Center, Leiden, Netherlands

Magnetic field correlation (MFC) imaging was implemented to quantitatively measure magnetic field inhomogeneities at 7 T. The feasibility of the technique was shown by comparing 3 T and 7 T data of healthy volunteers for different brain regions. MFC measurements were performed at multiple echo times to differentiate between macroscopic and microscopic field inhomogeneities. In vivo MFC imaging at 7 T was found to be more sensitive than at 3 T and it was possible to differentiate between brain regions with different scales of inhomogeneities.

15:00         4461.     Quantitative Analysis of Transverse Relaxation Due to Strongly Magnetized Micron-Sized Spheres Subjected to Unrestricted Diffusion in Gradient Echo and Spin Echo Imaging: Validation of Theory with Experiments and Monte Carlo Simulation

Peter Roland Seevinck¹, Koen Lucas Vincken¹, Gerrit Hendrik van de Maat¹, Chris J.G. Bakker¹

¹Image Sciences Institute, Department of Radiology , University Medical Center, Utrecht, Netherlands

The signal decay time course of FID and SE of an aqueous suspension of strongly magnetized micron-sized spheres was investigated in great detail to explore the influence of volume fraction, field strength and diffusion. Excellent agreement between MR experiments, Monte Carlo simulations and both the theory of the static dephasing regime and the strong field behavior was demonstrated based on R2, R2’ and R2* relaxation rates. This allows accurate prediction FID and SE signal behavior and quantification strongly magnetized spherical particles such as holmium- and iron-loaded microspheres and iron-labeled cells.

13:30         4462.     SPIO: Positive Contrast by Adiabatic IRON Using Repetitve Inversion Pulses

Carsten Warmuth^1,2, Arne Hengerer³, Robert Krieg³, Jörg Schnorr⁴, Michael Reinhardt⁵

¹Visage Imaging GmbH, Berlin, Germany; ²Siemens AG, Erlangen, Germany; ³Healthcare sector, Siemens AG, Erlangen, Germany; ⁴Department of Radiology, Charité, Berlin, Germany; ⁵Bayer Schering Pharma AG, Berlin, Germany

We present a fast positive contrast iron oxide particle imaging technique based on IRON. By using repetitive inversion pulses, suppression of resonant fat and water protons is achieved independent of tissue T1 relaxation times.

14:00         4463.     Observation of Time Dependent Magnetic Field Correlation in the Human Brain

Kathleen A. Williams¹, Kamila Szulc¹, Caixia Hu¹, Jens H. Jensen¹, Joseph A. Helpern^1,2

¹Department of Radiology, CBI, NYU School of Medicine, New York, USA; ²Center for Advanced Brain Imaging, The Nathan Kline Institute, Orangeburg, NY, USA

The magnetic field correlation (MFC) is a quantitative MRI metric that characterizes magnetic field inhomogeneities generated in the brain by iron-rich tissue structures. The theoretically predicted time dependence of the MFC has been previously shown in phantoms. Here the time dependence of the MFC is demonstrated in vivo for human brain. By fitting to a model, the observed time dependence is found to be consistent with field inhomogeneities that vary on a length scale of 10 to 20 microns, which is comparable to the size of known iron-rich tissue structures within the brain.

14:30         4464.     Quantification of the Regional Non-Hemin Iron in Human Brain in Vivo Through the Apparent Transverse Relaxation Rate of the Tissue Water at 4.7T

Fumiyuki Mitsumori¹, Hidehiro Watanabe¹, Nobuhiro Takaya¹

¹National Institute for Environmental Studies, Tsukuba, Ibaraki, Japan

Non-hemin iron (Fe) is known to cause neurodegeneration in several hereditary diseases. It has also been suspected to be closely related to the occurrence of Alzheimer�fs and Parkinson�fs diseases. In the present study we propose a method to estimate the concentration of brain [Fe] in vivo through the high correlation between [Fe] and the apparent relaxation rate (R₂^�õ) of the tissue water in human brain at 4.7T. The estimate was significantly improved when the contribution of macromolecular fraction to R₂^�õ was taken into account. The result of estimation on 54 healthy subjects will be presented.

15:00         4465.     Estimation of T2* in Severe Iron Overload Patients with Weighed Least Squares T2*-IDEAL

Huanzhou Yu¹, Ann Shimakawa¹, Scott B. Reeder², Jean H. Brittain³, Charles A. McKenzie⁴, Shreyas Vasanawala⁵

¹Applied Science Laboratory, GE Healthcare, Menlo Park, CA, USA; ²Departments of Radiology, Medical Physics, Biomedical Engineering and Medicine, University of Wisconsin, Madison, WI, USA; ³Applied Science Laboratory, GE Healthcare, Madison, WI, USA; ⁴Department of Medical Biophysics, University of Western Ontario, London, ON, Canada; ⁵Department of Radiology, Stanford University, Stanford, CA, USA

Hepatic iron overload is a common chronic liver disease. Excess amounts of iron reduce T2* of liver from above 20ms to shorter than 1ms. We have previously demonstrated that T2* mapping can be achieved simultaneously with water-fat separation using a multi-echo T2*-IDEAL algorithm. However, in the presence of very short T2*, the images at late echoes have poor SNR, leading to substantial errors. In this work, we introduce a weighted least squares algorithm, which automatically decreases the impact of later, noisier echoes. We demonstrate in simulations and in-vivo studies that T2* values as short as 1ms can be effectively estimated.

13:30         4466.     Assessment of Iron Concentration in a Two-Pool Model of Brain Tissue

Christian Langkammer^1,2, Thomas Seifert-Held¹, Michaela Soellinger¹, Christian Enzinger¹, Franz Fazekas¹, Rudolf Stollberger², Stefan Ropele¹

¹Department of Neurology, Medical University of Graz, Graz, Austria; ²Institute of Medical Engineering, Graz University of Technology, Graz, Austria

Using a ferritin phantom mimicking the composition of brain gray matter, we investigated the effect of iron load on the saturation line shape of water. Z-spectra were acquired under different saturation conditions by use of spoiled FLASH sequences with spectral presaturation. We observed a linear correlation between the maximum saturation effect (i.e., the direct saturation ratio) and iron concentration. Magnetization transfer and global susceptibility effects did not affect this relationship, favoring the use of this method for in vivo measurements of the iron content.

14:00         4467.     Towards Microcalcifcation Detection Using Susceptibility  Weighted Phase Imaging: Simulations and Phantom Experiments

Richard A. Baheza¹, E. Brian Welch², John C. Gore¹, Thomas E. Yankeelov¹

¹VUIIS, Vanderbilt University, Nashville, TN, USA; ²Philips Healthcare, Philips, Cleveland, OH, USA

The presence of calcium deposits in the breast can be an indicator of malignancy, so the ability to detect such abnormalities by MRI would be significant. Calcium deposits have a different magnetic susceptibility from surrounding tissues that can in principle be observed using susceptibility weighted imaging (SWI). Here we introduce a technique to locate a susceptibility induced signature within a data set by computing the cross-correlation between a template and the target image.

14:30         4468.     Multi-Echo Spin-Echo (MESE) Signal Behavior of Paramagnetic Holmium-166 Loaded Microspheres for Radiotherapy: Experiment and Simulation

Gerrit Hendrik van de Maat¹, Peter R. Seevinck¹, Koen L. Vincken¹, Hendrik de Leeuw¹, Chris J. Bakker^1,2

¹Image Sciences Institute, University Medical Centre Utrecht, Utrecht, Netherlands; ²Department of Radiology, University Medical Centre Utrecht, Utrecht, Netherlands

The potential of multi-echo spin-echo (MESE) measurements for quantification of holmium-166 loaded microspheres (HoMS) was explored. This was done by investigating the signal decay due to HoMS by MR experiments on a gel series containing HoMS and by Monte Carlo simulations. Results were compared to the theoretical model of strong field behavior and appeared to be in good agreement for a wide range of HoMS concentrations. The study indicates the potential of MESE for the quantification of HoMS concentrations up to 20mg/ml.

15:00         4469.     Simultaneous Spin/Gradient Echo Acquisitions in Conjunction with Arterial Spin Labeling for Quantification of T2* by Single Shot 3D Spiral GRASE Imaging

Johannes Gregori¹, Norbert Schuff^1,2, Matthias Günther³

¹Radiology & Biomedical Imaging, University of San Francisco, San Francisco, CA, USA; ²VA Medical Center , Center for Imaging of Neurodegenerate Diseases, CIND, San Francisco; ³Neurology, University Hospital Mannheim, University of Heidelberg, Heidelberg, Germany

Quantification of local T2*, which is used in fMRI as a surrogate for neuronal activity, can be difficult because static (R2’) and irreversible (R2) effects contribute to the decay. In this work, we developed a 3D mapping scheme based on GRASE, in which two spiral-out planar readouts per k-space partition are used to acquire a spin echo (governed by R2) and a time-shifted gradient echo (governed by R2’) on the ascending slope of the corresponding spin echo. From a combination of R2 and R2’, local T2* variations can be quantified. Measurements were also performed in conjunction with arterial spin labeling allowing to selectively measure T2* of labeled water.

14:00         4470.     Efficient and Accurate Modeling of Pulsed Magnetization Transfer.

Dirk K. Müller¹, André Pampel¹, Toralf Mildner¹, Torsten Schlumm¹, Thies H. Jochimsen¹, Harald E. Möller¹

¹Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany

Estimating quantitative magnetization transfer (MT) parameters using pulsed off-resonance saturation is a challenging task. We present a new fitting approach of estimating MT parameters for arbitrary pulse sequences using a fast and accurate description of the pulse sequence by solving the McConnell equations numerically using matrix algebra.

14:30         4471.     Reliability and Reproducibility of  High Resolution Magnetization Transfer Imaging on 3T

Ying Wu^1,2, Hongyan Du³, Kelly H. Zou⁴, Shawn Sidharthan¹, Eugene Dunkle, Robert Edelman^1,2, Ann Ragin²

¹Radiology, NorthShore University HealthSystem, Evanston, IL, USA; ²Radiology, Feinberg School of Medicine, Northwestern University, Chicago, IL, USA; ³Center on Outcomes, Research and Education, NorthShore University HealthSystem, Evanston, IL, USA; ⁴Global Medical R&D, Pfizer Inc, New York, NY, USA

The validity of quantitative MR methods for monitoring progression in Multiple Sclerosis, Alzheimer’s Disease and other chronic neurological diseases is limited by the reliability and reproducibility of the derived measurements. This investigation evaluated the reliability and reproducibility of a promising new high resolution strategy based on quantitative MT. The strategy was evaluated using conventional manually-defined region of interest measurements. In addition, the reliability and reproducibility of high resolution quantitative MT was also evaluated using automated methods aimed at eliminating operator bias further.

15:00         4472.     Quantitative Magnetization Transfer Imaging Using Non-Balanced SSFP

Monika Gloor¹, Klaus Scheffler¹, Oliver Bieri¹

¹Radiological Physics, University of Basel Hospital, Basel, Switzerland

Similar to balanced steady-state free precession (bSSFP), the signal of non-balanced SSFP (nb-SSFP, such as SSFP-FID or SSFP-echo) depends on magnetization transfer (MT) in tissues. Based on a binary spin-bath model, an extended SSFP signal equation is presented which takes MT effects into account. Using this new SSFP signal equation, quantitative MT imaging (qMTI) is demonstrated ex-vivo and in-vivo. High correlation with literature values is observed and furthermore, this method benefits from short acquisition times, high signal-to-noise ratios and off-resonance insensitivity.

15:30         4473.     Cardiac Magnetization Transfer Imaging Using SSFP at 3T

Oliver Michael Weber^1,2, Peter Speier³, Klaus Scheffler¹, Oliver Bieri¹

¹Radiological Physics, University of Basel, Basel, Switzerland; ²Siemens Healthcare, Zurich, Switzerland; ³Siemens Healthcare, Erlangen, Germany

An optimized bSSFP cine sequence was used to calculate magnetization transfer of the myocardium at 3T. MT values of 26% were found in normal myocardium, demonstrating the feasibility of cardiac MT imaging at 3T.

13:30         4474.     Intra- And Inter-Scanner Variability of Magnetization Transfer Ratio Using Balanced SSFP

Monika Gloor¹, Klaus Scheffler¹, Oliver Bieri¹

¹Radiological Physics, University of Basel Hospital, Basel, Switzerland

Magnetization transfer ratio (MTR) has become an important tool to study various tissue abnormalities, such as demyelination in brain white matter. Recently, a new technique for measuring MTR has been proposed based on balanced steady-state free precession (bSSFP) with modified radiofrequency pulses. The reproducibility and variability of MTR-bSSFP was analyzed on six healthy volunteers using two different 1.5T clinical systems. Intra-scanner MTR measurements were well reproducible (< ±0.3 pu) and inter-scanner variation is below 0.5 pu for optimal flip angle settings ([pu]: percentage units). MTR-bSSFP benefits further from relatively low B₁ sensitivity, high signal-to-noise ratios, and short overall acquisition times.

14:00         4475.     3D Amide Proton Transfer (APT) Imaging of the Whole Brain at 3T

He Zhu^1,2, Joseph S. Gillen^1,2, Peter B. Barker^1,2, Peter C.M. van Zijl^1,2, Jinyuan Zhou^1,2

¹Russell H Morgan Department of Radiology and Radiological Science, Johns Hopkins University, Maryland, MD, USA; ²F.M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Maryland, MD, USA

Amide proton transfer (APT) imaging is a novel chemical exchange saturation transfer (CEST) based MRI modality that gives contrast due to endogenous cytosolic protein and peptide content, as well as tissue pH, in vivo. In this abstract, a fast 3D APT technique with whole-brain coverage that is suitable for clinical applications is presented. Preliminary results show that the APT signal in the brain is about 1-2.2% of the water intensity, and the highest APT signal is detected in the cerebellum.

14:30         4476.     Correlation of Potential Myelin Measures from Quantitative Magnetisation Transfer (QMT) and Multi-Component Driven Equilibrium Single Pulse Observation of T₁ and T₂ (McDESPOT)

Rebecca Sara Samson¹, Sean C L Deoni², Claudia A M Wheeler-Kingshott¹

¹Department of Neuroinflammation, UCL Institute of Neurology, London, UK; ²Centre for Neuroimaging Sciences, Institute of Psychiatry, London, UK

It has previously been suggested that both the restricted proton fraction (f_b) measured from quantitative Magnetisation Transfer (qMT) and myelin water fraction (f_m) from multi-component relaxometry are related to myelin content in the brain. Multi-Component Driven Equilibrium Single Pulse Observation of T₁ and T₂ (mcDESPOT) is a promising alternative to spin-echo multi-component T₂ analysis. This pilot study investigates the correlation of f_m measured via the mcDESPOT technique with f_b at 1.5T in white and grey matter in three healthy volunteers.

15:00         4477.     An Improved Empirical Solution for Describing Chemical Exchange Saturation Transfer (CEST) MRI

Phillip Zhe Sun¹

¹Department of Radiology, Harvard Medical School, Athinoula A. Martinos Center for Biomedical Imaging, Charlestown, MA, USA

An empirical solution based on a 2-pool exchange model has been proposed to describe CEST MRI, providing tremendous insights for quantitative CEST imaging. Nevertheless, the empirical solution is only a first order approximation, and susceptible to non-negligible errors when quantifying CEST imaging of fast exchange groups under practical conditions (i.e. intermediate RF powers). To properly guide future development of more sensitive exogenous PARACEST/LIPOCEST agents, we modified the simplistic empirical equation with a second order correction term, which showed good agreement with numerical simulation.

13:30         4478.     Predicting Optimal Properties of CEST MRI Agents Under Practical Experimental Conditions

Phillip Zhe Sun¹, Guanshu Liu^2,3, Jinyuan Zhou^2,4, Peter van Zijl^2,4, Michael T. McMahon^2,4

¹Department of Radiology, Harvard Medical School, Athinoula A. Martinos Center for Biomedical Imaging, Charlestown, MA, USA; ²Russell H. Morgan Department of Radiology & Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD; ³F. M. Kirby Center for functional Brain Imaging, Kennedy Krieger Institute , Baltimore, MD, USA; ⁴F. M. Kirby Center for functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, USA

Chemical exchange saturation transfer (CEST) imaging provides a sensitivity enhancement mechanism that allows measurement of microenvironment properties through dilute labile groups. Because in general CEST MRI contrast increases with the chemical exchange rate, there has been great interest in developing DIACEST/LIPOCEST/PARACEST agents with fast exchange rates. Here, we have examined CEST MRI contrast under practical experimental conditions, and derived the optimal properties of CEST agents at different exchangeable proton shifts from water using both analytical solutions and numerical simulations. In addition we have validated these experimentally for two different types of exchangeable protons.

14:00         4479.     Quantitative Pulsed CEST

Kimberly L. Desmond¹, Greg J. Stanisz^1,2

¹Medical Biophysics, University of Toronto, Toronto, Ontario, Canada; ²Imaging Research, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada

A clinically applicable, quantitative chemical exchange saturation transfer (CEST) technique was developed and demonstrated in a phantom containing 1M ammonium chloride. The fitting algorithm consisted of a two-pool compartmental model of the Bloch equations including exchange and pulsed off resonance excitation modeled by a piecewise constant approximation. The results of the fit provided estimates for T2_a (T2 relaxation for the bulk water pool), M_0c (size of the CEST pool relative to bulk water), T2_c (T2 relaxation of CEST pool), and ω _0c (resonance frequency of the CEST pool).

14:30         4480.     Compensatory Amide Proton Transfer Ratio (CAPTOR) Imaging to Improve the Specificity of Tissue Acidosis MRI

Phillip Zhe Sun¹, A Gregory Sorensen¹

¹Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, Harvard Medical School, Charlestown, MA, USA

Amide proton transfer (APT) imaging has been shown sensitive to microenvironment pH and can detect ischemic tissue acidosis. However, the commonly used magnetization transfer (MT) asymmetry analysis is susceptible to concomitant RF irradiation effects (spillover, intrinsically asymmetric MT etc), and hence, may not be specific to pH-dependent APT mechanism. Here, we propose a compensatory APT ratio (CAPTOR) approach that subtracts the label scan (famide) from the mean of two compensatory images around the amide proton offset instead of a reference scan (-famide). Our data showed that CAPTOR had less heterogeneity among cerebral tissue types, and more specific to APT process.

15:00         4481.     In Vivo Myelin Water Imaging Using Z-Shimmed Multi-Gradient-Echo Pulse Sequences at 3T

Claudia Lenz¹, Markus Klarhöfer¹, Klaus Scheffler¹

¹Division of Radiological Physics, University of Basel Hospital, Basel, Switzerland

Quantitative imaging of the myelin water fraction (MWF) is able to show demyelinating processes and therefore provides insight into the pathology of white matter (WM) diseases such as multiple sclerosis. So far, mapping of the MWF most often was performed using a multi-echo spin-echo sequence and fitting the biexponential T2 decay with a non-negative least-squares algorithm. In this work, a different approach is presented using multi-gradient-echo pulse sequences. This method has been introduced by one study measuring formalin-fixed brains. We present a solution for in vivo measurements using z-shimming as a correction for magnetic field inhomogeneities.

13:30         4482.     Analytical Method for Correction of B1 Errors in High-Field Magnetization Transfer Ratio Mapping

Vasily L. Yarnykh¹

¹Radiology, University of Washington, Seattle, WA, USA

Magnetization Transfer Ratio (MTR) is known to strongly vary with B1. This circumstance restricts MTR applications in high magnetic fields. In this study, a simple analytical model of MTR dependence on B1 was derived. Based on this model, a correction algorithm employing a tissue-independent set of parameters was developed and tested for MTR mapping of the human brain at 3T. MTR correction was shown to be highly accurate across a wide range of B1 non-uniformities. Combination of this algorithm with fast B1 mapping enables whole-brain MTR mapping and histogram analysis on high-field scanners for a variety of neuroimaging applications.

14:00         4483.     Molecular Exchange in Breast Cells Studied with a New DW-MRI Method

Daniel Topgaard¹, Ingrid Åslund, Carin Malmborg², Anna Stenstam², Karin Bryskhe², Stina Oredsson

¹Physical Chemistry, Lund University, Lund, --, Sweden; ²Colloidal Resource

An extension of diffusion weighted (DW) MRI, with two DW periods and an intermediate mixing time, is shown to be sensitive to molecular exchange between slow and fast components. The method is applied to healthy and cancerous breast cells yielding significant differences between the various cell types.

14:30         4484.     Incidental Magnetization Transfer Contrast by Fat Saturation Preparation Pulses in Multi-Slice MR Imaging: An Illustration of Look-Locker EPI Sequences

Wanyong Shin¹, Hong Gu¹, Yihong Yang¹

¹Neuroimaging Research Branch, National Institute on Drug Abuse, Baltimore, MD, USA

It is demonstrated that the frequency selective fat saturation preparation pulses generate incidental magnetization transfer contrast (MTC) effects in multi-slice imaging. Simulation reveals that the fractional signal loss by these MTC effects are severer when smaller flip angle, shorter TR, and larger slice number are applied. Incidental MTC effects induce a signal attenuation (up to 30%) and bias on T1 measurement (up to 20%) when inversion recovery Look-Locker EPI sequences are used. Furthermore, it is shown that water-only excitation pulses have the potential to minimize the signal attenuation and provide unbiased T1 measurement without fat artifacts in MR images.

15:00         4485.     In Vivo Tumor Detection and Characterization by Fixed-Point Imaging Using Active Feedback Fields

Jon Furuyama¹, Dennis W. Hwang¹, Yung-Ya Lin¹

¹Chemistry and Biochemistry, UCLA, Los Angeles, CA, USA

The use of a CW pulse in the presence of strong radiation damping can be seen to produce unique contrast enhancement, and can be a useful supplement to conventional methods in the detection of tissues with small

14:00         4486.     Multi-Parametric Approach to Automatic Regional Analysis of Quantitative Relaxation Times in the Brain

Benjamin Segun Aribisala¹, Andrew M. Blamire¹

¹Newcastle Magnetic Resonance Centre, Newcastle University, Newcastle upon Tyne, Tyne and Wear, UK

A significant step in the analysis of imaging data is accurate definition of regions of interest (ROI) within a single tissue type. Analysis is often done by interactively defining ROI on each image under analysis (e.g. T1 or T2). This approach is sensitive to image resolution which introduces partial volume effects biasing the analysis. We propose a fully automatic multi-parametric approach whereby complementary information in multiple images is considered in order to classify each quantitative image into its tissue classes. We apply this method to brain data and demonstrate it is time efficient and largely free of partial volume effect.

14:30         4487.     Phase Contrast in the Post Mortem Rat Brain: Comparison with T2* and Histology

Ana Maria Oros-Peusquens¹, Sebastian Hirsch¹, Joerg Felder¹, Avdo Celik¹, M Cremer¹, N. Jon Shah¹

¹Institute of Neurosciences and Biophysics, Research Centre Juelich, Juelich, Germany

The origins of phase contrast were investigated in a post mortem rat brain with high isotropic resolution (60&#61549;m) MRI. The phase contrast after formalin fixation was characterised in the whole brain and compared with T2* maps and cellular structure based on histology. Phase images acquired on a post mortem rat brain allow for a more complete visualisation of fine layers in the hippocampus and with higher contrast (factor 10-20) than with magnitude images alone. For hippocampal structures myelin and iron content are not the main contrast-generating factor. Instead, the phase contrast seems more correlated with the cell density as described by the intensity of Nissl staining in histology.

15:00         4488.     How Reliable Are Current Practices in Reconstructing Relaxation Spectra for Detecting the T2 Myelin Water Signal When Applied to Real in Vivo T2 Decays?

Keith S. Cover¹

¹Physics and Medical Technology, VU University Medical Center, Amsterdam, North Holland, Netherlands

The results of this study demonstrated that, for in vivo T2 decays of white matter measured using current practices, bias in both Whittall’s and Mackay’s nonnegative least squares and fitting a few monoexponentials reconstruction algorithms can yield highly reproducible but false positive detections of the T2 myelin water signal. In contrast, data conserving reconstruction matrices (DCRM) yielded highly reliable results but required several times the signal to noise ratio (SNR). More in vivo measurements of a variety of structures and pathologies at high SNR are required to determine the most useful trade off between reliability of the reconstruction algorithm and the SNR.

15:30         4489.     Inconsistency in Interpretation of T₂ and Diffusion in White Matter

Sharon Peled¹

¹Radiology, Brigham and Women's Hospital, Boston, MA, USA

The short T2 component of the T2 curve measured in white matter is commonly assigned to myelin-associated water. A problem with this assignment is that water exchange times through the myelin sheath are not known – even minimal exchange would greatly reduce the size of this component to below the measured levels. Diffusion measurements indicate that such exchange exists. This incompatibility is described and solutions proposed.

13:30         4490.     Variations in T2, T2* and T1 Between White Matter Tracts at 7.0 T

Eleanor F. Cox¹, Pierre-Yves Herve², Peter J. Wright¹, Ashley Loftipur¹, Olivier Mougin¹, Susan E. Pritchard¹, Tomas Paus², Penny A. Gowland¹

¹SPMMRC, School of Physics & Astronomy, University of Nottingham, Nottingham, Nottinghamshire, UK; ²Brain & Body Centre, University of Nottingham, Nottingham, Nottinghamshire, UK

T2 and T2* have been shown to vary between white matter tracts. The aim of this study is to investigate the relationship between the relaxation times and known features of myelin density and axonal structure, by studying adjacent white matter tracts and comparing known differences in histology to the T2, T2* and T1 relaxation times of these tracts, measured at 7T.

14:00         4491.     Quantifying White Matter: Integrating Diffusion Tensor Imaging and Bound Pool Fractions

Nikola Stikov¹, Lee Michael Perry², John Mark Pauly¹, Brian A. Wandell², Robert F. Dougherty²

¹Electrical Engineering, Stanford University, Stanford, CA, USA; ²Psychology, Stanford University, Stanford, CA, USA

We explore combining two quantitative MR contrast mechanisms in order to better understand the structure of white matter in the brain. Diffusion tensor imaging (DTI) can identify white matter fascicles and measure their diffusivity. The bound pool fraction (BPF) estimates the proportion of protons bound to macromolecules, such as myelin. We combine these two in order to obtain concurrent information about the direction, diffusivity, and myelin content of white matter tracts in the corpus callosum and the optic radiation. We conclude that DTI and BPF are complementary techniques that combine to provide a more complete insight into tissue microstructure.

14:30         4492.     High Resolution Myelin Water Imaging in Rat Spinal Cord in Vivo with Actively Decoupled Implanted RF Coil

Piotr Kozlowski¹, Jie Liu², Jenny Ch Tso¹, Nathan Wolfe¹, Andrew C. Yung¹

¹UBC MRI Research Centre, Vancouver, BC, Canada; ²ICORD, Vancouver, BC, Canada

Myelin water imaging has been used to characterize rat spinal cord pathology. However, this technique requires very high SNR and homogeneity of the B1 field, making it challenging to apply this technique in rat spinal cord in vivo. We have designed and built an RF implanted coil, which is actively decoupled from a volume coil. Here we present preliminary results of high spatial resolution myelin water imaging using this system.

15:00         4493.     Intrinsic Connectivity Contrast: A Novel Contrast Mechanism for Investigating a Wide Range of Brain Disorders

R. Todd Constable¹, Yuko Hara², Nallakkandi Rajeevan¹, Fuyuze Tokoglu¹, Xenios Papademetris¹

¹Diagnostic Radiology, Yale University, New Haven, CT, USA; ²Biomedical Engineering, Yale University, New Haven, CT, USA

A novel contrast mechanisms based on resting state connectivity BOLD fMRI data is presented which delineates areas in the gray matter that exhibit altered connectivity relative to control subjects. This approach is applicable to any neurodegenerative disorder and provides a new tool for localizing disrupted tissues.

13:30         4494.     Pulsed Saturation Transfer for Quantifying CEST in the Presence of MT

Kimberly L. Desmond¹, Greg J. Stanisz^1,2

¹Medical Biophysics, University of Toronto, Toronto, Ontario, Canada; ²Imaging Research, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada

The quantification of the CEST effect in-vivo is complicated by the presence of immobile macromolecules, which additionally results in a large MT effect. The MT component has a significant impact on the magnitude of the CEST effect, which reduces the validity of asymmetry-based techniques. We propose a method by which the parameters for both CEST and MT pool can be determined independently by fitting a three pool model to data from a pulsed saturation transfer experiment. This method is tested on a set of three phantoms, containing 1M ammonium chloride, 2% agar, and both compounds combined.

14:00         4495.     Multeslice Gradient Echo CEST Imaging

W Thomas Dixon¹, Ileana Hancu¹, A Dean Sherry^2,3, James Ratnaker², Donald E. Woessner², Elena Vinogradov⁴, Robert E. Lenkinski⁴, David C. Alsop⁴

¹GE Global Research, Niskayuna, NY, USA; ²UT Southwestern Medical Center, Dallas, TX, USA; ³Chemistry, U Texas Dallas, Richardson, TX, USA; ⁴Beth Israel Deaconess Medical Center, Brookline, MA

Contrast Enhanced Saturation Transfer imaging usually uses long CEST RF pulses to achieve the maximum, steady-state effect. Most transmitters do not provide long pulses. Long pulses slow imaging greatly or require rapid imaging methods that often limit image quality. A 2D gradient echo sequence with a short CEST pulse inserted before each observe pulse retains all the flexibility of the original gradient echo sequence. Keeping pulse and wait durations constant, increasing the number of slices increases TR and increases sensitivity just as in 3D imaging. The CEST effect was independent of slice position in a uniform CEST phantom.

14:30         4496.     Using the Linewidth of the Amide Proton CEST Effect of MRI-PARACEST Agents for PH Mapping

Alex Xuexin Li^1,2, Craig K. Jones¹, Mojmir Suchy^1,3, Robert H.E. Hudson³, Ravi S. Menon^1,2, Robert Bartha^1,2

¹Center for Functional and Metabolic Mapping, Robarts Research Institute, London, ON, Canada; ²Department of Medical Biophysics, The University of Western Ontario, London, ON, Canada; ³Department of Chemistry, The University of Western Ontario, London, ON, Canada

A technology to measure pH using the amide protons site of a thulium complex with a DOTAM-Glycine-Lysine ligand: Tm³⁺-DOTAM-Gly-Lys was developed. The pH can be determined uniquely from the linewidth of the asymmetry curve of the CEST spectrum in a manner independent of contrast agent concentration, intrinsic magnetization transfer effect, and temperature for a given saturation pulse. pH maps with standard deviation less than 0.1 pH units were obtained from 10 mM Tm³⁺-DOTAM-Gly-Lys solutions with different pH.

15:00         4497.     Application of Parallel Imaging to Reduce SAR in CEST Experiments

Elena Vinogradov¹, Aaron K. Grant¹, Phillip M. Robson¹, Ilena Hancu², W Thomas Dixon², A Dean Sherry^3,4, Robert E. Lenkinski¹

¹Department of Radiology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, USA; ²GE Global Research, Niskayuna, NY, USA; ³Department of Chemistry, University of Texas at Dallas, Dallas, TX; ⁴Advanced Imaging Research Center, University of Texas Southwestern Medical Center, Dallas, TX, USA

Chemical Exchange Saturation Transfer (CEST) contrast employs selective presaturation of the exchanging pool and subsequent observation of the reduction of the free water signal. Many potential applications of CEST agents, in particular Paramagentic CEST agents (PARACEST) are limited by the RF power required, that may exceed FDA approved SAR limits. Combination of Parallel Imaging (PI) with CEST should allow increased TR, and, hence, reduced SAR, without sacrificing image resolution or total scan time. Preliminary data shown here demonstrate combination of CEST with PI, and shows that the PI reconstruction does not alter the observable CEST effects.

13:30         4498.     Amide Proton Transfer (APT) MRI: A 3T Vs. 7T Comparison

Guang Jia¹, Jinyuan Zhou², Seongjin Choi¹, Wenbo Wei¹, Steffen Sammet¹, Daniel James Clark¹, Stefan Fischer³, Michael V. Knopp¹

¹Department of Radiology, The Ohio State University, Columbus, OH, USA; ²Department of Radiology, Johns Hopkins University, Baltimore, MD, USA; ³Philips Medical Systems, Cleveland, OH, USA

The frequency difference between amide proton (3.5 ppm) and water resonance is 1043 Hz at 7T, much greater than 448 Hz at 3T, which may help reduce mixture of APT-MTR with the direct saturation effect to water signal. Our results shows better separation between the amide proton saturation and free water saturation profiles and a clearer dip that reflects the APT effect in the MT-spectrum of egg white at 7T than that at 3T. This study strongly suggests that the APT sequence should be extensively evaluated at ultra-high-field (7T) for tissue characterization.

14:00         4499.     Modeling the Influence of TR and Excitation Flip Angle on the MTR Obtained from 3D FLASH MRI

Gunther Helms¹, Henning Dathe², Peter Dechent¹

¹MR-Research in Neurology and Psychiatry, University of Göttingen, Göttingen, Germany; ²Biomechanics Group, Dept. of Orthodontics, Universityhospital, Göttingen, Germany

Modeling and optimization of the magnetization transfer ratios (MTR) obtained from FLASH MRI is often based on continuous-wave models that only implicitly account for excitation and relaxation during TR. These effects were explicitely modeled by an approximation of free relaxation between the RF pulses and verified on the human brain in vivo at 3T by variation of flip angle and TR. The influence of inhomogeneous RF fields could be readily incorporated. Choice of short TR and larger flip angles improved the MTR contrast and reduced the influence of RF inhomogeneity.

14:30         4500.     Optimization of Quantitative Magnetization Transfer Imaging Using a Selective Inversion Recovery Pulse

Ke Li^1,2, Zhongliang Zu^1,2, Junzhong Xu^1,2, John C. Gore^1,2, Heather M. Whitney^1,3, Daniel F. Gochberg^1,2

¹Vanderbilt University Institute of Imaging Science, Nashville, TN, USA; ²Department of Radiology and Radiological Sciences, Vanderbilt University, Nashville, TN, USA; ³Department of Physics and Astronomy, Vanderbilt University, Nashville, TN, USA

Quantitative magnetization transfer (qMT) imaging maps properties of the tissues that are usually interpreted in terms of two pools of protons, corresponding to free and immobilized fractions. The selective-inversion-recovery fast-spin-echo (SIR-FSE) qMT technique developed recently includes an inversion time (ti) which is varied between 3.5 ms and 10 s, while the delay before the next sequence repetition (td) is held constant. qMT parameters are determined by fitting the resulting recovery to a bi-exponential function of ti using an approximate solution. In the current study, we employ a new protocol that varies both ti and td and fits the data with minimal approximations. Cramer-Rao lower bounds (CRLB) are calculated to select the variations in both ti and td that will maximize the precision-per-unit-time. Monte Carlo simulations support this approach by showing a large reduction in the resulting qMT parameter uncertainties. The optimization results are also confirmed by measurements on a series of BSA phantoms with different percent weight.

15:00         4501.     Impact of Magnetization Transfer on Relaxometry Using Transient Steady-State Free Precession Imaging

Jan Ole Blumhagen¹, Francesco Santini¹, Oliver Weber¹, Klaus Scheffler¹, Oliver Bieri¹

¹Division of Radiological Physics, University of Basel Hospital, Basel, Switzerland

Several methods proposed the use of the transient response of balanced steady-state free precession (bSSFP) for quantification of relaxation times and spin densities. However, it has been shown that magnetization transfer (MT) affects the steady state of bSSFP and is dependent on pulse duration and repetition time. During transition to steady state also MT effects built up, thus it is likely that MT affects the estimation of relaxation parameters using transient SSFP methods. In this work, we will show exemplarily for inversion recovery bSSFP that significant deviations arise from MT.

14:00         4502.     The Use of Binomial Pulses for Suppression of Long T2 Signals

Atsushi Mark Takahashi¹

¹Applied Science Laboratory, GE Healthcare, Menlo Park, CA, USA

Ultra short TE (UTE) imaging has the promise of visualizing tissues not previously visible with MRI. However, contrast in UTE is typically fairly flat because little T2* decay occurs prior to data acquisition. In this work, we propose the use of binomial excitation pulses as a way to selectively excite spins based on their intrinsic T2.

14:30         4503.     A Flexible Design Algorithm for Single-Shot 2D Circular/Elliptical OVS RF Pulses

Xiaoping Wu¹, Nate Powell¹, Malgorzata Marjanska¹, Michael Garwood¹, Kâmil Ugurbil¹, Pierre-François Van de Moortele¹

¹Center for Magnetic Resonance Research and Radiology Department, University of Minnesota, Minneapolis, MN, USA

Outer Volume Suppression (OVS) is used to saturate unwanted MR signals in spectroscopy. Typically, OVS is performed with multiple conventional (e.g., sinc) pulses, suffering from low efficiency and high SAR. We have previously

15:00         4504.     High Resolution Reduced-FOV Single-Shot FSE Imaging Using 2DRF Pulses with Less Blurring and Low SAR

Jing Yuan¹, Tzu-Cheng Chao², Yi Tang¹, Lawrence P. Panych³

¹Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA; ²National Taiwan University, Taipei; ³Department of Radiology, Brigham and Women's Hospital, Harvard Medical School,, Boston, MA, USA

RFOV imaging was developed for SS-FSE sequence using 2DRF pulses to reduce echo train length (ETL) and scan time. High resolution, low SAR and less blurring can be achieved in RFOV images due to the shorter ETL. In this study, scan time and blurring reductions were demonstrated in phantom images with reduction factors from two to five. 2.5X higher resolution (320*320) RFOV brain images were achieved within an even shorter time compared to that required for low-resolution (128*128) full FOV images. Significant blurring reduction results were also compared to full FOV images with the same nominal resolution.

15:30         4505.     Selective Composite Adiabatic Refocusing Pulses for 3D RARE Imaging

Ziqi Sun¹

¹Davis Heart and Lung Research Institute, The Ohio State University, Columbus, OH, USA

A compact selective composite adiabatic refocusing pulse with nonlinear phase dispersion compensation capability was developed for 3D RARE imaging. The composite adiabatic pulse consisted of two adiabatic half passage (AHP) pulses with phase reversal in the second AFP pulse. A full echo train and significantly improved signal sensitivity and uniformity were observed using the selective composite adiabatic refocusing pulse in comparison to that of an amplitude-modulated selective refocusing pulse in 3D RARE imaging.

13:30         4506.     An On-The-Fly Radiofrequency Pulse for Bilateral Excitation with Independently Modulated Phase

Ernesto Staroswiecki^1,2, John M. Pauly², Bruce Lewis Daniel¹, Brian Andrew Hargreaves¹

¹Radiology, Stanford University, Stanford, CA, USA; ²Electrical Engineering, Stanford University, Stanford, CA, USA

In this abstract we present a method to excite multiple slabs with arbitrary radiofrequency pulses while modulating the phase of each slab independently, allowing the simultaneous acquisition of multiple volumes without imaging the non-excited volume between the slabs. We show an example of a bilateral acquisition where the peak B1 amplitude is only marginally higher than that of a single slab with the same RF pulse, and that results in negligible signal loss.

14:00         4507.     Multi-Frequency Parallel Transmit Fopr Multi-Slice Scan Acceleration

Laura Sacolick¹, Florian Wiesinger¹, Schulte F. Rolf¹, Dong Chen¹, Guido Kudielka¹, Wolfgang Loew¹, Mika W. Vogel¹

¹Imaging Technologies, GE Global Research, Munich, Bavaria, Germany

The majority of applications thus far for parallel RF transmission in MRI have focused on addressing problems of B1 homogeneity. Here an eight-channel parallel transmit system is demonstrated for a novel application in reducing scan time as well as SAR. The multiple transmit channels are used to produce RF pulses simultaneously at multiple carrier frequencies. These are used to excite and simultaneously acquire four slices, which are then reconstructed by SENSE. The rungs of a body TEM coil assigned to excite each slice are chosen to minimize SAR, and are phase shimmed to maximize B1 field homogeneity. Multi-frequency parallel transmit is used here to give significant scan acceleration that can be combined with multiple receive coil acceleration, and applied in a wide variety of applications.

14:30         4508.     Counteracting RF Inhomogeneity on the Upper Legs at 3T Using Strongly Modulating Pulses

Nicolas Boulant¹, Alexandre Vignaud², Pierre G. Carlier³

¹CEA, NeuroSpin, Gif sur Yvette, France; ²Siemens Healthcare, St Denis, France; ³Pitié Salpêtrière University Hospital, Institute of Myology , Paris, France

Strongly modulating pulses have been used to homogenize the flip angle on upper legs at 3 T. The muscle and fat tissues were discriminated in the pulse design algorithm via their resonance frequencies to achieve different flip angles. By implementing a “double Ernst angle”, we demonstrate a higher signal to noise ratio for both tissues (on average) in addition to a better signal uniformity in spoiled gradient echo experiments. Measurements of the flip angle using the AFI sequence are also reported.

15:00         4509.     RF Pulses for in Vivo Spectroscopy at High Field Designed Using Optimal Control

Gerald B. Matson^1,2, Karl Young^1,3, Lana G. Kaiser^1,3

¹CIND, Veterans Affairs Medical Center, San Francisco, CA 94121, USA; ²Pharmaceutical Chemistry, University of California, San Francisco, CA, USA; ³Radiology, University of California, San Francisco, San Francisco, CA, USA

Localized in vivo spectroscopy at high magnetic field strength (> 3 T) is susceptible to localization artifacts such as the chemical shift artifact and the spatial interference artifact for J-coupled spins. This latter artifact results in regions of anomalous phase for J-coupled spins. The use of selective RF pulses with increased bandwidth can minimize the signals with anomalous phase, while further suppression of anomalous phase signals can be accomplished with selective saturation pulses. Here we demonstrate through examples that optimal control methods can provide new, improved bandwidth localization and saturation pulses for improved localized spectroscopy at high field.

13:30         4510.     Electrodynamic Analysis of SAR and Transmit Homogeneity for RF Shimming on a Dielectric Cylinder

Riccardo Lattanzi^1,2, Yudong Zhu¹, Daniel K. Sodickson¹

¹Center for Biomedical Imaging, NYU School of Medicine, New York, NY, USA; ²Sackler Institute of Graduate Biomedical Sciences, NYU School of Medicine, New York, NY, USA

This work explores the electrodynamic behavior of transmit homogeneity and SAR for RF shimming. Ultimate SAR was calculated for uniform target excitation profiles both in a transverse and in a coronal plane through the center of a homogeneous cylinder. The trade-off between profile fidelity and SAR minimization was investigated in the ultimate case, as well as in the case of transmit arrays with increasing numbers of elements. Our simulations enforced homogeneity not only in the amplitude but also in the phase of the excitation profile, and the results suggest that this becomes a particularly stringent constraint at high field strengths.

14:00         4511.     Slice-Selective Array-Optimized Composite Pulse for Simultaneous Improvement in Excitation Uniformity and Reduction of SAR

Bu Sik Park^1,2, Kyunghyun Sung³, John McGarrity², Zhipeng Cao^2,4, Sukhoon Oh², Mark Watson⁵, Scott Schilliak⁵, Labros Petropolous⁵, Christopher M. Collins^2,4

¹Bioengineering , The Pennsylvania State University, Hershey, PA, USA; ²Radiology, The Pennsylvania State University, Hershey, PA, USA; ³Radiology, Standford University, Stanford, CA, USA; ⁴Bioengineering, The Pennsylvania State University, Hershey, PA, USA; ⁵MR Instruments Inc., Minneapolis, MN, USA

To overcome challenges of inhomogeneous transmit B1 distribution and high specific absorption rate (SAR) at high field, we present a slice-selective array-optimized composite pulse designed to both improve B1 uniformity and reduce SAR using an 8-channel transmit head array loaded with a head model, and compare our result with the conventional quadrature driving method at 3T. The sequence was implemented on a 3D MRI simulator currently under development, and showed improvement in excitation uniformity and SAR compared to a conventional birdcage coil.

14:30         4512.     Experimental Estimation of Local SAR in a Multi-Transmit System

Ulrich Katscher¹, Christian Findeklee¹, Tobias Voigt²

¹Philips Research Europe - Hamburg, Hamburg, Germany; ²Technical University Karlsruhe, Karlruhe, Germany

SAR management is a widely discussed topic at high main fields, particularly in the framework of multi-transmit systems. Typically, local SAR is estimated via patient model simulations, which, however, are too extensive for clinical routine. This study investigates a possible alternative local SAR estimation by post-processing the RF transmit element's sensitivities. To this goal, the recently published approach "Electric Properties Tomography" was adapted for non-quadrature RF excitation, typically applied in multi-transmit systems, e.g., for RF shimming. Local SAR distributions obtained in phantom experiments with an 8-channel transmit system at 3T show a high correlation with corresponding local SAR simulations.

15:00         4513.     Imaging Conductivity and Local SAR of the Human Brain

Tobias Voigt¹, Olaf Doessel¹, Ulrich Katscher²

¹Institute of Biomedical Engineering, University of Karlsruhe, Karlsruhe, Germany; ²Philips Research Europe, Hamburg, Germany

The approach "Electric Properties Tomography" is applied to estimate electric conductivity and local SAR distributions in vivo. Mean conductivity values for different brain areas agree quantitatively with literature values. Corresponding local SAR generated by a Tx/Rx body coil at 1.5T is shown.

13:30         4514.     A Novel Three Dimensional Radiofrequency Pulse for Small Voxel Excitation

Natee Ina¹, Suwit Saekho^1,2

¹Radiological Technology, Chiang Mai University, Muang, Chiang Mai, Thailand; ²Biomedical Engineering Center, Chiang Mai University, Muang, Chiang Mai, Thailand

The three dimensional Radio Frequency pulse (3D RF) has benefit to many applications. The major limitation of this pulse is long pulse length. A novel 3D RF pulse design was based on the small tip angle approximation combining the multi-shot and the half-pulse scheme for a small volume excitation. Numerical integration of the Bloch equation demonstrated that the two shot pulses along the x-y direction weighted with two half pulses along z- direction for cylindrical excitation at the diameter and thickness of 12 mm and 10 mm respectively, provides 34% shorter pulse width compared to that of conventional two-shot pulse.

14:00         4515.     A Dual-Band Three-Dimensional Tailored RF Pulse for Simultaneous Susceptibility Artifact and B1+ Inhomogeneity Reduction

Cungeng Yang¹, Weiran Deng¹, V. Andrew Stenger¹

¹Department of Medicine, University of Hawaii, Honolulu, HI, USA

Susceptibility artifacts and B1+ inhomogeneity are major limitations in high field MRI. Three-dimensional (3D) RF pulses are useful for reducing B1+ inhomogeneity and spectral spatial pulses have been shown to reduce the through-plane signal loss susceptibility artifact. We present a dual-band 3D RF pulse design for simultaneously reducing susceptibility induced signal loss and B1+ inhomogeneity. The method is demonstrated in T2* weighted brain images at 3T using an RF body coil. The pulses were observed to work well for multiple slices and several volunteers. The method can easily be extended to parallel transmission applications or more complex B1+ profiles.

14:30         4516.     Correction of Parallel Transmit RF Pulses at 9.4 T Using Measured Gradient Waveforms

Xiaoping Wu¹, J. Thomas Vaughan¹, Kâmil Ugurbil¹, Pierre-François Van de Moortele¹

¹Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, MN, USA

Parallel Transmission (PT) allows for applying accelerated multidimensional spatially selective RF pulses through independent Transmit (Tx) RF channels, a feature especially promising at very high magnetic fields to compensate for B1 field inhomogeneity. However, gradient waveform distortions due to gradient system imperfections and/or eddy current can result in very poor excitation profiles. It has been shown that such degradation of excitation patterns can be effectively reduced by using measured k-space trajectories for RF pulse design. In the present work, we conducted PT experiments on a 9.4 T human scanner and measured k-space trajectories to calculate corrected RF pulses, resulting in dramatic excitation accuracy improvement.

15:00         4517.     An Efficiency Metric for K-Space Trajectories

Christopher J. Hardy¹, Luca Marinelli¹

¹GE Global Research, Niskayuna, NY, USA

To maximize speed, k-space trajectories should be designed to allow efficient traversal under conditions of constrained gradient amplitude and slew rate. We employ a rate remapping algorithm to always push gradient amplitude or slew rate to the limit over the whole course of the trajectory, creating time-optimal gradient waveforms. Given constant extent of 2D k-space coverage, the average density of coverage is assumed to be proportional to the total length of the trajectory. Trajectory length divided by the time of traversal can then be taken as an efficiency metric. Various trajectories are compared using this metric.

14:00         4518.     Microstructural Assessment of Dental Tissues by Quantitative MRI Using Ultra-Short Echo Times (UTE): Initial In-Vivo Evaluation

Saïd Boujraf^1,2, Christian Hofmann^1,3, Johannes Ulrici⁴, Erich Hell⁴, Bernd Haller³, Volker Rasche¹

¹Department of Internal Medicine II, University Hospital of Ulm, Ulm, Germany; ²Department of Biophysics and Clinical MRI Methods, Faculty of Medicine and Pharmacy of Fez, Fez, Morocco; ³Department of paraodontology, Dental University Clinic of Ulm, Ulm, Germany; ⁴Sirona Dental Systems GmbH, Bensheim, Germany

This work demonstrates the application of in-vivo UTE-MRI for assessing hard-tissue features of teeth with sufficient spatial-resolution and tissue-contrast within reasonable scan time.

14:30         4519.     In-Vitro Quantification of Dental Filling Induced Artifacts in Dental Magnetic Resonance Imaging Using Ultrashort Echo Time (UTE) at 3 Tesla

Saïd Boujraf^1,2, Christian Hofmann^1,3, Regina Maschka³, Johannes Ulrici⁴, Erich Hell⁴, Bernd Haller³, Volker Rasche¹

¹Department of Internal Medicine II, University Hospital of Ulm, Ulm, Germany; ²Department of Biophysics and Clinical MRI Methods, Faculty of Medicine and Pharmacy of Fez, Fez, Morocco; ³Department of paraodontology, Dental University Clinic of Ulm, Ulm, Germany; ⁴Sirona Dental Systems GmbH, Bensheim, Germany

The aim of this study is assessing the sensitivity of UTE-MRI to artifacts induced by dental filling-materials, and to compare the resulting artifact levels in UTE Images to conventional 3D-spin-echo and 3D-gradient-echo techniques

15:00         4520.     Microstructural Assessment of Dental Tissues by Quantitative MRI Using Ultra-Short Echo Times (UTE): In-Vitro Evaluation

Saïd Boujraf^1,2, Christian Hofmann^1,3, Johannes Ulrici⁴, Erich Hell⁴, Bernd Haller³, Volker Rasche¹

¹Department of Internal Medicine II, University Hospital of Ulm, Ulm, Germany; ²Department of Biophysics and Clinical MRI Methods, Faculty of Medicine and Pharmacy of Fez, Fez, Morocco; ³Department of paraodontology, Dental University Clinic of Ulm, Ulm, Germany; ⁴Sirona Dental Systems GmbH, Bensheim, Germany

The aim of this study is to proof the feasibility of UTE-MRI for assessing teeth morphology, including enamel, dentin, pulpa and root-material.

15:30         4521.     Imaging of Short T2 Species Using a Dual Adiabatic Inversion Recovery Ultrashort TE (DIR UTE) Sequence

Jiang Du¹, Christine B. Chung¹, Mark Bydder¹, Won Bae¹, Atsushi M. Takahashi², Graeme M. Bydder¹

¹Radiology, University of California-San Diego, San Diego, CA, USA; ²Global Applied Science Laboratory, GE Healthcare Technologies, Menlo Park, CA, USA

Imaging of short T2 tissues requires not only a short echo time (TE) but also efficient suppression of long T2 tissues in order to maximize the short T2 contrast and dynamic range. This paper introduces a novel method of long T2 suppression using two long adiabatic inversion pulses, with the first pulse inverting the longitudinal magnetization of long T2 water protons and the second one inverting the longitudinal magnetization of fat protons. Ultrashort TE (UTE) acquisition starts when both inverted magnetization reach the nulling point. The feasibility of this technique was demonstrated on phantom, cadaveric samples and healthy volunteers.

13:30         4522.     Improved Cardiac Shim Using Field Map Estimate from Multi-Echo Dixon Method

Peter Kellman¹, Diego Hernando², Saurabh Shah³, Sven Zuehlsdorff³, Andreas Greiser⁴, Renate Jerecic³

¹National Institutes of Health, Bethesda, MD, USA; ²University of Illinois, Urbana, IL, USA; ³Siemens Medical Solutions, Chicago, IL, USA; ⁴Siemens Medical Solutions, Erlangen, Germany

Rapid field variation across the heart cause by tissue-air interface presents a challenge to cardiac imaging. An improved method for shimming is presented which estimates the fieldmap in the presence of rapid field variation and in regions containing both fat and water. The fieldmap is estimated using a multi-echo GRE acquisition and VARPRO solution for fat and water components. The multi-echo GRE sequence was implemented with parallel imaging reconstruction as a real-time, free-breathing, non-ECG gated 2D multi-slice acquisition, acquiring the volume in <7s. With the proposed shim method, the field variation in the user defined heart region was reduced significantly.

14:00         4523.     Fat Suppressed FSE Pelvic MR Imaging with Two New FSE Based Dixon Pulse Sequences: A Comparison of IDEAL FSE and FSE Triple Echo Dixon

Russell Norman Low^1,2, Neeraj Panchal^1,2

¹Sharp and Children's MRI Center, San Diego, CA, USA; ²San Diego Imaging Medical Group, San Diego, CA, USA

We review our initial clinical experience with two new Dixon-based FSE sequences for fat suppressed pelvic MRI. Traditional fat-suppressed FSE imaging is often degraded by artifacts caused by magnetic field inhomogeneities. The large FOV required for pelvic imaging can exacerbate these artifacts. IDEAL FSE and Fast Spin Echo Triple Dixon (FTED) are two new sequences which generate fat-only and water-only images insensitive to magnetic field inhomogeneities. We explore the use of IDEAL FSE and FTED for pelvic MRI and compare them to conventional fat suppressed FRFSE T2-weighted imaging for image quality, homogeneity of fat suppression, artifacts, and lesion detection.

14:30         4524.     Comparison of Susceptibility-Weighted Imaging and Magnetic Field Correlation Imaging

Caixia Hu¹, Jens H. Jensen¹, Kathleen Williams¹, Maria Fatima Falangola¹, Joseph A. Helpern¹

¹Radiology, New York University, New York, NY, USA

Susceptibility-weighted imaging and magnetic field correlation imaging are two recently introduced techniques that yield image contrast sensitive to magnetic field inhomogeneities generated by iron rich regions within the brain. Here these two techniques are directly compared, and it is shown that they quantify distinct aspects of the pattern of field inhomogeneities.

15:00         4525.     Fast, High Resolution T2* Mapping Using 3D MGE and 3D EPI, with 3D Correction for Macroscopic Dephasing Effects

Simon Robinson¹, Markus Barth², Jorge Jovicich¹

¹Centre for Mind/Brain Sciences, University of Trento, Trento, Trentino, Italy; ²Centre for Cognitive Neuroimaging, Donders Institute for Brain, Cognition and Behaviour, Nijmegen, Netherlands

T2* maps reflect local iron concentration but are limited in usefullness by dephasing due to static magnetic field gradients. We explore the utility of high resolution 3D EPI and 3D MGE sequences in reducing dephasing, and the correction of residual effects using a 3D model based on fieldmaps, which are calculated from phase information. Results are presented for the basal ganglia, which are of clinical interest in a number of pathologies. 3D MGE with acceleration factor 2, the fastest sequence tested, emerges as a reliable means to perform whole-brain submillimetre T2* mapping at 4T in just over 5 minutes.

13:30         4526.     Reference Phantom Validation of T2-Mapping: Maximum Likelihood Estimation of T2 from Magnitude Phased-Array Multi-Echo Data

Clemens Bos¹, Adri Duijndam¹, Julien Sénégas²

¹Philips Healthcare, Best, Netherlands; ²Philips Research Europe, Hamburg, Germany

High-resolution T2 maps are often biased by noise signal that is rectified and mimics persistent long T2 signal. Maximum likelihood estimation with explicit inclusion of noise statistics in the signal model has the potential to mitigate this bias, and to minimize the variance of the resulting T2-map. Maximum likelihood estimation of T2 was validated on the eurospin reference phantom. T2 values were accurate well within 5%. Maximim likelihood estimation is compatible with phased-array coils and parallel imaging. In conclusion, maximum likelihood estimation has utility for accurate and precise pixel-wise calculation of T2.

14:00         4527.     T_1ρ -Weighted MRI Senses Partial Pressure of Oxygen

Dharmesh R. Tailor¹, James E. Baumgardner², Walter R. T. Witschey³, Ravinder Reddy¹

¹Radiology, University of Pennsylvania, Philadelphia, PA, USA; ²Anesthesia, University of Pennsylvania, Philadelphia, PA, USA; ³Biochemistry & Molecular Biophysics, University of Pennsylvania, Philadelphia, PA, USA

T_1ρ-Weighted MRI is demonstrated as a technique for monitoring partial pressure of oxygen, an important physiologic parameter with obvious functional and clinical implications, with a sensitivity that is higher than that of both T₁ and T₂ weighting. The possibility of T_1ρ-weighted MRI for imaging localized P_O2 changes is also demonstrated in the rat and human brain.

14:30         4528.     MR Contrast Generated by Altering Parameters of Adiabatic Pulses: Theoretical Simulations and in Vivo  MRS Results

Silvia Mangia¹, Timo Liimatainen¹, Michael Garwood¹, Pierre-Gilles Henry¹, Shalom Michaeli¹

¹CMRR, Dept. of Radiology, University of Minnesota, Minneapolis, MN, USA

Rotating frame relaxation rates are measured during radiofrequency (RF) irradiation, and therefore can be altered by choosing different RF settings. The present work focuses on rotating frame relaxation methods based on adiabatic pulses, exploring how MR contrast can be created by using different pulse modulation functions, maximum amplitude or bandwidth of the adiabatic pulse. Results from theoretical simulations and animal MRS experiments are presented.

15:00         4529.     Contrast Enhancement for Early Tumor Detection by Active Feedback Self-Nutation

Sophia Y. Yang¹, Fang Liu¹, Dennis W. Hwang¹, Yung-Ya Lin¹

¹Chemistry and Biochemistry, UCLA, Los Angeles, CA, USA

Active feedback fields can selectively lock spin components with a small frequency shift and concentration difference within an inhomogeneous sample to generate contrast, which would otherwise be difficult to obtain in conventional imaging. From our experiment with early tumor mice, we show that this concept can be applied for early tumor detection.

13:30         4530.     Intra and Inter-Site Reproducibility of Myelin Water Volume Fraction Values Derived Using McDESPOT

Sean CL Deoni¹, Rebecca Samson², Claudia A.M. Wheeler-Kingshot²

¹Centre for Neuroimaging Sciences, London, UK; ²Department of Neuroinflammation, UCL Institute of Neurology, London, UK

Multi-component Driven Equilibrium Single Pulse Observation of T1 and T2 (mcDESPOT) is a promising alternative to conventional T2-based approaches to multi-component relaxometry. An important attribute of any quantitative imaging method is the technique’s reproducibility, both across multiple time-points and across different imaging centres. Here we quantify the intra and inter-site reproducibility of mcDESPOT-derived myelin water volume fraction estimates, demonstrating high repeatability with intra and inter-site coefficients of variations of 0.069 and 0.07, respectively. Further, voxel-wise paired t-tests revealed no time-point or imaging site-specific biases.

14:00         4531.     Evaluation of K-T SENSE for Cardiac Imaging of Rats at 9.4T

Anthony N. Price¹, Kenneth K. Cheung¹, Shaihan J. Malik², Mark F. Lythgoe¹

¹Centre for Advanced Biomedical Imaging, UCL Institute of Child Health, University College London, London, UK; ²Imaging Sciences Department, Hammersmith Hospital Campus , Imperial College London, London, UK

In recent years parallel imaging and dynamic undersampling have become routinely available in clinical cardiac MRI. However, these techniques have been slow to translate to the pre-clinical setting. In this abstract we present a full implementation and quantitative evaluation of up to 5x dynamically undersampled data using a four-channel array coil on a 9.4T experimental system. k-t SENSE reconstruction can provide an accurate estimation of LV ejection fraction in rats, using complete short axis cine data of 0.2 x 0.2 x 1mm resolution taking only 2-3 minutes to acquire.

14:30         4532.     Preserving Signal Contrast in Multi-Slice Black Blood Fast Spin Echo

Ananth J. Madhuranthakam¹, Jesse L. Wei², Jean H. Brittain³, Neil M. Rofsky², David C. Alsop²

¹Applied Science Laboratory, GE Healthcare, Boston, MA, USA; ²Radiology, Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, MA, USA; ³Applied Science Laboratory, GE Healthcare, Madison, WI, USA

When double inversion recovery (DIR) is used with single shot fast spin echo (SSFSE) to achieve multi-slice black blood images, the signal contrast is preserved in the first slice but degrades in later acquired slices due to the absence of sufficient recovery time. This is especially problematic for long T₁ species, for instance, lesions with moderately prolonged T₁ causing difficulty in diagnosis due to alterations in normally seen contrast. In this work, we demonstrate a technique to achieve black blood by using a preparation sequence to suppress fast moving spins while preserving signal contrast in all single shot slices.

15:00         4533.     Comparison of Black Blood and Bright Blood Cardiac MR Imaging by Prospective- And Retrospective ‘wireless’ Gating Methods for Evaluation of Mouse Heart Function at 9.4T

Sandra Margaretha Bovens^1,2, Bernard C. te Boekhorst¹, Meike van den Berk¹, Kees W. van de Kolk¹, Arno Nauerth³, Niels Gadellaa¹, Marcel G. Nederhoff^1,2, Gerard Pasterkamp¹, Michiel ten Hove¹, Cees J. van Echteld⁴

¹Cardiology, University Medical Center Utrecht, Utrecht, Netherlands; ²InterUniversity Cardiology Institute of the Netherlands, Utrecht, Netherlands; ³Bruker Biospin MRI GmbH, Germany; ⁴Novartis Institutes for BioMedical Research, Basel, Switzerland

CMR is often used in mice, for the determination of cardiac function. We introduced a fast, multislice black blood approach with a ‘wireless’ retrospective gating method. The CNR of retrospective gating methods are higher than the prospective methods. There are no significant differences between the bright- and black blood methods, or the prospective and retrospective methods. The introduced ‘wireless’ multislice black blood method is fast and the SNR is comparable to that of bright blood methods. The inter-observer variability decreases, thus showing a fast, reliable ‘wireless’ alternative for the determination of cardiac function with a black blood sequence.

14:00         4534.     Optimization of Variable-Density Cartesian Sampling for Time-Resolved Imaging

Reed F. Busse¹, Kang Wang², James H. Holmes¹, Jean H. Brittain¹, Frank R. Korosec³

¹Applied Science Laboratory, GE Healthcare, Madison, WI, USA; ²Medical Physics, University of Wisconsin, Madison, WI, USA; ³Radiology, University of Wisconsin, Madison, WI, USA

A number of Cartesian sampling patterns were considered to optimize system performance for time-resolved imaging applications. Two k-t sampling density functions were compared, a two-value “top-hat” function and a continuous function proportional to 1/kr. The effect k-t sampling distribution, including bit-reversed time ordering and point-wise k-space interleaving was also assessed. Modeling studies simulating rapid contrast arrival in an angiography application were performed. Reconstruction errors due to temporal dynamics – temporal blur and coherent artifacts – were found to be lowest for the approach with 1/kr k-t sampling density and bit-reversed point-wise interleaving of k-t samples.

14:30         4535.     A Low Curvature Uniformly Sampled Trajectory for 3D Imaging: The Interlocking Loops Trajectory

Eric Wong¹

¹University of California, San Diego, La Jolla, CA, USA

K-space trajectories with uniform sampling and low curvature allow for fast sampling. We introduce here a novel 3D trajectory based on two interlocking loops that uniformly samples a sphere, with low curvature at all points.

15:00         4536.     Shells with Integrated RadiaL and Spiral (SWIRLS): An Improved Shells K-Space Trajectory

Yunhong Shu¹, Matt A. Bernstein¹

¹Radiology, Mayo Clinic, Rochester, MN, USA

An improved method to sample k-space on a spherical surface is described. The novel trajectory, named shells with integrated radial and spiral (SWIRLS), eliminates the need for two helical spiral interleaves that cover the “polar icecap” region in the original shells trajectory. Phantom and volunteer experiments demonstrate the feasibility of the SWIRLS method. It has improved temporal efficiency compared to the shells trajectory, while overcoming the problems with phase discontinuity.

15:30         4537.     On Optimal Encoding of Flow in Three-Directional Phase-Contrast Sequences

Francesco Santini¹, Michael Markl², Klaus Scheffler¹

¹Radiological Physics, University of Basel Hospital, Basel, Switzerland; ²Medical Physics, University Hospital Freiburg, Freiburg, Germany

Three-directional flow encoding by means of phase-contrast sequence is becoming popular for clinical applications. In order to keep reasonable scan times, there is often the need to compromise on temporal resolution of the acquired dataset. In this work, an analysis of the behavior of the standard flow encoding method with respect to the dynamics of the velocity waveform is provided. An alternative reconstruction algorithm, implementing a “view sharing” principle along the velocity direction is presented, which improves the time resolution at no additional cost. The optimized reconstruction method is validated in vivo.

13:30         4538.     Improved Radial Spoiled Gradient Echo Imaging with Randomized RF Phases and Gradient Moments

Wei Lin¹, Hee Kwon Song¹

¹Department of Radiology, University of Pennsylvania, Philadelphia, PA, USA

Conventional spoiled gradient echo imaging with quadratic RF spoiling generates non-ideal signal intensities, particularly at short repetition times and large flip angles. This translates to significant errors in various quantitative applications based on T₁-weighted image contrast. This work proposes a novel spoiling scheme, based on random gradient moments and RF phases, in conjunction with the radial acquisition scheme to suppress artifacts caused by TR-to-TR signal variations. The proposed method achieves ideal spoiling within a wide range of flip angles (0⁰-90⁰) and repetition times (5-3000 ms). Simulations and phantom experiments demonstrate the superior performance over conventional RF spoiling.

14:00         4539.     Effects of Relaxation During RF Pulses on the Signal Intensity of Spoiled Gradient Echo Sequences

Nicolas Boulant¹

¹CEA, NeuroSpin, Gif sur Yvette, France

We present an analysis of the steady state signal in the presence of relaxation during RF pulses in spoiled gradient echo sequences. It is shown that a minor attenuation of the magnetization vector, of the order of one to some percent, can have dramatic consequences on the measured signal in the short TR/long T1 applications, yielding in some cases a substantial loss of signal at the Ernst angle. Measurements on a phantom at 3 teslas are performed to verify this analysis.

14:30         4540.     Simultaneous Acquisition of Gradient Echo / Spin Echo BOLD and Perfusion with a Separate Labeling Coil

Christopher Glielmi¹, Qin Xu¹, Richard Cameron Craddock^2,3, Xiaoping Hu¹

¹Department of Biomedical Engineering, Georgia Institute of Technology / Emory University, Atlanta, GA, USA; ²School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, USA; ³Department of Psychiatry and Behavioral Sciences, Emory University, Atlanta, GA, USA

We present a new sequence that complements arterial spin labeling with simultaneous acquisition of gradient echo and spin echo BOLD. These three contrasts have complementary characteristics and simultaneous acquisition facilitates comparison of each measure’s specificity and reproducibility.

15:00         4541.     Recovery of Signal Loss Due to In-Plane Susceptibility Gradients in Gradient Echo EPI by Acquiring Extended Phase-Encoding Lines

Kwan-Jin Jung^1,2

¹BIRC, University of Pittsburgh, Pittsburgh, PA, USA; ²Bioengineering Department, University of Pittsbrguh, Pittsburgh, PA, USA

The in-plane susceptibility gradient causes an echo shift which results in signal loss and ripple artifact in gradient echo imaging. This become more severe in the gradient echo EPI due to a low gradient strength applied in the phase-encoding direction during the long echo train. An approach of using the z-shim in the PE direction was developed. However, this method has the same disadvantage of increasing the scan time two or three times as in the through-plane z-shimming. A new approach is to acquire the echo signals that are shifted outside the regular data acquisition time by acquiring more echoes outside the regular TDAQ. This will recover the signal loss as well as remove the ripple artifact without increasing the scan time significantly.

13:30         4542.     Field Map Measurements Using a TrueFISP Sequence

Sven Zuehlsdorff¹, Peter J. Weale¹, Saurabh Shah¹, Andreas Greiser², Renate Jerecic¹

¹Siemens Medical Solutions USA, Inc., Chicago, Il, USA; ²Siemens AG Healthcare Sector, Erlangen, Germany

A homogeneous static magnetic field (B0) is essential for many clinical applications requiring TrueFISP imaging. Typically, 3D gradient echo (GRE) sequences are used estimate the B0-field inhomogeneity. However, these approaches are typically susceptible to motion (such as cardiac or respiratory motion) and flow. In this study, the spectral response function of a 2D TrueFISP sequence was utilized to estimate deviations of the B0-field. Field maps of different body regions such as brain and heart have been obtained.

14:00         4543.     Simultaneous B1 and T1 Mapping Based on Modified "Actual Flip-Angle Imaging"

Tobias Voigt¹, Ulrich Katscher², Kay Nehrke², Olaf Doessel¹

¹Institute of Biomedical Engineering, University of Karlsruhe, Karlsruhe, Germany; ²Philips Research Europe, Hamburg, Germany

We present a new, fast, and accurate simultaneous B1 and T1 mapping method based on the "Actual Flip angle Imaging" (AFI) sequence. In terms of accuracy and signal to noise ratio it seems to outperform standard AFI B1 mapping. The quantitative T1 estimation of the presented approach is in good agreement with reference T1 measurements. The approach has been successfully tested with healthy volunteers.

14:30         4544.     Modulated Repetition Time Look Locker (MORTLL):  a Method for Rapid High Resolution Three Dimensional T1 Mapping

Neville Dali Gai¹, John Butman¹

¹Radiology & Imaging Sciences, National Institutes of Health, Bethesda, MD, USA

A modification of the Look-Locker (LL) technique that enables high resolution T1 mapping over the physiologic range of intracranial T1 values is presented. The modifications include the use of a 3D balanced SSFP acquisition (for high SNR and resolution) along with variable repetition time to allow effective full recovery of longitudinal magnetization. The technique makes it possible to image a volume of thin contiguous slice with high resolution and accuracy using a simple fitting procedure for three LL phases in a reasonable time and is particularly well suited for imaging long T1 species. The T1 values obtained in WM, GM and CSF are compared with those obtained using inversion recovery spin echo imaging (gold standard) as well as with a 8 phase LL with a three parameter model and correction described in literature. T1 values obtained in multiple phantoms and from six volunteers showed excellent agreement with IR-SE T1 values.

15:00         4545.     A Multi-Spectral Three-Dimensional Acquisition Technique for Imaging Near Metal Implants

Kevin M. Koch¹, John E. Lorbiecki¹, R Scott Hinks¹, Kevin F. King¹

¹Applied Science Laboratory, GE Healthcare, Waukesha, WI, USA

Metallic implants used in bone and joint arthroplasty induce severe spatial perturbations to the B0 magnetic field used for high-field clinical magnetic resonance. These perturbations distort slice-selection and frequency encoding processes applied in conventional two-dimensional techniques and hinder the diagnosis of complications from arthroplasty. Here, a method is presented whereby multiple three-dimensional fast-spin-echo images are collected at offset RF transmission and reception frequencies. It is demonstrated that this technique can be used to generate a composite image that is devoid of slice-plane distortion and possesses greatly reduced distortions in the readout direction, even in the immediate vicinity of metallic implants.

13:30         4546.     Phase Accrual During Excitation in Ultrashort TE (UTE) Imaging:  an Alternate Definition of TE for Phase Measurements

Jing-Tzyh Alan Chiang¹, Michael Carl², Mark Bydder¹, Jiang Du¹, Robert F. Mattrey¹, Graeme Bydder¹

¹Radiology, University of California San Diego, San Diego, CA, USA; ²Applied Science Laboratory, GE Health Care

MR imaging of tissues with submillisecond T2’s is possible with UTE (ultrashort TE) sequences that utilize half RF excitation pulses and radial projection readouts. For UTE sequences, TE is customarily defined to be from the end of the half RF pulse to the beginning of radial projection readout. We present data here that show phase accrual in UTE imaging to be more than expected for this customary definition of TE. Numerical calculations show this is largely due to phase accrual before the end of the RF pulse, and we suggest an alternate definition of TE for phase calculations in UTE.

14:00         4547.     Assessment of Motion Sensitized Driven Equilibrium (MSDE) Improvement for Whole Brain Application

Makoto Obara^1,2, Masatoshi Honda³, Yutaka Imai³, Marc Van Cauteren¹, Kagayaki Kuroda²

¹Philips Electronics Japan, Minato-ku, Tokyo, Japan; ²Graduate School of Information Science and Engineering, Tokai University, Hiratsuka, Kanagawa, Japan; ³Department of Radiology, Tokai University, Isehara, Kanagawa, Japan

Improvement of MSDE for whole brain application was assessed in human volunteers study. The signal obtained by the improved MSDE sequence was compared with a conventional MSDE sequence at cerebrum area. The improved MSDE sequence achieved both better image homogeneity than when acquired with the conventional MSDE and the signal drop typical in conventional MSDE is absent. It is indicating that improved MSDE sequence is less sensitive to B0 and B1 inhomogeneity and EC effects. Therefore, it may be appropriate for whole brain application.

14:30         4548.     Non-Contact Tracking of Involuntary Head Motions by Ultra-Wideband Radar for Improved High- And Ultra-High Field Magnetic Resonance Brain Imaging

Florian Thiel¹, Mathias Hein², Jürgen Sachs², Ulrich Schwarz², Tomasz Lindel¹, Frank Seifert¹

¹Physikalisch-Technische Bundesanstalt (PTB), Berlin, Germany; ²Ilmenau University of Technology

Subject motion appears to be a limiting factor in numerous MR imaging applications, especially at high and ultrahigh fields. For head imaging the subject’s ability to maintain the same head position for a considerable period of time places restrictions on the total acquisition time. In particular, head tremor, may render certain high-resolution techniques inapplicable. Furthermore, if the very high spatial resolution offered by ultrahigh-field MR scanners shall be exploited, the displacements caused by respiration and cardiac activity have to be considered. Thus, we propose applying a novel method, based on an ultra-wideband radar technique to monitor involuntary head displacements.

15:00         4549.     Non-Linear Effects at High-Speed Moving Table Imaging

Ulrich Katscher¹, Oliver Lips¹

¹Philips Research Europe - Hamburg, Hamburg, Germany

Continuously moving table imaging is an alternative to multi-station techniques for, e.g., MR angiography. In the past years, the provided table speed increased from typically ~1 cm/s up to ~5 cm/s. Assuming a long-term continuation of this trend of increasing table speed, one has to keep in mind that at table speeds approaching ~3&[middot]10⁹ cm/s, non-linear, relativistic effects has to be taken into account. This study investigates possible consequences, particularly on scan time and RF field distortions due to the Lorentz transformation. The study is based on simulations due to the current lack of suitable MR systems.

14:00         4550.     SVD Eigenimage Based SENSE

Yu Li¹, Feng Huang¹, Wei Lin¹, Charlie Saylor¹, Arne Reykowski¹

¹Advanced Concept Development, Invivo Diagnostic Imaging, Gainesville, FL, USA

In this study, an eigenimage theory was developed for SENSE. Using singular value decomposition (SVD), the full FOV image can be reconstructed by summing several eigenimages weighted by some coefficients calculated from the eigenvalues of a sensitivity matrix. Based on this theory, a data-driven regularization technique was proposed to overcome the challenges of inaccurate sensitivity maps and noise amplification in SENSE reconstruction. This is a new technique that can offer simplicity and efficiency in SENSE g-factor control and artifact suppression.

14:30         4551.     High Pass SENSE

Feng Huang¹, Yu Li¹, Wei Lin¹, Charlie Saylor¹, Arne Reykowski¹

¹Invivo Corporation, Gainesville, FL, USA

A method to reduce noise/artifact level in images reconstructed by SENSE through artificial image sparsity is presented. The sparsity of an image can be artificially increased through a high pass filter in k-space. With constraint of sparsity, the image reconstructed by regularized SENSE with high pass filtered data can be efficiently denoised. G-factor map is used to produce the sparsity-regularization map. Experiments show that the proposed method reconstructs images with reduced noise level than conventional SENSE, and 1D net acceleration factor 4 can be achieved with an 8-channel coil.

15:00         4552.     An Unsupervised Method to Enhance Both SNR and Edges for PPI

Weihong Guo¹, Feng Huang²

¹Department of Mathematics, University of Alabama, Tuscaloosa, AL, USA; ²Advanced Concept Development, Invivo Corporation, Gainesville, FL, USA

Partially parallel imaging (PPI) techniques reduce acquisition time at the cost of signal to noise ratio (SNR). In this work, an unsupervised adaptive method is proposed to reduce noise/artifact level, as well as to sharpen edges. This method is based on Non-local Means (NL-Means). Results of the application to GRAPPA, with both phantom and in vivo data, demonstrate that the proposed method is able to increase SNR, to preserve the fine structures, and to sharpen the edges at the same time.

15:30         4553.     GRAPPA Operator for Wider RadiaL Band (GROWL)

Wei Lin¹, Feng Huang¹, Yu Li¹, Charles Saylor¹, Arne Reykowski¹

¹Invivo Corp., Philips Healthcare, Gainesville, FL, USA

A self-calibrated parallel imaging reconstruction method is proposed for azimuthally undersampled radial dataset. GRAPPA extrapolation operators were used to widen each radial view into a band consisting of several parallel lines, followed by a standard regridding procedure. Self-calibration is achieved by regridding central k-space region, where Nyquist criterion is satisfied, to a rotated Cartesian grid. Applying the proposed reconstruction method to in vivo radial gradient echo images demonstrated a removal of most streaking artifacts at data reduction factors R = 4 and 8, using an eight-element coil array.

13:30         4554.     Atlas Based Sparsification of Image and Theoretical Estimation (ABSINTHE)

Eric Pierre¹, Nicole Seiberlich², Stephen Yutzy¹, Jean Tkach^2,3, Mark Griswold^2,3

¹Biomedical Engineering, Case Western Reserve University, Cleveland, OH, USA; ²Radiology, Case Western Reserve University; ³University Hospitals of Cleveland

In GRAPPA, higher acceleration factors can be achieved with less noise enhancement when reconstructing sparse images. In this study, the ABSINTHE technique has been developed to render undersampled in vivo brain images sparser by removing the “normal” brain information using Principal Component Analysis with a training set of similar brain images. The resulting undersampled sparse images are then reconstructed using GRAPPA. The effectiveness of ABSINTHE for reconstructing undersampled simulated and in-vivo data is demonstrated, and an improved image quality in terms of lower artifact powers is shown for ABSINTHE in comparison to the standard GRAPPA technique.

14:00         4555.     Reconstruction Efficiency and Accuracy Improvement in Real Time Dynamic Parallel Imaging Through Explicit Tracking of Calibration Information

Roger Nana¹, Xiaoping Hu¹

¹Biomedical Engineering, Emory University / Georgia Institute of Technology, Atlanta, GA, USA

This paper presents an approach for improving image reconstruction in dynamic parallel imaging in cases with cyclic changes in coil calibration information. The variation of the calibration information is ascertained by a consistency measure, and the cyclic nature of the variation is used to improve the reconstruction efficiency and accuracy by allowing time frames to share calibration information and avoid using inconsistent calibrating frames. The method is demonstrated using TGRAPPA with free-breathing cardiac imaging data. The method is simple and robust and can be applied to other real-time parallel dynamic image reconstructions

14:30         4556.     O-Space Imaging: Tailoring Encoding Gradients to Coil Profiles for Highly Accelerated Imaging

Pelin Aksit Ciris¹, Jason Peter Stockmann¹, Lick-Kong Tam¹, Robert Todd Constable^1,2

¹Biomedical Engineering, Yale University, New Haven, CT, USA; ²Diagnostic Radiology & Neurosurgery, Yale University, New Haven, CT, USA

O-space imaging is a novel MRI method which achieves very short imaging times by collecting highly independent information at each acquisition, through selecting encoding gradients that complement spatial localization provided by coils. Instead of conventional phase encoding, alternative encoding schemes with projection directions orthogonal to coil sensitivity profiles and each other, are employed. O-space imaging was evaluated using a radially varying gradient field and a circumferentially distributed coil array, with various noise levels, number of coils, and encoding schemes. Results identified superior encoding schemes, and indicate feasibility of high acceleration factors with accurate knowledge of the field.

15:00         4557.     Parallel Imaging Technique Using Localized Gradients (PatLoc) Reconstruction Using Orthogonal Mode Decomposition

Fa-Hsuan Lin¹, Thomas Witzel¹, Jonanthan Polimeni¹, Juergen Hennig², Gerrit Schultz², John W. Belliveau¹, Lawrence L. Wald¹

¹A. A. Martinos Center, Massachusetts General Hospital, Charlestown, MA, USA; ²Department of Diagnostic Radiology, medical Physics, University Hospital Freiburg, Freiburg, Germany

Using conventional imaging sequence, the parallel imaging technique using localized gradients (PatLoc) system requires combining different gradient coils to generate two sets of gradient fields in order to complete “frequency” and “phase” encodings. Previously it was suggested using two orthogonal circular symmetric (OCS) fields to accomplish this requirement. Here we propose to use singular value decomposition (SVD) to automatically decompose the gradient information and suggest the linear combinations for two orthogonal gradients for the pulse sequence. Such orthogonal mode (OM) PatLoc fields were demonstrated with improved image reconstruction in up to 4-fold accelerated acquisitions.

13:30         4558.     Comparison of TSENSE, K-T SENSE and PINOT Fast Imaging Methods on Cine MRI

Lei Hou Hamilton¹, Javier Acebrón Fabregat², David Moratal², Senthil Ramamurthy³, Sebastian Kozerke⁴, Marijn E. Brummer⁵

¹School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, USA; ²Universidad Politécnica de Valencia, Valencia, Spain; ³Children’s Healthcare of Atlanta, Atlanta, GA, USA; ⁴University and ETH Zurich, Zurich, Switzerland; ⁵School of Medicine, Emory University, Atlanta, GA, USA

In this study we compare the performance of three MRI fast imaging methods, TSENSE, k-t SENSE and PINOT on real cardiac cine images. These methods combine parallel imaging with UNFOLD, k-t BLAST and Noquist, respectively. These methods have been implemented under similar net reduction factors. A direct comparison of reconstructed images and an analysis of noise levels show that TSENSE and k-t SENSE have better SNR but tend to blur edges, while PINOT, with no filtering or interpolation, favorably preserves edge detail at a cost of decreased SNR.

14:00         4559.     Comparison of Kt-SENSE and Kt-GRAPPA Applied to Cardiac Cine and Phase Contrast Imaging

Bernd André Jung¹, Sebastian Kozerke²

¹Dept. of Diagnostic Radiology, Medical Physics, University Hospital, Freiburg, Germany; ²Institute for Biomedical Engineering, ETH Zürich, Switzerland

To reduce scan time in time-resolved imaging, advanced parallel imaging techniques have been introduced such as kt-SENSE, kt-GRAPPA, and PEAK-GRAPPA as an extension of kt-GRAPPA. In this work, kt-SENSE was compared to PEAK-GRAPPA in terms of error, noise behavior and temporal fidelity for cardiac Cine and phase contrast data.

14:30         4560.     Optimized Parallel Imaging for Dynamic PC-MRI with Multi-Directional Velocity Encoding

Hsu-Hsia Peng¹, Simon Bauer², Hsiao-Wen Chung¹, Juergen Hennig², Bernd Jung², Michael Markl²

¹Department of Electrical Engineering, National Taiwan University, Taipei, Taiwan; ²Department of Diagnostic Radiology, Medical Physics, University Hospital Freiburg, Freiburg, Germany

In dynamic phase contrast (PC) MRI using standard GRAPPA several central k-space (ACS) lines are acquired separately for each velocity direction and each cardiac phase for calculating the reconstruction weights (WS). We developed two methods, which calculated WS with reduced ACS. The effects on image quality and flow quantification were compared to fully-sampled data, standard GRAPPA, and TGRAPPA. The results show that the two proposed methods can improve scan efficiency while maintaining image quality and accuracy of measured flow or tissue velocities. Compared to TGRAPPA, the proposed methods were more accurate in evaluating flow velocity. In conclusion, the proposed reconstruction strategies are promising for dynamic multi-directionally encoded acquisitions.

15:00         4561.     Extending GRAPPA Kernels to 4D: Application on Time-Resolved 3D Phase Contrast Imaging

Bernd André Jung¹, Simon Bauer¹, Michael Markl¹

¹Dept. of Diagnostic Radiology, Medical Physics, University Hospital, Freiburg, Germany

Improved scanner performance and recent methodological development enable multi-dimensional data acquisition such as time-resolved 3D phase contrast imaging with three-directional velocity encoding. To speed up acquisition times for 3D data acquisition, parallel imaging techniques have been introduced. For time-resolved imaging, parallel imaging techniques have been developed such as kt-SENSE and kt-BLAST, kt-GRAPPA, and PEAK-GRAPPA as an extension of kt-GRAPPA. Previous applications of these methods focused on time-resolved 2D imaging. The aim of this work was to extend spatio-temporal PEAK-GRAPPA acceleration to reconstruction for time-resolved 3D phase contrast data acquisition to fully exploit data redundancy along all spatial and temporal directions.

13:30         4562.     TV Regularization for High-Pass GRAPPA with Higher Net Acceleration Factor

Xiaojing Ye¹, Yunmei Chen¹, Feng Huang²

¹Department of Mathematics, University of Florida, Gainesville, FL, USA; ²Invivo Corporation, Gainesville, FL, USA

High-pass GRAPPA (hp-GRAPPA) suppresses the central calibration signal to reduce image support. When the number of ACS lines is limited, this suppression results in insufficient calibration signal which causes residual aliasing artifacts. We propose a total variation regularized GRAPPA technique to calculate supplemental calibration signal for hp-GRAPPA. The experimental results, with comparisons with conventional GRAPPA and hp-GRAPPA, show that the proposed method can generate images with lower noise/artifact level when only 32 ACS lines are used with reduction factor 4. This work enables hp-GRAPPA with limited ACS lines, and hence increases the net acceleration factor while preserving the image quality.

14:00         4563.     In-Vivo-Sensitivity-Based Regularization of Parallel MR Image Reconstruction

Qi Duan¹, Ricardo Otazo¹, Jian Xu², Daniel K. Sodickson¹

¹Radiology, Center for Biomedical Imaging, NYU School of Medicine, New York, NY, USA; ²Siemens Medical Solutions USA Inc., New York, NY, USA

This abstract demonstrates the technical feasibility of regularization of parallel MR image reconstruction using in vivo coil sensitivities. This new approach not only combines coil sensitivity calibration and regularization of the inverse problem, but also provides better signal-to-noise ratio performance than a corresponding two-step scenario without substantially increasing residual aliasing artifacts. Ultimately, this approach could be applied to any SENSE-based parallel MRI technique, especially at high acceleration factor and when pure coil sensitivity profiles are difficult to estimate.

14:30         4564.     In Vivo Receive Sensitivity Measurement

Jinghua Wang¹, Maolin Qiu¹, June Watzl¹, Robin de Graff¹, Robert Todd Constable¹

¹Diagnostic Radiology, Yale School of Medicine, New Haven, CT, USA

Inhomogeneous receive sensitivity, which is object-dependent, strongly affects the uniformity of MRI signal intensity acquired with parallel imaging techniques at high field. Various methods have been proposed to estimate the receive sensitivity in order to reduce spatial variations in signal intensity caused by the receive sensitivity nonuniformities. A primary challenge is to separate the contribution of tissue contrast, inhomogeneous transmit field and receive sensitivity to signal intensity. Here, we propose a novel rotating object method to estimate the receive sensitivity of a transmit/receive coil. The proposed method is validated using phantom experiments at 4.0 and 7.0 T.

15:00         4565.     Updating Pre-Scan Sensitivity Maps with the Minimum Number of ACS Lines

Feng Huang¹, Yu Li¹, Wei Lin¹, Charlie Saylor¹, Arne Reykowski¹

¹Invivo Corporation, Gainesville, FL, USA

There are advantages to use pre-scann sensitivity maps, such as intensity correction, higher net acceleration factor, etc. However, if there is motion between pre-scan and the target acquisition, then serious aliasing artifacts may occur because of the misregistered sensitivity maps. It is shown in this work that the misregistration can be corrected with as few as 3 extra auto-calibration signal (ACS) lines. The quality of the reconstructed image can be significantly improved with the updated sensitivity maps.

14:00         4566.     Reconstruction of Retrospectively-Gated Cardiac Data Using a Combination of GRAPPA, SPACE-RIP, UNFOLD and an Adaptive Regularization Scheme

Tzu-Cheng Chao^1,2, William Scott Hoge³, Jing Yuan⁴, Hsiao-Wen Chung², Bruno Madore⁴

¹Department of Radiology, Brigham and Women's Hospital, Harvard Medical School,, Boston, MA, USA; ²Department of Electrical Engineering, National Taiwan University, Taipei, Taiwan; ³Department of Radiology,, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA; ⁴Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA

A novel retrospectively-gated reconstruction approach is proposed for accelerated cardiac cine imaging. The approach combines advantageous features from fast-imaging methods such as GRAPPA, SPACE-RIP, UNFOLD, kt-SENSE, SHRUG and GEYSER. The result is a reliable algorithm, tested up to acceleration factors of 6.3 (subsampling factor of 8), and featuring fast reconstruction processing.

14:30         4567.     High Spatial Frequencies Are More Dynamic Than Low Spatial Frequencies in Cardiac Motion

Mireia Sanz Blasco¹, Sumati Krishnan², David Moratal¹, Senthil Ramamurthy³, Marijn E. Brummer²

¹Universitat Politècnica de València, València, Spain; ²Pediatrics/CRC, Emory University, Atlanta, GA, USA; ³Children's Healthcare of Atlanta, Atlanta, GA, USA

A spatiotemporal spectral analysis is presented of kt-space in cine images in the presence of several types of cardiac motion. The temporal spectral content of each k-space view is quantified by a single spectral coefficient. Paradoxically, comparison of these coefficients between k-space views for phantom simulation data and in short-axis cardiac MRI data reveals greater relative changes in high spatial frequencies than in low spatial frequencies.

15:00         4568.     Real-Time Shallow-Breathing Cardiac MRI Using Patient-Adaptive Parallel Imaging

Behzad Sharif¹, John Andrew Derbyshire², Anthony Z. Farnesh², Robert J. Lederman², Yoram Bresler¹

¹Coordinated Science Lab, Department of Electrical & Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL, USA; ²Cardiovascular Branch, NHLBI, National Institutes of Health, DHHS, Bethesda, MD, USA

The goal of real-time cardiac MRI is to reconstruct high quality images from data acquired with no cardiac synchronization. The previously proposed PARADISE imaging technique achieves high resolutions for non-gated breath-hold imaging by optimally adapting data acquisition and image reconstruction to the signal support information and coil sensitivities. We propose a modified PARADISE scheme that allows for shallow breathing and can capture the breathing-induced beat-to-beat cardiac motion variability. By comparing the results to breath-hold gated cine, it is demonstrated that the proposed technique achieves high quality artifact-free images with desirable spatiotemporal resolution.

15:30         4569.     An Auto-Calibrated K-T Approach for Fast 3D Dynamic Contrast-Enhanced Abdominal Imaging

Peng Lai¹, Anja C. Brau¹, Philip J. Beatty¹, Shreyas Vasanawala², Manojkumar Saranathan³

¹Applied Science Laboratory, GE Healthcare, Menlo Park, CA, USA; ²Department of Radiology, Stanford University, Stanford, CA, USA; ³Applied Science Laboratory, GE Healthcare, Rochester, MN, USA

This study proposed a new fast imaging approach, kats ARC (Autocalibrating Reconstruction for Cartesian Sampling with k- & adaptive-t-space data synthesis) for accelerating 3D dynamic contrast-enhanced MRI. The proposed method utilizes a new undersampling pattern compatible with 2D acceleration and exploits contrast kinetics information in autocalibration signals to adaptively determine the temporal interpolation window. Its effectiveness was investigated in comparison with conventional fast imaging methods based on simulation and validated in a patient study. Both simulation and in-vivo studies showed that kats ARC could improve overall image quality and temporal fidelity than conventional methods. This work demonstrates that kats ARC is a promising technique for highly-accelerated 3D DCE abdominal imaging.

13:30         4570.     Accelerated Acquisition and Reconstruction of Non-CPMG Fast Spin Echo Sequences

Ken-Pin Hwang^1,2, Patrick Le Roux¹, Xiaoli Zhao³, Zhiqiang Li³

¹Applied Science Laboratory, GE Healthcare, Waukesha, WI, USA; ²Department of Imaging Physics, University of Texas MD Anderson Cancer Center, Houston, TX, USA; ³MR Engineering, GE Healthcare, Waukesha, WI, USA

Non-CPMG FSE produces a stable echo train in the presence of phase variations at the beginning of the train, but creates two signal families that must both be fully acquired for recombination. A close examination of the signals suggest that they are related by a slowly varying function that can be estimated with just the low order portion of k-space. This forms the basis for a new method for accelerating non-CPMG acquisitions by eliminating artifacts that occur with single encoding of non-CPMG data. The technique is demonstrated for both Cartesian single shot FSE and multishot PROPELLER applications.

14:00         4571.     Spline-Based Variational Reconstruction of Variable Density Spiral K-Space Data with Automatic Parameter Adjustment

Benedicte Delattre¹, Jean-Noël Hyacinthe¹, Jean-Paul Vallée¹, Dimitri Van De Ville^2,3

¹Faculty of Medicine, University of Geneva, Geneva, Switzerland; ²Biomedical Imaging Group (BIG), Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland; ³Work supported in part by the Center for Biomedical Imaging (CIBM), Geneva and Lausanne, Switzerland

Small-animal cardiac imaging is very challenging because we face with several problems like resolution or flux artifacts. One possible way to assess them is the use of non-Cartesian acquisition scheme like variable density spiral. Regridding reconstruction, which is the most popular alternative, however introduces noticeable artifacts due to k-space interpolation, especially when dealing with undersampled trajectories. We propose a variational approach where the image is described by a spline model and where an automatic adjustment of the regularizing weight is implemented. We evaluate our framework for various degrees of the spline model and different orders of derivation of the regularizer.

14:30         4572.     Spiral MR Reconstruction Using FOCUSS

Hisamoto Moriguchi¹, Yutaka Imai¹

¹Radiology, Tokai University, Isehara, Kanagawa, Japan

It has recently demonstrated that the focal underdetermined system solver (FOCUSS) can successfully be adapted to angularly undersampled projection reconstruction (PR). Images reconstructed using PR-FOCUSS are of high quality. In this study, FOCUSS can be extended to spiral trajectories with reduced spiral interleaves. The newly proposed method is referred to as �eSpiral FOCUSS�f. Images reconstructed using Spiral FOCUSS substantially reduce aliasing artifacts from those reconstructed using conventional reconstruction methods, e.g. gridding. The algorithm can also be applied to general non-uniform sampling methods. Spiral FOCUSS is a quite useful reconstruction technique that permits faster spiral acquisition while maintaining image quality.

15:00         4573.     Automatic Regularization for Magnetic Resonance Inverse Imaging

Aapo Nummenmaa¹, Matti S. Hamalainen¹, Fa-Hsuan Lin^1,2

¹Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, MA, USA; ²Institute for Biomedical Engineering, National Taiwan University, Taipei, 106, Taiwan

We propose a simple method for automatic regularization of dynamic magnetic resonance Inverse Imaging (InI). Regularization is interpreted in a Bayesian way, as a variance parameter of a Gaussian prior, and marginal likelihood is used to estimate these parameters. The proposed method is compared to the presently used ad hoc regularization of InI by using empirical data from a visual stimulation experiment. Possible extension of the method for dynamic modeling of the regularization parameters is discussed.

13:30         4574.     Optimal Combination and Filtering for 7 T – Phase Images

Shan Yang¹, Kai Zhong¹, John Grinstead², Oliver Speck¹

¹Biomedical Magnetic Resonance, Otto-von-Guericke-University, Magdeburg, Germany; ²Siemens Medical Solutions USA Inc., Portland, USA

Images based on signal phase showed superior and complementary contrast at 7 Tesla compared to magnitude images. However, due to the strong RF inhomogeneity at 7 T and a lack of a body coil, it is difficult to obtain sensitivity map for each individual array element. In this study, we systematically compared the effects of different filter kernel sizes (&#61555;) on the resulting phase images using 4 different combination and filtering methods.These parameters result in optimized phase image quality at 7 T even for other resolution settings.

14:00         4575.     Improved Coil Combination for Homodyne-Corrected Phased Array Images

Ryan Fobel¹, Greg J. Stanisz¹

¹Sunnybrook Research Institute, Toronto, ON, Canada

Homodyne detection is a reconstruction method commonly used to reduce the number of high-frequency phase encodes for faster imaging or to correct for off-centered, partial echoes for shorter effective TEs. It also has useful properties for low-SNR imaging and negative polarity signals (e.g. Inversion recovery). When multichannel data sets are combined with the root-sum-of-squares technique, many of its benefits are negated because all of the phase information is lost. This study derives a simple modification to the root-sum-of-squares method that preserves phase information. A significant reduction to the noise floor and magnitude bias is demonstrated.

14:30         4576.     A Selective Acceleration and Multi-Resolution Approach to Multi-Echo Imaging

Suchandrima Banerjee¹, Cornelius Von Morze², Douglas AC Kelley¹, Philip J. Beatty¹, Daniel B. Vigneron², Eric T. Han¹

¹Applied Science Laboratory, GE Healthcare, Menlo Park, CA, USA; ²Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA, USA

Multi-echo sequences provide a time-efficient means to acquire – in a single scan - multiple images, each at a different clinically relevant contrast weighting. Parallel imaging has the potential to further increase the time efficiency of such sequences, but its application can be limited by the signal-to-noise ratio (SNR) available at the longest acquired echo time. To overcome this limitation, this work proposes a scheme that selectively applies parallel imaging to the higher SNR echoes, while reducing the spatial resolution and maintaining Nyquist sampling of the lower SNR echoes. We demonstrate the efficacy of this method for two dual echo applications.

15:00         4577.     Improved Visualization of the Subthalamic Nuclei by Reducing Susceptibility Induced Signal Losses in T2* Weighted Multi-Gradient-Echo Images

Steffen Volz¹, Elke Hattingen², Christine Preibisch^1,3, Thomas Gasser⁴, Ralf Deichmann¹

¹Brain Imaging Center, University Frankfurt, Frankfurt, Hessen, Germany; ²Department of Neuroradiology, University Hospital, Frankfurt, Hessen, Germany; ³Abteilung für Neuroradiologie, Klinkum rechts der Isar der TU München, München, Germany; ⁴Department of Neurosurgery, University Hospital, Frankfurt, Hessen, Germany

T2*-weighted gradient echo images yield good contrast of iron-rich brain structures. However, they are also sensitive to macroscopic field inhomogeneities, limiting anatomical information from affected areas. An image postprocessing method based on the evaluation of phase information is presented. A pixelwise image intensity correction of the original images acquired at different TE is performed in a first step and a combination of these images using optimized weighting factors is performed in a second step. The resulting images feature reduced signal losses in areas with major field gradients, while a contrast-to-noise (CNR) analysis in deep brain structures demonstrates good contrast.

13:30         4578.     Effect of Reordering on Dynamic Contrast Enhancement

Vibhas Deshpande¹, Gabriel Acevedo-Bolton², Helmuth Schultze-Haakh³, David Saloner², Gerhard Laub³

¹Siemens Medical Solutions USA Inc., San Francisco, CA, USA; ²VA, San Francisco, CA, USA; ³Siemens Medical Solutions USA Inc., CA, USA

Various reordering schemes with different benefits can be used for dynamic contrast-enhanced imaging. The effect of the reordering schemes on the contrast dynamics is however unclear. The purpose of this study was to evaluate contrast dynamics with linear, centric and radial centric-out k-space trajectory. Data were acquired in a flow phantom with a gradient-echo sequence and short and long injections. For short injections, radial centric-out trajectory shows artifacts because of varying T1’s during the scan. Artifacts are reduced with a longer injection. Contrast-enhancement is less with radial centric-out as compared to linear or centric if T1’s vary during the scan.

14:00         4579.     Improved Image Quality by Noquist Acceleration of Cardiac MRI

Senthil Ramamurthy¹, David Moratal², James W. Parks³, Stamatios Lerakis⁴, Marijn E. Brummer⁴

¹Children Research, Children's Healthcare of Atlanta, Atlanta, GA, USA; ²Universitat Politècnica de València, Valencia, Spain; ³Cardiology, Sibley Heart Center, Atlanta, GA, USA; ⁴Emory University, Atlanta, GA, USA

The Noquist method for accelerated cardiac imaging takes advantage of the spatiotemporal redundancy whenever the field of view contains static regions. Notable characteristics of Noquist compared to alternative methods include full preservation of spatiotemporal resolution. This study reports first results of a implementation of Noquist for a prospectively gated dynamic cardiac imaging method (GE FastCARD). A conventional image and a Noquist-accelerated image with the same acquisition parameters are compared. For the same acquisition time, Noquist-accelerated image shows clear improvement in image clarity. The improvement in image resolution clearly outweighs the SNR penalty associated with Noquist.

14:30         4580.     Comparison of Prospective ECG-Gated MRI with Wireless, Retrospective Gated MRI for Evaluation of Heart Function in Healthy and Infarcted Mouse Heart

Sandra Margaretha Bovens^1,2, Bernard C. te Boekhorst¹, Krista den Ouden¹, Kees W. van de Kolk¹, Arno Nauerth³, Marcel G. Nederhoff^1,2, Gerard Pasterkamp¹, Michiel ten Hove¹, Cees J. van Echteld⁴

¹Cardiology, University Medical Center Utrecht, Utrecht, Netherlands; ²InterUniversity Cardiology Institute of the Netherlands, Utrecht, Netherlands; ³Bruker Biospin MRI GmbH, Germany; ⁴Novartis Institutes for BioMedical Research, Basel, Switzerland

To assess (murine) left- and right ventricular heart function, ECG-triggered CMRI is often used. In mice with a myocardial infarction or cardiac hypertrophy it is sometimes challenging to get a good ECG signal. Therefore we compared a ‘wireless’ retrospective gating method with a prospective gating method in mice with and without infarcted hearts. The SNR and CNR of the retrospectively gated data are higher than that of the prospectively gated data. The ‘wireless’ retrospective gating method gives comparable results with regard to the cardiac function parameters, in the control mice as well as the mice with myocardial infarction.

15:00         4581.     A Navigator Gated Free-Breathing FSE for Black Blood Cardiac Imaging

Liuquan Cheng¹, Yi Wang², Pascal Spincemaille², Thanh D. Nguyen², Qian Zhao¹, Na Yang¹

¹Radiology, Chinese PLA General Hospital, Beijing, China; ²Radiology, Weill Medical College of Cornell University, NY, NY, USA

Breath-holding acquisition is one of the obstacle for cardiac MRI especially on the patient with heart insufficiency. A navigator echo gated, double/triple IR FSE was developed to acquire the black blood T2WI images in free-breathing mode.

14:00         4582.     Recoverability Bounds for Parallel Compressive Sensing MRI

Joshua D. Trzasko¹, Armando Manduca¹

¹Center for Advanced Imaging Research, Mayo Clinic, Rochester, MN, USA

Compressive Sensing (CS) and parallel imaging are two distinct techniques in MR imaging that allow for accelerated acquisition while retaining image quality. Recent advances have shown that these two methods may be naturally combined to provide even taster exams. In this work, we investigate the theoretical signal recovery properties of the hybrid model and derive a relationship between recoverability, the sampling model, and coil sensitivity profiles.

14:30         4583.     Investigation of Sparsifying Transforms for Compressed Sensing in MRI Reconstruction

Christopher Baker¹, Kevin King², Dong Liang¹, Leslie Ying¹

¹Eletrical Engineering, University of Wisconsin at Milwaukee, Milwaukee, WI, USA; ²Global Applied Science Lab, GE Healthcare, Waukesha, WI, USA

In compressed sensing (CS) MRI reconstruction, the level of sparsity and incoherence achieved by the transform affects the under-sampling that can be performed. This work investigates contourlets and the discrete cosine transform (DCT) as sparsifying transforms for CS reconstruction and compares them with the widely used wavelet transform. Results show that the contourlet transform performs about the same as the wavelet, while the DCT on small image patches outperforms the wavelet in CS reconstruction of MR images. The observation suggests that use of a DCT on small image patches may improve the CS reconstruction quality.

15:00         4584.     Sparse Image Reconstruction Using the Generalized Sampling Theorem for MR Angiography

Nicole Seiberlich¹, Hyun Jeong², Timothy J. Carroll^2,3, Mark A. Griswold¹

¹Radiology, Case Western Reserve University, Cleveland Heights, OH, USA; ²Biomedical Engineering, Northwestern University, Chicago, IL, USA; ³Radiology, Northwestern University, Chicago, IL, USA

GST-MRA, a novel method to reconstruct highly undersampled sparse images using ideas from the Generalized Sampling Theorem, is introduced here. In order to reduce the number of image pixels to be reconstructed, a soft mask is created using the composite image, and only those pixels which contain signal in this mask are reconstructed. This method is demonstrated for the reconstruction of high frame rate images of an AVM patient with a 2D acceleration factor of >20. In addition, parallel imaging in the form of coil sensitivity maps can also be incorporated into the method, further increasing the reconstruction fidelity.

15:30         4585.     Evaluation of Continuous Approximation Functions for the L₀-Norm for Compressed Sensing

Carlos A. Sing-Long^1,2, Cristian A. Tejos^1,2, Pablo Irarrazaval^1,2

¹Departamento de Ingenieria Electrica, Pontificia Universidad Catolica de Chile, Santiago, R.M., Chile; ²Biomedical Imaging Center, Pontificia Universidad Catolica de Chile, Santiago, R.M., Chile

Compressed Sensing allows reconstructing signals from some of its Fourier coefficients, if they are sparse in some representation. It is usually implemented as an l₁-minimization, although it was recently shown that the reconstruction process can be accelerated and the undersampling rate increased by using continuous approximations of the l₀-norm. We evaluated the performance of four different approximation functions in terms of reconstruction error, number of iterations to convergence and size of the reconstructed signal’s support. We observed that their convergence properties differed significantly, and we recommend a rational function with discontinuous derivative at the origin.

13:30         4586.     Comparative Evaluation of  L₁  Vs L_p Minimization Techniques for Compressed Sensing MRI

Juan Carlos Ramirez Giraldo¹, Joshua D. Trzasko¹, Armando Manduca¹

¹Center for Advanced Imaging Research, Mayo Clinic, Rochester, MN, USA

Compressed sensing(CS) can provide accurate reconstructions from highly undersampled data for certain types of MR scans, allowing faster acquisition times. Standard CS is based on l₁-norm minimization, which offers mathematical guarantees of global convergence. It is known that fewer samples are required for CS based on minimization of an l₀-norm, but this is mathematically more difficult and no convergence guarantees exist. We compare standard l₁-norm CS with two algorithms that approximate the l₀-norm with the l_p-norm (p between 0 and 1), with different sampling pattern densities and parameterizations. Despite the lack of theoretical guarantees, both l_p-norm algorithms always outperformed standard l₁-norm CS.

14:00         4587.     Compressed Sensing Reconstruction Based on Maximum Intensity Projection Images

Dong-Hyun Kim^1,2, Sung-Min Gho¹

¹Electrical and Electronic Engineering, Yonsei University, Shinchon-Dong, Seoul, Korea; ²Radiology, Yonsei University, Shinchon-Dong, Seoul, Korea

Compressed sensing algorithm is best fit for images that can have sparse representations in image domain. The 3D volumetric information is best viewed using a maximum intensity projection (MIP) image format. Here, work on combining MIP image with the compressed sensing algorithm is presented, and investigate the usefulness of this approach.

14:30         4588.     Compressed Sensing in Dynamic Enhanced Lung Imaging: A Comparison with K-T BLAST

Jia-Shuo Hsu¹, Su-Chin Chiu¹, Tzu-Cheng Chao², Yi-Ru Lin³, Hsiao-Wen Chung¹

¹National Taiwan University, Taipei, Taiwan; ²Brigham and Women's Hospital, Harvard Medical School,; ³Naitonal Taiwan University of Science and Technology

Dynamic contrast-enhanced lung imaging, with high resemblance with dynamic cardiac imaging for having high intensity variation constrained in relatively limited image space together with corresponding sparsity in the corresponding x-f space, may serve as a potential candidate for the CS techniques. This work shows while k-t BLAST was shown to be a well-performing alternative for dynamic CE lung imaging [4], CS may also be an encouraging candidate as it slightly outperforms k-t BLAST while remaining relatively stable in different phases

15:00         4589.     Incorporating Self-Referenced Information Into Compressed Sensing in Dynamic Imaging

Tzu-Cheng Chao^1,2, Bruno Madore², Ming-Long Wu³, Jing Yuan², Hsiao-Wen Chung¹

¹Department of Electrical Engineering, National Taiwan University, Taipei, Taiwan; ²Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA; ³Brain Imaging and Analysis Center, Duke University, Durham, NC, USA

Compressed Sensing is a newly developed fast imaging method aimed at robustly recovering the signal from undersampled datasets. In this work, we propose a modified sampling scheme to facilitate the reconstruction algorithm, based on an Orthogonal Matching Pursuit, for dynamic imaging purposes. Cardiac cine and fMRI data were used to test the approach and evaluate performance. The proposed modifications enabled significant reductions in computation time (about a 1.4 to 2 fold reduction).

13:30         4590.     Compressed Sensing for Active Feedback Contrast-Enhanced in Vivo Tumor Imaging

Luan Vu¹, Jon Furuyama¹, Thomas Goldstein², Stanley Osher², Yung-Ya Lin¹

¹Chemistry and Biochemistry, UCLA, Los Angeles, CA, USA; ²Applied Mathematics, UCLA, Los Angeles, CA, USA

The ability to successfully reconstruct images without having to collect the entire k-space is valuable for parameter optimization and can lead to a higher success rate for the detection of early tumors. Other k-space trajectories can be used to further improve the performance of the algorithm, such as concentrating the kept data points towards the center of k-space, while deleting more points around the edges (results not shown). As a result, strategic k-space sampling (beyond purely random) can be employed to further accelerate data acquisition.

14:00         4591.     Blade by Blade Compressed Sensing for PROPELLER

Eric Aboussouan¹, Jim Pipe¹

¹Barrow Neurological Institute, Phoenix, AZ, USA

This work proposes to use compressed sensing to fill missing lines of individual PROPELLER blades which are then gridded together. By making use of the inherent data redundancy of PROPELLER, such a scheme is thought to greatly reduce the risk of losing image features, which can be a concern with iterative reconstruction methods. This method could be used in applications where parallel imaging is not suitable (e.g. for PROPELLER spine imaging).

14:30         4592.     Self-Adjusted Regularization Ratio for Robust Compressed Sensing

Feng Huang¹, Yunmei Chen²

¹Advanced Concept Development, Invivo Corporation, Gainesville, FL, USA; ²Department of Mathematics, University of Florida, Gainesville, FL, USA

A self-adjustment technique is proposed in this work to automatically optimize the ratio between regularization term and data fidelity term in regularized reconstruction framework. Using compressed sensing (CS) as an example, experiments with both phantom and in vivo data sets demonstrated that the proposed method made the regularized reconstruction framework less sensitive to the choice of regularization parameter. This work dramatically reduces the difficulty of parameter decision and increases the practicability of regularized reconstruction techniques.

15:00         4593.     Compressed Sensing for Motion Artifact Reduction

Joelle Karine Barral¹, Dwight George Nishimura¹

¹Electrical Engineering, Stanford University, Stanford, CA, USA

Navigators can effectively track rigid-body motion of limited amplitude. However, data associated with significant motion need to be discarded, which often results in unacceptable artifacts. We propose to use a pseudo-random trajectory and compressed sensing theory to reconstruct datasets where data corrupted by motion and detected by navigators have been rejected. When data are acquired with a pseudo-random trajectory, motion occurring over several TRs results in a randomly undersampled dataset that can be accurately reconstructed. Simulation and experimental results are presented.

13:30         4594.     Reconstruction of Magnetic Resonance Inverse Imaging Using the Minimum L-1 Norm Constraint

Fa-Hsuan Lin^1,2, Thomas Witzel¹, Jonanthan Polimeni¹, John W. Belliveau¹

¹A. A. Martinos Center, Massachusetts General Hospital, Charlestown, MA, USA; ²Institute of Biomedical Engineering, National Taiwan University, Taipei, Taiwan

Magnetic resonance inverse imaging (InI) can obtain an unprecedented temporal resolution by using a minimal gradient encoding and solving inverse problems from multiple channels of the coil array measurements. To reconstruct InI data, additional constraint must be supplied to obtain a unique solution. Here we used the minimum L-1 norm constraint to reconstruction InI fMRI data. Compared to the minimum L-2 norm results, the minimum L-1 norm reconstructions are more spatially focal and have higher sensitivity in detecting the activity of human visual system.

14:00         4595.     An Improved Approach in Applying Compressed Sensing in Parallel MR Imaging

Bing Wu¹, Richard Watts², Rick Millane¹, Philip Bones¹

¹Depeartment of Electrical and Computer Engineering, University of Canterbury, Christchurch, New Zealand; ²Department of Physics and Astronomy, University of Canterbury, Christchurch, New Zealand

A new approach that improves the effectiveness of compressed sensing (CS) in parallel MR imaging (pMRI) is proposed. Instead of directly combining the CS recovery process with parallel imaging operators as has been previously reported, a two step reconstruction is used. A conventional SENSE-like reconstruction which gives a high fidelity reconstruction with high noise level is first made, then a CS reconstruction based on the prior knowledge of the SENSE reconstruction is performed to reduce the noise level, hence achieving a high fidelity reconstruction with low noise level. It is demonstrated using in-vivo data set that this new approach outperforms SENSE reconstruction or CS reconstruction alone.

14:30         4596.     A Novel Fast Dynamic Cardiac Data Reconstruction Method Using Prior Knowledge and Adaptive Matching Pursuits

Muhammad Usman¹, Philip G. Batchelor¹

¹Division of Imaging Sciences, Kings College London, London, UK

Recently, variants of matching pursuit methods such as Regularized Orthogonal Matching Pursuit (ROMP), Compressive Sampling Matching Pursuit (CoSAMP)and Sparsity Adaptive Matching Pursuit (SAMP) have been proposed and shown to give exact reconstruction for sufficiently sparse signals. Compared to the conventional OMP, these variants offer faster reconstructions. The SAMP has an additional advantage that it does not require any prior knowledge about the signal sparsity which is the case when recovering the practical compressible signals. Based on SAMP algorithm and the prior knowledge obtained from the sliding window reconstruction from the under-sampled data, we propose a scheme that can provide faster dynamic cardiac MR reconstructions compared to exisiting iterative schemes such as OMP.

15:00         4597.     HTGRAPPA: TGRAPPA Based B1-Weighted Image Domain Reconstruction for Real-Time MRI

Haris Saybasili^1,2, Peter Kellman¹, Mark A. Griswold³, J. Andrew Derbyshire¹, Michael A. Guttman¹

¹NIH/NHLBI, Bethesda, MD, USA; ²Biomedical Engineering Institute, Bogazici University, Istanbul, Turkey; ³Case Western Reserve University, USA

We present a new parallel imaging algorithm based on TGRAPPA for real-time MRI, called HTGRAPPA and its real-time, low-latency implementation suitable for interventional MR applications. Our method calculates GRAPPA coefficients in the k-space, and transfers them in the image domain. These image domain GRAPPA weights were combined into composite unmixing coefficients using adaptive B1-weight estimates and optimal noise weighting. That way, convolution operations in the k-space are avoided during the reconstruction, thus blazingly fast reconstruction speeds are achieved. More than 70 frames per second reconstruction performance is achieved on 30 coil, rate 4 dataset (up to 265x faster than TGRAPPA).

14:00         4598.     A Real-Time Motion Compensation Package and Active Marker Headband for Brain MRI

Melvyn Boon King Ooi¹, Sascha Krueger², William J. Thomas³, Srirama V. Swaminathan⁴, Truman R. Brown^1,3

¹Biomedical Engineering, Columbia University, New York, USA; ²Philips Research Europe, Hamburg, Germany; ³Radiology, Columbia University, New York, USA; ⁴Philips Medical Systems, Cleveland, OH, USA

A prospective, intra-image compensation strategy for rigid-body motion is presented. A short tracking-sequence, interleaved into the imaging-sequence, measures the positions of three active-markers integrated into a headband worn by the subject. During head motion, the rigid-body transformation that realigns these markers is fed back to update the image-plane – maintaining it at a fixed orientation relative to the head – before the next segment of k-space is acquired. The complete package requires minimal additional hardware and can be flexibly incorporated into multiple imaging-sequences, promoting transferability to clinical practice. Improvements in image quality are evaluated in 3D-MPRAGE brain MRI during volunteer motion.

14:30         4599.     In-Vivo Applications of Optical Real-Time Motion Correction Using a Monovision System

Murat Aksoy¹, Matus Straka¹, Stefan Skare¹, Rexford Newbould², Samantha Holdsworth¹, Juan Santos³, Roland Bammer¹

¹Department of Radiology, Stanford University, Stanford, CA, USA; ²GlaxoSmithKline, London, UK; ³Department of Electrical Engineering, Stanford University, Stanford, CA, USA

Real time motion correction using an external optical tracking system has been suggested as a way to perform rigid head motion correction for MRI. In this study, we demonstrated the motion correction efficacy of our real-time monovision system that uses a single camera mounted inside the magnet bore on the head coil. Experiments were carried out using a gradient echo and a spin echo sequence with the subject performing varying degrees of in-plane and through plane motion. The results show that the current system is effective in removing both in plane and through plane motion.

15:00         4600.     Image Instability Evaluation and Motion Correction for High-Resolution MRI of the Rat Retina

Xiaodong Zhang¹, yingxia Li¹, Timothy Q. Duong²

¹Yerkes National Primate Research Center, Emory University, Atlanta, GA, USA; ²Research Imaging Center, University of Texas Health Science Center at San Antonio, San Antonio, TX, USA
This study systemically examined these potential animal and hardware issues, and implemented solutions to maximize stability for high-resolution structural, physiological and functional MRI of the rat retina. Studies were performed on phantoms and the in vivo retinas to distinguish between potential animal and hardware issues. Solutions to maximize retinal MRI stability and protocols to evaluate and verify temporal stability are detailed.

15:30         4601.     Rotation Correction with Self-Navigated MRI

Jason Mendes¹, Dennis L. Parker¹

¹UCAIR, University of Utah, Salt Lake City, UT, USA

This work provides a demonstration that in-plane rotational motion that occurs during the acquisition of MRI data can be quantified using a self-navigation technique. The proposed technique compares adjacent sets of measurement lines in k-space to detect and quantify object rotation. This method can be applied to any segmented sequence that samples k-space in sets of equally spaced lines. It does not require any patient preparation or the acquisition of separate navigator data. The method only requires a slight increase in the field of view along the phase encoding direction, but the increase required is small enough that resolution or total scan time are not significantly affected.

13:30         4602.     A Kalman Filtering Framework for Prospective Motion Correction

Julian Maclaren¹, Oliver Speck², Jürgen Hennig¹, Maxim Zaitsev¹

¹Dept. of Diagnostic Radiology, Medical Physics, University Hospital Freiburg, Freiburg, Germany; ²Biomedical Magnetic Resonance, Otto-von-Guericke University, Magdeburg, Germany

Tracking head position and prospectively adjusting the imaging volume is becoming an increasingly popular means of preventing motion artefacts in MRI. However, the technique imposes stringent requirements on the accuracy of the tracking system used. For high-resolution imaging, these requirements are difficult to achieve in practice. We present a Kalman-filtering based approach that improves pose estimation and prediction and enables estimation of residual errors. This makes retrospective correction of residual motion artefacts possible. The result is a reduction in the required accuracy of the tracking system itself.

14:00         4603.     Reducing Artifacts in Dynamic MRI Movies Using a Spline Interpolated Sliding Window Technique

Eric Frederick^1,2, Iga Muradyan¹, Mirko Hrovat³, Hiroto Hatabu¹, Samuel Patz¹

¹Radiology, Brigham and Women's Hospital, Boston, MA, USA; ²Physics, University of Massachusetts at Lowell, Lowell, MA, USA; ³Mirtech, Brockton, MA, USA

A spline interpolated sliding window technique is proposed to reduce intensity pulsation artifacts in rectilinearly sampled dynamic MRI movies. The technique is tested on both simulated MRI data and experimentally with a dynamic phantom.

14:30         4604.     Pitfalls of Complex Summation and Its Variant Method in Synthesizing the Phase-Cycled SSFP Images to Suppress the Band Artifact

Kwan-Jin Jung^1,2

¹BIRC, University of Pittsburgh, Pittsburgh, PA, USA; ²Bioengineering Dept, University of Pittsburgh, Pittsburgh, PA, USA

The SSFP image suffers from the band artifact. The complex summation (CS) method has been reported to be better than the maximum intensity projection (MIP) in suppressing the band artifact. However, it was noted recently that CS has an inherent deficiency in suppressing the band artifact due to phase incoherence across the phase-cycled images. It was described that a magnitude-weighted CS method (MWCS) was more reliable than CS and SoS (square-of-sum) in suppressing the band artifact. In this abstract, the pitfalls of CS and its variant MWCS are studied further with an experimental demonstration and simulation. It is found that the band suppression of synthesis methods depends on the flip angle and relaxation times. The CS and MWCS were demonstrated to perform worse than MIP in the phantom experiment. Besides, CS had an abnormal image intensity due to the chemical shift.

15:00         4605.     Azimuthal Sorting in Tandem with Elliptical Reordering (ASTER): A New K-Space Reordering Scheme for Reduced Motion Sensitivity

Manojkumar Saranathan¹, Ersin Bayram², Ananth Madhuranthakam³

¹Applied Science Laboratory, GE Healthcare, Rochester, MN, USA; ²MR Engineering, GE Healthcare, Waukesha, WI, USA; ³Applied Science Laboratory, GE Healthcare, Boston, MA, USA

MRI is very sensitive to motion, particularly in 3D acquisitions where temporally localized motion can propagate across the whole reconstructed volume. In cardiac and abdominal imaging applications, subjects are required to suspend their respiration for a period of 15-25s to minimize ghosting and blurring artifacts. Breath-holding poses difficulties in pediatric and elderly subjects. We propose ASTER (Azimuthal Sorting in Tandem with Elliptical Reordering), a novel k-space reordering scheme that minimizes sensitivity to motion whilst permitting flexible combination with magnetization preparation schemes. We demonstrate the validity of our scheme using PSF simulations and volunteer data from abdominal imaging.

13:30         4606.     In-Vivo Flow-Artifact Suppression Using Parallel Spatially Selective Excitation

Johannes Thomas Schneider^1,2, Martin Haas², Jürgen Hennig², Sven Junge¹, Wolfgang Ruhm¹, Peter Ullmann¹

¹Bruker BioSpin MRI GmbH, Ettlingen, Germany; ²Dept. of Diagnostic Radiology, Medical Physics, University Hospital Freiburg, Freiburg, Germany

Parallel spatially-selective excitation / TransmitSENSE allows specific excitation of defined regions. In the present study this technique was successfully applied to the removal of motion artifacts in rat imaging: Signal from vessels in the rat head which showed up as ghost artifacts superimposed on the rat’s brain could be efficiently suppressed by selective excitation of the rat brain only. These in-vivo results underline the potential of TransmitSENSE for the prevention of artifacts originating from flow or motion.

14:00         4607.     Prospective Motion Correction with Stereoscopic Optical Tracking at 7T

Daniel Stucht^1,2, Peter Schulze¹, Maxim Zaitsev³, K. A. Danishad¹, Ilja Y. Kadashevich¹, Oliver Speck¹

¹Biomedical Magnetic Resonance, Otto-von-Guericke University, Magdeburg, Germany; ²IBMI, Otto-von-Guericke University, Magdeburg, Germany; ³Dept. of Radiology, Medical Physics, Universital Hospital Freiburg, Freiburg, Germany

In this study an implementation of prospective motion correction on a 7T MR scanner is presented. The potentially higher resolution in 7T requires longer acquisition times without patient motion. The reduction of motion artefacts is shown for motion between different scans and motion during the aquisition of image data in different resolutions up to 1024x1024. The results of in vivo and phantom measurements are shown. The achievable artifact-free image resolution is limited by the accuracy of the tracking data.

14:30         4608.     The Effect of Magnetization Transfer on Rapid T2 Estimation with Phase-Cycled Variable Nutation SSFP

Hendrikus Joseph Alphons Crooijmans¹, Klaus Scheffler¹, Oliver Bieri¹

¹Radiological Physics, University of Basel Hospital, Basel, BS, Switzerland

The transversal relaxation time (T2) found by means of two phase-cycled variable nutation steady-state free precession measurements analysis (DESPOT2) is subject to magnetization transfer (MT) effects when imaging tissues. To explore this possible effect, human brain is measured with a balanced SSFP sequence using several RF pulse durations and compared the results with a multi-contrast spin echo reference T2 measurement. It is found that MT can have a severe influence on the found T2 value.

15:00         4609.     Correction of Off-Resonance Effects in Multi-Component Driven Equilibrium Single Pulse Observation of T1 and T2 (McDESPOT)

Sean CL Deoni¹

¹Centre for Neuroimaging Sciences, London, UK

mcDESPOT is a promising new approach to whole-brain multi-component relaxometry. However, as the method utilizes data acquired using the rapid steady-state free-precession imaging sequence, it is sensitive to off-resonance effects leading to substantive errors in the derived myelin water fraction and relaxation time estimates. Here we suggest a simple approach allowing the near complete removal of these artifacts and demonstrate the efficiency of the technique in vivo.

13:30         4610.     A New Synthesis Method of Phase-Cycled SSFP Images to Remove the Band Artifact by Combining Complex Summation and Maximum Intensity Projection Techniques

Kwan-Jin Jung^1,2

¹BIRC, University of Pittsburgh, Pittsburgh, PA, USA; ²Bioengineering Dept, University of Pittsburgh, Pittsburgh, PA, USA

The complex summation technique (CS) is known to be effective in reducing the band artifact as a synthesis method of multiple phase-cycled images. However, it was noted that CS is not reliable due to phase incoherence in the multiple phase-cycled images. On the other hand, the maximum intensity projection (MIP) and spectral decomposition synthesis (SDS) techniques do not have the phase-related problem even though they might be less effective in reducing the band artifact. Here, the advantages of CS and MIP are combined as a more effective synthesis method to take advantage of both techniques in reducing the band artifact without the detrimental effect of the phase incoherence. This new technique was confirmed by a phantom experiment at 3T to perform much better than other existing methods in reducing both the band artifact and image noise.

14:00         4611.     ¹⁹F/¹H Simultaneous 3D Radial Imaging of Atherosclerotic Rabbits Using Self-Navigated Respiratory Motion Compensation

Jürgen Rahmer¹, Jochen Keupp¹, Shelton D. Caruthers^2,3, Oliver Lips¹, Todd A. Williams³, Samuel A. Wickline³, Gregory M. Lanza³

¹Philips Research Europe, Hamburg, Germany; ²Philips Healthcare, Andover, MA, USA; ³Washington University, St. Louis, MO, USA

MR molecular imaging of low doses of tracer material often requires long scan times and makes motion compensation strategies desirable. 3D radial imaging with golden section profile interleaving allows auto-navigated motion compensation. Here, it is applied to simultaneous 19F/1H imaging of the aorta of atherosclerotic rabbits after systemic administration of 19F-based angiogenesis-targeted nanoparticles. Tracking of respiratory motion is

14:30         4612.     High Frame Rate Cardio-Respiratory Imaging Using Model-Based Reconstruction

Pierre-André Vuissoz^1,2, Freddy Odille^1,3, Brice Fernandez^1,4, Maelene Lohezic^1,4, Adnane Benhadid^1,2, Damien Mandry^1,5, Jacques Felblinger^2,6

¹IADI, Nancy-Université, Nancy, France; ²U947, INSERM, Nancy, France; ³Centre for Medical Image Computing, University College London, London, UK; ⁴Global Applied Science Lab., GE healthcare, Nancy, France; ⁵CHU de Nancy, Nancy, France; ⁶CIC-IT 801, INSERM, Nancy, France

The recently described Generalised Reconstruction by Inversion of Coupled Systems (GRICS) algorithm enables to build a model of image deformation correlated to physiological sensors such as ECG or respiratory belt. Since these physiological signals are recorded at high sampling rate, this enables to reconstitute a real time high resolution (256x256) movie of thoracic motion with quality comparable to breath-hold acquisitions. 18 reconstitutions of thoracic motion of 16s long at 25fps. have been computed. Both breathing and cardiac beating of the healthy subject can be observed in different orientations, in particular cardiac short axis and both vertical and horizontal long axes.

15:00         4613.     On the Feasibility of Accelerating Self-Gated Cine Cardiac Imaging in Rodents Using SENSE

Christof Baltes*¹, David Ratering*¹, Markus Rudin^1,2

¹Institute for Biomedical Engineering, University and ETH Zurich, Zurich, Switzerland; ²Institute of Pharmacology and Toxicology, University Zurich, Zurich, Switzerland

Self-gated MR imaging has been demonstrated to be a useful tool for cardiac MRI in small rodents. However, the method suffers from long acquisition times. For this reason, the feasibility of accelerating self-gated cardiac imaging in rodents using SENSE was investigated in this work. Fully sampled cine cardiac images of the rat were acquired using the IntraGate technique. Computer simulations exploring the effect of residual respiratory motion artifacts on the SENSE reconstruction revealed that higher demands on the respiratory compensation are required for the accelerated compared to the fully sampled acquisition.

14:00         4614.     Removal of Residual Motion Artifacts in FMRI Using Constrained Independent Component Analysis

Kiran Kumar Pandey¹, Douglas C. Noll¹

¹Biomedical Engineering, University of Michigan, Ann Arbor, MI, USA

Image-registration in fMRI only corrects for bulk movements, leaving secondary artifacts - spin history, motion induced field inhomogeneity changes, interpolation errors - untouched. Secondary artifacts increase variance and reduce sensitivity of fMRI. Methods using motion parameters as “Nuisance Explanatory Variables” or ICA are used to isolate/remove residual motion artifacts. Use of ICA requires isolation and then identification of motion components before removal. We used constrained ICA to isolate/identify these components in one step using motion parameters as references. Use of cICA efficiently and effectively removed residual motion artifacts without “manual” identification step and, increased the sensitivity of BOLD fMRI studies.

14:30         4615.     Wii – Highly Sensitive Tracking of Patient Motion

Olga Tymofiyeva¹, Michael Ledwig¹, Kurt Rottner², Ernst-Juergen Richter², Peter Michael Jakob¹

¹Dept. of Experimental Physics 5, University of Würzburg, Würzburg, Bavaria, Germany; ²Dept. of Prosthodontics, Dental School, University of Würzburg, Würzburg, Bavaria, Germany

Numerous medical MRI applications suffer from patient motion and require motion control. A simple and low-cost motion tracking system is proposed which is based on the infrared light detection camera built-in in a Wii Remote (Nintendo), connected via Bluetooth to a computer for analysis. The proposed system allows for detecting patient motion with sensitivity below 40µm and can become a solution for a wide range of MRI applications, in which real-time patient motion control is required.

15:00         4616.     Prospective Head Movement Correction for High-Resolution MRI Using a Single Camera

Lei Qin^1,2, Fenghua Jin², Yang Tao², Jeff H. Duyn¹

¹NINDS, National Institutes of Health, Bethesda, MD, USA; ²Univ of Maryland, College Park, MD, USA

We propose a novel prospective motion correction method for MRI based on positional tracking with a single video camera. A short training scan, using whole-brain EPI during intentional head motion, serves to relate camera images of the human face to head position. With this information, motion during a real-time MRI scan is estimated by correlating each newly captured camera image with the one from the training data. The corresponding motion parameters are fed back to the MRI scan computer to adjust scan parameters. Results show the system is able to correct motion for high-resolution anatomical MRI.

15:30         4617.     On the Application of Phase Correction and Use of K-Space Entropy in Partial Fourier Diffusion-Weighted EPI

Samantha J. Holdsworth¹, Stefan Skare¹, Roland Bammer¹

¹Lucas MRS/I Center, Stanford University, Stanford, CA, USA

It is well-known that diffusion-weighted (DW) imaging is very sensitive to the effects of brain motion, even in single-shot (ss)-EPI. Pulsatile brain motion that occurs during the application of the DW gradients can result in the non-linear dispersion of k-space, corresponding to signal dropout and shading in the image domain. Here we explore the use of k-space entropy as a metric to identify k-space corrupted by non-linear brain motion; the use of peripheral cardiac gating and non-gating; phase correction applied before both homodyne and POCS reconstruction; as well as the number of overscans that should be used to avoid significant artifacts due to pulsatile brain motion.

13:30         4618.     Comparison of Physiological Trigger Modes for DWI in the Abdomen

Tobias Binser¹, Harriet C. Thoeny², Chris Boesch¹, Peter Vermathen¹

¹Dept. of Clinical Research, University of Bern, Bern, Switzerland; ²Dept. of Radiology, Inselspital Bern, Bern, Switzerland

A major problem in abdominal diffusion-weighted imaging (DWI) is related to physiological motion artifacts caused by respiration and cardiac pulsation. In the present study we compared different physiological triggering methods, namely respiratory versus combined respiratory-cardiac gating techniques in DWI of the kidney. The main objective was to evaluate if respiratory-cardiac double triggering is advantageous for the determination of diffusion parameters, especially for the stability of microperfusion components in addition to apparent diffusion coefficients (ADC). The results from repeated measurements show a significant improvement of data stability for respiratory-cardiac double triggering over respiratory triggering alone.

14:00         4619.     Comparison of Quantitative Artifact Level in TGRAPPA and TSENSE Reconstruction During Deep Breathing

Mihaela Jekic¹, Yu Ding², Yiu-Cho Chung³, Subha V. Raman⁴, Jennifer Dickerson⁴, Sven Zuehlsdorff³, Sonia Nielles-Vallespin³, Orlando P. Simonetti

¹Biomedical Engineering, The Ohio State University, Columbus, OH, USA; ²Davis Heart and Lung Research Institute, The Ohio State University; ³Siemens Healthcare; ⁴The Ohio State University Medical Center

We quantitatively compared the artifact performance of TGRAPPA and TSENSE real-time cine imaging under the condition of deep breathing, which can cause the chest wall to move in and out of the FOV and also can create a mismatch between the coil sensitivity map and the actual coil position. We applied an autocorrelation-based approach to quantify the artifacts in 30 myocardial slices, exploiting their spatially fixed nature at 1/3 and 2/3 FOV for rate 3 acceleration. We found that TGRAPPA performed better than TSENSE in terms of quantitative artifact level, which was in agreement with qualitative physician-assigned artifact scores.

14:30         4620.     Respiratory Navigation Scheme for Free-Breathing 3D SPGR Liver Imaging: Technical Feasibility

Anja C.S. Brau¹, Yuji Iwadate², Ersin Bayram³, Phillip M. Young⁴, Shreyas Vasanawala⁴, Robert J. Herfkens⁴

¹Applied Science Lab, GE Heathcare, Menlo Park, CA, USA; ²Applied Science Lab, GE Healthcare, Hino, Japan; ³GE Healthcare, Waukesha, WI, USA; ⁴Department of Radiology, Stanford University Medical Center, Palo Alto, CA, USA

Breath-held 3D dynamic contrast-enhanced gradient echo imaging is an important component of abdominal MR exams. However, for patients who cannot breath-hold, motion artifacts can compromise diagnostic utility. Thus a respiratory-navigated acquisition is desirable; however, a navigated acquisition should remain sufficiently fast to capture contrast dynamics and minimize the impact on image contrast. In this work, we present the preliminary technical feasibility of a respiratory navigation scheme applied to a conventional fat-suppressed 3D SPGR sequence as well as to a 3D chemical species-based water-fat separation sequence.

15:00         4621.     Simple Method for Free-Breathing Multi-Slice T2w-TSE Liver Imaging Without PACE

Sang-Young Zho¹, Jaeseok Park², Dong-Hyun Kim^1,2

¹Electrical and Electronic Engineering, Yonsei University, Shinchon-Dong, Seoul, Korea; ²Radiology, Yonsei University, Shinchon-Dong, Seoul, Korea

PACE technique enables liver imaging with T2w-TSE sequence for patients who have difficulty holding their breath. One of limitation of PACE is relatively long scan time. If we do imaging liver without it, scan time will be reduced. For this, one projection line is added to conventional 2DFT TSE sequence and shows reduced respiratory motion artifact.

13:30         4622.     3-D Prospective Motion Correction for MR Spectroscopy

Brian Keating¹, J C. Roddey², Weiran Deng¹, Anders Dale², Nathan White³, V Andrew Stenger¹, Thomas Ernst¹

¹Dept. of Medicine, University of Hawaii at Manoa, Honolulu, HI, USA; ²Dept. of Neurosciences, University of California at San Diego, La Jolla, CA, USA; ³Cognitive Science, University of California at San Diego, La Jolla, CA, USA

Patient motion during MR spectroscopy (MRS) acquisitions compromises the spectral quality. Therefore, we adapted a 3D image-based prospective motion correction module for use with a 1H PRESS sequence. A spiral navigator is acquired immediately prior to the MRS water suppression module, to obtain three orthogonal images from which head motion is estimated. By applying the appropriate rotations and translations, the voxel can be made to remain stationary with respect to the brain. Spectra acquired during head motion demonstrate improved spectral quality, including a reduction in lipid contamination, lower line width, and spectral reproducibility, when motion correction is applied.

14:00         4623.     Real-Time Prospective Rigid-Body Motion Correction with the EndoScout Gradient-Based Tracking System

Andre Jan Willem van der Kouwe^1,2, Barry Fetics³, Dmitry Polenur³, Abraham Roth³, Erez Nevo³

¹Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, MA, USA; ²Department of Radiology, Harvard Medical School, Brookline, MA, USA; ³Robin Medical Inc., Baltimore, MD, USA

Motion during MR acquisitions causes considerable wasted scan time and confounds clinical diagnosis. We present an integrated solution for tracking and correcting rigid body head motion in real-time using the EndoScout gradient-based tracking system and the Siemens 3T TIM Trio MR scanner. The EndoScout probe consists of six small coils, and its location and orientation in the switching fields can be determined by the voltages induced in the coils. By rigidly affixing the probe to an object, the location of the object can be tracked in real-time. We demonstrate that the system provides real-time prospective motion correction during MR imaging.

14:30         4624.     Real-Time Optical Motion Correction for Diffusion Tensor Imaging

Murat Aksoy¹, Matus Straka¹, Stefan Skare¹, Rexford Newbould², Samantha Holdsworth¹, Juan Santos³, Roland Bammer¹

¹Department of Radiology, Stanford University, Stanford, CA, USA; ²GlaxoSmithKline, London, UK; ³Department of Electrical Engineering, Stanford University, Stanford, CA, USA

Due to its prolonged acquisition time and inherent motion sensitivity, motion correction for DTI is a must for clinically acceptable image quality. In this study, we used a real-time optical motion correction system that relies on monovison to correct for rigid head motion artifacts in DTI. The proposed system employs a single camera mounted on the head coil and a planar marker with a checkerboard pattern on it. Relative changes in marker pose are used to update the gradient and RF waveforms in real-time. Results show that the proposed system is effective in correcting for pixel misregistration and signal dropouts caused by patient motion in DTI.

15:00         4625.     Variable Sampling Density Spherical Navigator Echoes (VSD SNAV) for Prospective 3D Alignment

Junmin Liu^1,2, Maria Drangova^1,2

¹Imaging Research Laboratories, Robarts Research Institute, London, Ontario, Canada; ²Medical Biophysics, Schulich School of Medicine and Dentistry, University of Western Ontario, London, Ontario, Canada

An improved spherical navigator-based technique for prospective inter-scan realignment is presented. Instead of acquiring a set of orbital trajectories to cover the entire surface of a k-space sphere to determine rotation, combined with a linear navigator to determine translation, a two-shot variable density helical spiral spherical navigator (VSD SNAV) was adapted. Using the VSD-SNAV resulted in reduction of the navigator acquisition time from 19 s to 50 ms, while maintaining accuracy for translations up to 20 mm and rotations as large as 20 degrees. In vitro and in vivo results are presented.

13:30         4626.     Self-Navigator for Free-Breathing 3D SSFP Cardiac Imaging: Center of Mass Along the Slice Direction Versus the Image Signal Sum Using the Center of K-Space.

Pascal Spincemaille¹, Jing Liu¹, Thanh Dang Nguyen¹, Martin Prince¹, Yi Wang¹

¹Radiology, Weill Medical College of Cornell University, New York, NY, USA

The k-space center signal has been used for self-navigating of respiratory motion that is very useful for SSFP imaging. However, this signal is slice location and geometry dependent and is affected by blood flow and heart motion, which may lead to ineffective motion sensitization. In this study, the kz axis is repeatedly sampled and Fourier transformed to generate a projection of the image volume onto the slice axis. The center-of-mass of this projection is proposed as a respiratory gating signal and is shown to provide superior respiratory motion artifact suppression when compared to the use of the center-of-k-space signal only.

14:00         4627.     Practicality Makes a Comeback: Dynamic MRI Without the Overhead

Uygar Sümbül¹, John Mark Pauly¹

¹Electrical Engineering, Stanford University, Stanford, CA, USA

A causal, continuously auto-calibrating acceleration algorithm is presented for dynamic MRI. The algorithm does not need any extra data acquisition to calibrate itself and works with arbitrary readout trajectories. The reconstruction is fast and it adapts almost instantaneously to changing imaging parameters. Therefore the algorithm is suitable for applications such as catheter tracking and image guided therapy. In-vivo cardiac MRI results demonstrate that the algorithm decreases the temporal blur when compared to the sliding window reconstruction.

14:30         4628.     Computational Model of the Left Ventricle Motion Using Tagged Magnetic Resonance Imaging (TMRI)

Mohammed Dahawy Alenezy¹, Tareq M. Alrefae², Jack Shi¹, Mehmet Bigen³

¹Physics and Astronomy, University of Kansas, Lawrence, KS, USA; ²Kuwait University, Khaldiya, Kuwait, Kuwait; ³Radiology, Medical School of South Carolina, USA

A computational model is proposed to study the motion of left ventricle (LV) using tMRI data. The LV wall motion is decomposed into the deformation component ,along the short axis and the rotation component, along the long axis of LV. rotational motion is represented by 2-D non-linear rotational transformation and the deformation is described by a 1/r function. The tMRI data of Eight different subjects were studied . It was found that there is a significant difference in the deformation parameter between the diabetic and the healthy rats and human. There is no significant difference in the average twist between human and rat heart.The torsion of rat heart is, however, found to be substantially larger than that of human.

15:00         4629.     Free Breathing Black-Blood Systolic Imaging Using Heart Rate Prediction and Motion Compensated Reconstruction

Maelene Lohezic^1,2, Brice Fernandez^1,2, Julien Oster^2,3, Damien Mandry^2,4, Olivier Pietquin^3,5, Pierre-Andre Vuissoz^2,3, Jacques Felblinger^2,3

¹Global Applied Science Lab., GE Healthcare, Nancy, France; ²IADI, Nancy-Université, Nancy, France; ³U947, INSERM, Nancy, France; ⁴CHU de Nancy, Nancy, France; ⁵IMS Research Group, SUPELEC-Metz Campus, Metz, France

High resolution black-blood systolic imaging is difficult to obtain due to long acquisition time incompatible with breath hold and to preparation time needed. A new method allowing free breathing black-blood systolic imaging is presented and validated on five healthy volunteers. It combines RR interval prediction, respiratory motion estimation and motion compensated reconstruction. The proposed method showed improvements when compared to the average of 3 free breathing acquisitions and to breath-hold images. Results have been obtained on 256x256 acquisitions. They can be extended to higher resolutions, allowing accurate examination of heart structures, even right ventricle wall or papillary muscles.

14:00         4630.     3-D Multi-Coil Motion Correction for Spiral Projection Imaging

Kenneth O. Johnson¹, Eric Aboussouan¹, James G. Pipe¹

¹Keller Center for Imaging Innovation, Barrow Neurological Institute, Phoenix, AZ, USA

Motion correction has been successfully demonstrated for single coil data using Spiral Projection Imaging. Extending motion correction to multi-coil data poses additional challenges. This work investigates an approximation method applied to multi-coil data, that proves successful on synthesized data.

14:30         4631.     Quantitative Evaluation and Optimization of 3D PROspective MOtion (PROMO) Through Offline Simulation

Jian Zhang^1,2, Daniel Rettmann³, Eric Han⁴, Cooper Roddey⁵, Nate White⁵, Joshua Kuperman⁵, Juan Santos¹, Anders Dale^5,6, Ajit Shankaranarayanan⁴

¹Department of Electrical Engineering, Stanford University, Stanford, CA, USA; ²Department of Radiology, Stanford University, Stanford, CA, USA; ³Global Applied Science Lab, GE Healthcare, Rochester, MN, USA; ⁴Global Applied Science Lab, GE Healthcare, Menlo Park, CA, USA; ⁵Department of Neuroscience, University of California, San Diego, La Jolla, CA, USA; ⁶Department of Radiology, University of California, San Diego, La Jolla, CA, USA

In this work, quantitative off-line evaluations have been performed to study the accuracy and stability of PROMO. With the aid of our off-line PROMO simulation package, various impacting factors, including navigator prescription parameters, navigator reconstruction filters, image noises, and motion patterns, have been studied. An optimal PROMO parameter configuration has been proposed. And both simulation and in vivo results show that this optimization improves the algorithm performance significantly. Furthermore, since our offline system contains independent coordinate translation and gradient adjustment simulation module, it can be easily applied to evaluate other motion correction algorithms as well.

15:00         4632.     Geometric Distortion Correction in Echo Volumar Imaging

Wenlong Qian¹, Paul M. Glover¹, Richard W. Bowtell¹

¹SPMMRC, University of Nottingham, Nottingham, UK

Echo volumar imaging (EVI) is highly sensitive distortion due to magnetic field inhomogeneities because of its long echo train length. This sensitivity can be reduced by using parallel imaging to shorten the echo train length, but significant distortion can still remain in images acquired in vivo. Here we have therefore extended two distortion correction methods that have previously been used with EPI to EVI. The first is based on using a field map and the second involves point spread function (PSF) mapping. Images obtained from phantom and human brain show that correction using a PSF-based deconvolution is robust and effective.

15:30         4633.     Evaluation of EPI Geometric Distortion Correction Using Phase Labeling for Additional Coordinate Encoding (PLACE)

Sheeba Arnold¹, Markus Vogler^2,3, Oliver Hinds¹, Susan Whitfield-Gabrieli⁴, Michael Hamm⁵, Josef Pfeuffer⁵, Christina Triantafyllou^1,6

¹Athinoula A. Martinos Imaging Center, McGovern Institute for Brain Research, MIT, Cambridge, MA, USA; ²Siemens HealthCare, Erlangen, Germany; ³University of Applied Sciences Hof, Germany; ⁴Department of Brain and Cognitive Sciences, MIT, Cambridge, MA, USA; ⁵Siemens Medical Solutions USA Inc., Charlestown, MA, USA; ⁶Athinoula A. Martinos Center for Biomedical Imaging, Radiology, MGH, Harvard Medical School, Charlestown, MA, USA

Among the methods proposed for EPI geometric distortion correction Phase Labeling for Additional Coordinate Encoding (PLACE) has the advantage that it does not require post-processing steps because the correction happens real-time during acquisition. In this study we evaluate the PLACE method quantitatively, first with a finger-tapping task by inducing artificial distortions, and then compared it

13:30         4634.     Compensation of Respiration-Induced Off-Resonance in Real-Time Spiral k-t BLAST Imaging

Holger Eggers¹

¹Philips Research Europe, Hamburg, Germany

Free breathing leads not only to direct motion artifacts in abdominal real-time imaging, but also to respiration-induced field inhomogeneity artifacts. Their compensation would normally require a dynamic field mapping and a complex off-resonance correction. In this work, a simple, approximate approach is proposed for this purpose instead and is demonstrated in real-time spiral k-t BLAST imaging. It essentially involves a derivation of resonance frequency offsets from phase variations in the low frequency k-space samples of the imaging data and a subsequent, corresponding demodulation of the imaging data before their reconstruction for each receive coil element.

14:00         4635.     Free-Breathing Dynamic Magnetic Resonance Imaging of the Abdomen

Frederik J. de Bruijn¹, Yan Zhang^1,2, André M. Sprengers³, Aart J. Nederveen³, Jaap Stoker³, Rolf M. Lamerichs¹

¹Philips Research, Eindhoven, Netherlands; ²Leiden Institute of Advanced Computer Science, Leiden University, Leiden, Netherlands; ³Department of Radiology, Academic Medical Center, Amsterdam, Netherlands

Abdominal magnetic resonance imaging (MRI) is complicated by tissue motion as well as by weak tissue contrast. The major source of motion is respiration, which can be dealt with by breath-hold imaging or by respiratory triggering. Nevertheless, these commonly used methods are known to have their limitations in terms of patient burden as well as in image quality. We propose a method that allows 3-dimensional (3D) motion-compensated imaging of the bowel in the presence of continued respiratory motion and normal bowel peristalsis. The method is based on a computationally efficient spatiotemporal recursive search cube matching algorithm which allows accurate real-time volumetric motion estimation and compensation for the respiratory motion. We are currently evaluating the application to motion-compensated reconstruction of dynamic contrast-enhanced data of the colon.

14:30         4636.     Improved R2* Measurement  Accuracy with Absolute SNR Truncation and Optimal Coil Combination

Xiaoming Yin^1,2, Saurabh Shah³, Peter Kellman⁴, Andrew C. Larson^1,5

¹Department of Radiology, Northwestern University, Chicago, IL, USA; ²Department of Electrical Engineering and Computer Science, Northwestern University, Evanston , IL, USA; ³Siemens Medical Solutions,, Chicago, IL, USA; ⁴Laboratory of Cardiac Energetics, NIH, Bethesda, MD, USA; ⁵Department of Electrical Engineering and Computer Science, Northwestern University, Evanston, IL, USA

R2* is typically estimated via mono-exponential fitting of signal decay within a series of GRE images combined by root sum-of-square (RSS) approach. However, RSS approaches rectify and bias noise in later TEs, resulting in systematic fitting errors. Our work investigated the accuracy of low SNR R2* measurements. By combining a truncation model with array coil combination methods, we found that SNR-based truncation can be an effective method to reduce measurement error. Given that the SNR threshold providing minimum error in the truncation process may be unknown to us, the optimal B1-weighted combined method is the optimal choice for R2* measurement.

15:00         4637.     A General Method for Correction of Intensity Inhomogeniety in Two Point Dixon Imaging

Ola Friman¹, Olof Dahlqvst Leinhard^2,3, Peter Lundberg^2,3, Magnus Borga^3,4

¹MeVis Research, Bremen, Germany; ²Department of Medicine and Health, Linköping University, Linköping, Sweden; ³Center for Medical Image Science and Visualization (CMIV), Linköping University, Linköping, Sweden; ⁴Department of Biomedical Engineering, Linköping University, Linköping, Sweden

Two point Dixon imaging can be used for quantitative fat estimation. However, field inhomogeneities pose a problem that needs to be corrected for before quantitative measurements can be obtained. We present a general framework for field inhomogeneitiy correction by fitting a set of smooth 3D spatial basis functions to voxels with high fat content. By choosing the number of basis functions, the smoothness constraint of the field can be controlled. The method is evaluated by measuring the FWHM of the fat peak in histograms for different number of basis functions. It is also compared to a previous method with good results.

13:30         4638.     Scanner Calibration for Multisite Geometric Accuracy: How to Do It

Berkay Kanberoglu¹, Ted Trouard², Lina Karam¹, Josef P. Debbins³

¹Electrical Engineering, Arizona State University, Tempe, AZ, USA; ²Biomedical Engineering, University of Arizona, Tucson, AZ, USA; ³Keller Center for Imaging Innovation, Barrow Neurological Institute, Phoenix, AZ, USA

This study uses available processing methods to build a "pipeline" which can geometrically calibrate multiple MRI systems and then independently verify the calibration for use in longitudinal volumetric brain studies.

14:00         4639.     PET Effects on MR Data Quality in an Integrated MR-PET Human Brain Scanner

Thomas Benner¹, Andre J. W. van der Kouwe¹, Ciprian Catana¹, Michael Hamm², A. Gregory Sorensen¹, Bruce R. Rosen¹

¹Radiology, Athinoula A. Martinos Center, Charlestown, MA, USA; ²Siemens Medical Solutions, Charlestown, MA, USA

New combined MR-PET systems allow simultaneous acquisition of MR and PET data. We sought to evaluate the effect of the PET insert of a BrainPET prototype on the quality of the MR data with regard to RF noise, shim, signal-to-noise ratio (SNR) and temporal SNR. Two sets of scans were performed: with the PET insert in operating mode in the scanner bore and with the PET insert switched off outside the bore. The result show that the PET insert does not appear to have any significant effect on the MR scanner data regarding RF noise, shim, SNR and temporal SNR.

14:30         4640.     Motion Induced Magnetic Susceptibility and Field Map Estimation in FMRI

Desmond Yeo¹, Jeffery Fessler², Boklye Kim³

¹tbyeo@umich.edu; ²fessler@eecs.umich.edu; ³boklyek@umich.edu, Ann Arbor, MI, USA

Head rotation during fMRI time series acquisition may cause nonlinear field-inhomogeneity changes and lead to dynamic geometric distortion. The use of image registration to estimate dynamic field inhomogeneity maps from a static field map is not sufficient. This work retrospectively estimates magnetic susceptibility induced field maps of an object affected by out-of-plane rotation, given a static field map and the associated object motion parameters. It estimates a susceptibility map from a static field map using regularized image restoration techniques, and applies rigid body motion to the estimation. The dynamic field map is then computed using susceptibility voxel convolution.

15:00         4641.     Performance of Stereo Vision and Retro-Grate Reflector Motion Tracking Systems in the Space Constraints of an MR Scanner

Brian Armstrong¹, Brian Andrews-Shigaki², Robert T. Barrows¹, Todd P. Kusik¹, Thomas Ernst², Oliver Speck³

¹Electrical Engineering and Computer Science, UW-Milwaukee, Milwaukee, WI, USA; ²Medicine, University of Hawaii, Honolulu, HI, USA; ³Biomedical Magnetic Resonance, Otto-von-Guericke University, Magdeburg, Germany

Motion tracking and prospective compensation is one approach to reducing motion artifact in MR scans. Optical motion tracking is a leading candidate method to obtain the needed real-time motion data. The repeatability and accuracy of two optical motion tracking systems, stereo vision (SV) and retro-grate reflector (RGR), are investigated under the space constraints of an MR scanner. An optical bench and precision tilting rotary table provide geometric stability and highly accurate presentation of poses. Results indicate that achieving sufficiently accurate motion tracking in an MR bore remains a challenge for prospective motion compensation.

13:30         4642.     Combine Reconstructions Using Non-Local Operator and Its Application in PPI

Weihong Guo¹, Hu Cheng², Feng Huang³

¹Department of Mathematics, University of Alabama, Tuscaloosa, AL, USA; ²Department of Psychological and Brain Sciences, Indiana University, IN, USA; ³Advanced Concept Development, Invivo Corporation, Gainesville, FL, USA

Given a set of reconstructions with different noise/artifact distribution, it is possible to generate an image with higher signal to noise ratio (SNR) than each single reconstruction through weighted summation. An efficient weight calculation method is proposed in this work to automatically and adaptively integrate advantages of GRAPPA and SENSE reconstructions.Quantitative and qualitative results have shown advantages of the proposed model.

14:00         4643.     Retrospective Self-Navigated Cine Imaging Using the Unused Echo in Alternating TR SSFP

Hsu-Lei Lee¹, Ajit Shankaranarayanan², Krishna S. Nayak^1,3

¹Ming Hsieh Department of Electrical Engineering, University of Southern California, Los Angeles, CA, USA; ²Global Applied Science Lab, GE Healthcare, Menlo Park, CA, USA; ³Keck School of Medicine, University of Southern California, Los Angeles, CA, USA

In alternating TR SSFP, a low-resolution readout of the target scan plane can be acquired during the often-unused short TR. We collect data from free-breathing continuous wideband SSFP sequence and demonstrate successful retrospective self-navigation using this signal. This method can avoid temporal blurring caused by respiration in a cardiac scan without using a separate navigator sequence, removes the dependence on patients’ ability to perform breath-holds, and allows for longer scan time to acquire higher spatial and temporal resolution images.

14:30         4644.     Cardiac and Respiratory Motion Compensated Reconstruction Driven Only by 1D Navigators

Freddy Odille¹, Sergio Uribe², Tobias Schaeffter², David Atkinson¹

¹Centre for Medical Image Computing, University College London, London, UK; ²Division of Imaging Sciences, King's College London, London, UK

Motion compensated reconstruction for arbitrary motion is a difficult task requiring prior knowledge about motion. One way to deal with this is to constrain the time component of motion to be correlated with certain motion sensors, such as bellows, ECG, or navigators. Here we propose an extended analysis of MR-navigators, based on 1D nonrigid registration on a per-coil basis, and principal component analysis of the resulting 1D motion fields. The method allows detection of motion eigenmodes from the imaging data themselves, which are then used to constrain a motion-compensated reconstruction for free-breathing cardiac imaging, without the need for external ECG.

15:00         4645.     Real-Time Adaptive Motion Correction for Coronary MR Angiography

Markus Henningsson^1,2, Sergio A. Uribe Arancibia¹, Andrea J. Wiethoff^1,3, Philip G. Batchelor¹, Rene M. Botnar¹

¹Division of Imaging Sciences, King's College London, London, UK; ²NIHR Biomedical Research Centre at Guy’s & St Thomas’ Hospital and King’s College London , London, UK; ³Philips Healthcare, Reigate, UK

Artifacts due to residual coronary motion remain an impediment to diagnostic coronary MRA in a small but substantial amount of patients. The purpose of this study was to investigate a method (AF) that determines a patient specific correction factor which continuously updates throughout the course of the scan. 7 healthy volunteers were scanned and the LCA visibility was analyzed, and the result showed that the AF performs better than a constant 0.6 factor with a navigator on the diaphragm, as well as a 1.0 factor with a navigator on the heart.

14:00         4646.     Real-Time Motion Corrected MRS Using EPI Navigators

Aaron T. Hess¹, Matthew Dylan Tisdall^2,3, Ernesta M. Meintjes¹, Andre J. W. van der Kouwe^2,3

¹Human Biology, University of Cape Town, Cape Town, South Africa; ²Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, MA, USA; ³Department of Radiology, Harvard Medical School, Brookline, MA, USA

In this abstract we present a real-time motion navigation technique for MRS. The technique uses a low resolution echo planar imaged (EPI) volume acquisition (duration < 500ms), inserted in the relaxation time of each TR before the spectroscopy excitation. The EPI images are processed online and co-registered using PACE in order to provide real-time feedback to adjust the imaging coordinate system for rigid body translation and rotation. Our results demonstrate this technique can be incorporated into single voxel spectroscopy without increasing acquisition duration or impacting on the spectral quality.

14:30         4647.     Phase Based Navigator Echo Analysis with a Virtual Large Loop RF Coil

Yuji Iwadate¹, Kenichi Kanda², Akira Nabetani¹, Tetsuji Tsukamoto¹

¹Japan Applied Science Laboratory, GE Yokogawa Medical Systems, Hino, Tokyo, Japan; ²MR Engineering, GE Yokogawa Medical Systems, Hino, Tokyo, Japan

Phase based navigator echo analysis is useful in navigator gated imaging but sometimes affected by an inhomogeneous sensitivity of a receiver RF coil. We propose a method that uses virtual large loop coil to obtain a homogeneous phase background. By adding two navigator data sets vectorially, sensitivity of the virtual large loop coil had a homogeneous phase, which resulted in accurate motion detection with phase based navigator echo analysis. Future work will involve the phase compensation to apply to a larger number of coil elements of phased arrays.

15:00         4648.     The Effects of Induced B₀ and B₁ Magnetic Field Perturbations on Efforts to Image Near Embedded Metal Hardware

Kevin M. Koch¹, Kevin F. King¹, Graeme C. McKinnon¹

¹Applied Science Laboratory, GE Healthcare, Waukesha, WI, USA

Recent developments have demonstrated an improved capability of MR imaging near metal implants. These new methods are capable of removing bulk image distortions near commonly utilized metal implants. Here, we explore the more nuanced image intensity effects due to B₁ perturbations and severe B₀ gradients on these recently demonstrated imaging approaches. Computational field estimates are utilized to simulate MR images near simple metal implant geometries. These simulations generally agree with empirically measured images, and therefore help explain the remaining limitations to imaging near metal implants.

15:30         4649.     Quantitative Evaluation of Metal Artifact with New Turbo Spin Echo Imaging Techniques

Chen Lin¹, Brian Dale², Larry Friggle³, Leverett Neville¹, Annette Douglas-Akinwande¹, Kenneth Buckwalter¹

¹Radiology, Indiana University School of Medicine, Indianapolis, IN, USA; ²Siemens Healthcare; ³Radiloogy, Clarian Health Partners, Indianapolis, IN, USA

The artifact from metal hardware was evaluated and quantified for 2D TSE with Cartesian and BLADE (radial) k-space samplings as well as for 3D TSE with constant and variable refocusing flip angle schedules. These results can be used to optimize imaging of patients with orthopedic hardware.

13:30         4650.     Rigid Body Motion Detection with Lissajous Navigator Echoes

Marcus G. Ullisch^1,2, Tony Stöcker¹, Mark A. Elliott³, Kaveh Vahedipour¹, N. Jon Shah^1,4

¹Institute for Neurosciences and Biophysics, Research Centre Juelich, Juelich, Germany; ²Department of Psychiatry and Psychotherapy, University Hospital Aachen, Aachen, Germany; ³Department of Radiology, University of Pennsylvania, Philadelphia, PA, USA; ⁴Faculty of Medicine, Department of Neurology, RWTH Aachen University, JARA, Aachen, Germany

Spherical Navigator Echoes have been used successfully as a means of estimating rotation in MRI but suffer from one shortcoming. Near the pole the slewrate of the gradient waveform increases so rapidly, that the pole caps have to be left unsampled. Thus, rotation estimation errors occur when features rotate out of the sampled area. For rotations perpendicular to the windings of the spiral a reduced accuracy compared to rotations along the windings has been reported. Here we present an alternative sampling scheme for the spherical surface which is able to sample the complete surface while staying within the slewrate limits of the hardware.

14:00         4651.     Noise Bias Correction for Signal Averaged Images

Elena Olariu¹, Arturo Cardenas-Blanco², Ian Cameron^1,3

¹Physics, Carleton University, Ottawa, Ontario, Canada; ²Ottawa Health Research Institute; ³Diagnostic Imaging, Ottawa General Hospital

Clinical MR images are corrupted by noise which may reduce the reliability of quantitative analyses. The extraction of the true MR signal intensity from noisy MR magnitude images is confounded by a bias, which will be referred to here as Rician Bias (RB), caused by noise rectification in the magnitude calculation for low intensity pixels. Averaging in the image domain reduces the effective noise but not the noise bias. For low SNR a post-processing scheme to correct the noise bias combined with a limited amount of signal averaging is preferable. The RB correction method discussed here, which is an implementation of the theory developed by Koay and Basser, has been previously described2. The results are extended here to consider the effect of signal averaging and inaccuracies in the value of óg used.

14:30         4652.     PARAMAP: An Automated Imaging Analysis Tool for Quantitative CEST Molecular Imaging: Validation in Vitro

Julien Flament¹, Benjamin Marty¹, Sébastien Mériaux¹, Julien Valette¹, Christelle Medina², Caroline Robic², Marc Port², Franck Lethimonnier¹, Fawzi Boumezbeur¹

¹NeuroSpin, I2BM, Commissariat à l'Energie Atomique, Gif-sur-Yvette, France; ²Research Division, Guerbet, Roissy-Charles de Gaulle, France

CEST agents are promising new contrast agents for MR molecular imaging. Yet, their susceptibility to parameters such as B₀ or B₁ is a real issue to achieve truly quantitative CEST imaging. In this study, we acquired B₀ and B₁ maps along with the CEST on and off-resonance images to generate quantitative CEST maps using a home-made imaging analysis software called PARAMAP. PARAMAP is based on the numerical simulation of the Bloch equations modified for chemical exchange incorporating B₀ and B₁ dependencies. The efficiency of our analysis tool was verified in vitro.

15:00         4653.     Rapid Motion Detection Using Pre-Rotated Baseline Spherical Navigator Echoes

Junmin Liu^1,2, Maria Drangova^1,2

¹Imaging Research Laboratories, Robarts Research Institute, London, Ontario, Canada; ²Medical Biophysics, Schulich School of Medicine and Dentistry, University of Western Ontario, London, Ontario, Canada

A new spherical navigator echo (SNAV) registration method is presented. Our algorithm starts from collecting a set of pre-rotated baseline SNAVs, which randomly and uniformly cover a predetermined rotation range. The rotation angles between a transformed and baseline position are determined by identifying the minimum of a cost function calculated by subtracting the transformed SNAV from the set of pre-rotated baseline SNAVs. Preliminary results demonstrate that accurate translation and rotation parameters can be extracted in less than 50 ms, compared to the 5 seconds required for the previously used iterative registration approaches.

13:30         4654.     Increasing the Effectiveness of the UNFOLD Technique by Removing Phase Trends

Yanle Hu¹

¹Imaging Research Center, University of Texas at Austin, Austin, TX, USA

To increase temporal resolution in dynamic MRI studies, under-sampling techniques are typically used. One way to remove aliasing artifacts associated with under-sampling is to use the UNFOLD technique, which purposely interleaves under-sampled k-space trajectories across time frames to move aliasing peaks away from true signal peaks in the spectrum, and then filter out aliasing peaks to get alias-free images. The phase drift in signal may shift and broaden aliasing peaks. Removing linear and quadratic phase trends before the implementation of UNFOLD can ensure that aliasing artifacts are removed effectively without increasing the number of frequency components to be filtered out.

14:00         4655.     TV Regularization for Segmented GRAPPA with Higher Net Acceleration Factor

Yunmei Chen¹, Xiaojing Ye¹, Haili Zhang¹, Jiangli Shi¹, Feng Huang²

¹Department of Mathematics, University of Florida, Gainesville, FL, USA; ²Invivo Corporation, Gainesville, FL, USA

Segmented GRAPPA is superior to GRAPPA but requires significant amount of ACS lines. We propose a total variation regularized GRAPPA technique to produce a full calibration k-space with limited ACS lines. In the next step, the full calibration k-space data is used as calibration signal for segmented GRAPPA. The experimental results, with comparisons with GRAPPA and high-pass GRAPPA, show that the proposed method can generate images with lower artifacts/noises level when only 32 ACS lines are used with reduction factor 4. This work enables segmented GRAPPA with limited ACS lines, and hence increases the net acceleration factor while preserving the image quality.

14:30         4656.     MPRAGE Using EPI Navigators for Prospective Motion Correction

Matthew Dylan Tisdall^1,2, Aaron Hess³, Andre van der Kouwe^1,2

¹Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, MA, USA; ²Radiology, Harvard Medical School, Brookline, MA, USA; ³Faculty of Health Sciences, University of Cape Town, Cape Town, South Africa

We present initial results from a novel method for real-time motion correction in MPRAGE imaging. We use high-speed, 32^3 EPI volumes as navigators, interspersed in the dead time of the MPRAGE sequence. We estimate rigid body motion from the EPI volumes and correct the MPRAGE's acquisition location before each readout block.

15:00         4657.     Iterative Image Reconstruction for PROPELLER-MRI: Through-Plane Motion Considerations

Ashish A. Tamhane¹, Mark Anastasio¹, Konstantinos Arfanakis¹

¹Department of Biomedical Engineering, Illinois Institute of Technology, Chicago, IL, USA

In PROPELLER image reconstruction using gridding, through-plane motion artifacts are minimized by assigning low “quality” weights to affected blades. Iterative image reconstruction can reduce artifacts that arise from data inconsistencies and minimize image noise by use of regularization strategies. “Quality” weights similar to those used in gridding can also be incorporated in iterative image reconstruction in order to reduce the effects of through-plane motion. The goal of this study was to assess the performance of iterative PROPELLER image reconstruction with un-weighted and weighted least squares cost functions in the presence of through-plane motion, and compare it to that of gridding.

13:30         4658.     Correction for Chemical Shift Artifacts in Magnetic Resonance Electrical Impedance Tomography

Mark Jason Hamamura¹, Orhan Nalcioglu¹, Lufti Tugan Muftuler¹

¹Tu & Yuen Center for Functional Onco-Imaging, University of California, Irvine, CA, USA
Chemical shift artifacts in magnetic resonance electrical impedance tomography (MREIT) degrade the accuracy of the reconstructed conductivity. In this study, we investigated the use of a modified 3-point Dixon technique to remove these artifacts in a simple fat/water phantom. The results demonstrate that this technique can be used to correct for chemical shift artifacts in MREIT.

14:00         4659.     DCE-MRI Tumor Registration by Using TV-L1 Optical Flow

Michael Hofer¹, Thomas Pock², Karin Kapp³, Thomas Bauernhofer⁴, Franz Ebner⁵, Rudolf Stollberger¹

¹Institute of Medical Engineering, Graz University of Technology, Graz, Austria; ²Institute for Computer Graphics and Vision, Graz University of Technology, Graz, Austria; ³Department of Radiotherapy-Radiooncology, Medical University Graz, Graz, Austria; ⁴Department of Oncology, Medical University Graz, Medical University Graz, Austria; ⁵Department of Radiology, Medical University Graz, Graz, Austria

Non-rigid image registration, independent from contrast changes in dynamic time series, is implemented using a TV-L1 optical flow algorithm to compensate physiological and unintentional motion to get a better quantification of DCE-MRI data. Contrast invariant structure images of the whole sequence are created and registered to a template image which is defined by the structure image of the first component of the principal component analyses. The performance of the algorithm was evaluated using real patient DCD-MRI tumor sequences where movement artifacts of tumor and surrounding tissue could be markedly reduced which subsequently improved the quantification of DCE-MRI data.

14:30         4660.     Self-Navigated IDEAL Water-Fat Separation

Yun Jiang¹, Jeffrey Tsao¹

¹Global Imaging Group, Novartis Institutes For BioMedical Research, Cambridge, MA, USA

IDEAL water-fat separation is being adopted in quantitative analysis of adiposity. The presence of motion during acquisition causes motion artifacts, which can result in quantification inaccuracies. To overcome this challenge, a double-echo navigator technique was incorporated in the IDEAL sequence to monitor the signal fluctuation caused by motion. Retrospective motion correction led to a substantial reduction of motion artifacts, thereby improving the accuracy and robustness of the quantification.

15:00         4661.     Impact of B₁ Inhomogeneities on AIF Selection in DCE-MRI at 3 Tesla

Robert Merwa¹, Gernot Reishofer², Thorsten Feiweier³, Karin Kapp⁴, Franz Ebner², Rudolf Stollberger¹

¹Institute of Medical Engineering, Graz University of Technology, Graz, Austria; ²Department of Radiology, Medical University of Graz, Graz, Austria; ³MED MR PLM AW Neurology, Siemens AG Healthcare Sector, Erlangen, Germany; ⁴Department of Radiation Therapy, Medical University of Graz, Graz, Austria

Dynamic contrast-enhanced MRI was performed in combination with a special sequence in order to determine B1 inhomogenities. AIF and tissue concentrations were calculated and the kinetic parameters Ktrans and Ve were determined with a generalized kinetic model. The absolute deviation of the maximum values of two comparable AIFs can be improved by a factor greater than 10 and the root mean square deviation concerning the two AIFs can be decreased by a factor greater than 5 if B1 inhomogeneities are corrected. Also the deviations of Ktrans and Ve in respect of the two AIFs are significantly lower.

14:00         4662.     Morphometry of Intratumoral Enhancement Patterns on 4D Spectral Images for Differential Diagnosis of Breast Tumors in Dynamic Contrast-Enhanced MRI

Sang Ho Lee^1,2, Jong Hyo Kim^2,3, In Chan Song^2,3, Yun Sub Jung^1,2, Jeong Seon Park⁴, Woo Kyung Moon³

¹Interdisciplinary Program in Radiation Applied Life Science, Seoul National University College of Medicine, Seoul, Korea; ²Institute of Radiation Medicine, Seoul National University Medical Research Center, Seoul, Korea; ³Department of Radiology, Seoul National University College of Medicine, Seoul, Korea; ⁴Department of Radiology, Hanyang University College of Medicine, Seoul, Korea

This study demonstrates the importance of quantitatively capturing spatio-temporal properties of intratumoral enhancement patterns for MR-based breast tumor diagnosis. MR-time-series images were registered for motion compensation and tumor regions were segmented semi-automatically from our proposed perfusion index map enhancing tumor contrast. Eigenvalues were obtained for voxel-wise temporal enhancement curves within tumor by using singular value decomposition (SVD), generating eigenvalue maps. The spatial variations of eigenvalues within tumor were captured by 3D geometric moment invariants (GMIs). The potential of our SVD-based GMI features in differentiation of benign and malignant tumors is validated by classification performance using least square support vector machine.

14:30         4663.     Application of Vessel Exclusion for Development of Computer-Aided Diagnosis for Breast MRI

Muqing Lin¹, Ke Nie¹, Jeon-Hor Chen^1,2, Orhan Nalcioglu¹, Min-Ying Lydia Su¹

¹Tu & Yuen Center for Functional Onco-Imaging, University of California, Irvine, CA, USA; ²Department of Radiology, China Medical University, Taichung, Taiwan

The purpose is to evaluate the reduction of vessel contamination in the hot-spot search of breast lesions on DCE-MRI. This is particularly important for development of automated CAD that identified lesions and characterizes the suspicious impression. The blood vessels show strong contrast enhancements and the wash-out pattern, which is a strong malignancy feature. The initial hot spot search was performed using a kernel of 3x3 pixels throughout the entire breast. The vessels were identified using a computer algorithm-based method based on 2D and 3D. We have shown that our algorithm can successfully exclude vessels and reduce false positive detection.

15:00         4664.     Association of Carotid Atherosclerosis Eccentricity with High Risk Plaque Features: An MRI Study

Huijun Chen¹, Li Dong¹, William S. Kerwin¹, Wei Yu², Hideki Ota¹, Hunter Underhill¹, Xihai Zhao¹, Zhaoqi Zhang², Zach Miller¹, Thomas S. Hatsukami³, Chun Yuan¹

¹Department of Radiology, University of Washington, Seattle, WA, USA; ²Department of Radiology, Beijing Anzhen Hospital, Beijing, China; ³Department of Surgery, University of Washington, Seattle, WA, USA

Multi-contrast weighted MRI identified plaque features in carotid artery, such as intraplaque hemorrhage (IPH) and thin or ruptured fibrous cap, have been reported to be predictors of subsequent stroke. On the other hand, plaque eccentricity of carotid artery has been found to be associated with ipsilateral cerebrovascular events in an ultrasonography study. In this study, we sought to assess the association between plaque shape and high risk plaque features in carotid artery, and a novel eccentricity index (EI) of plaque using MR imaging was established. The result shows that the plaques with high risk features (presence of IPH and thin or ruptured fibrous cap) have relatively high EI, which suggests the proposed EI from carotid MRI could be a potential predictor of high risk plaque.

15:30         4665.     Depiction of SPIO in Atherosclerotic Plaque Using True Resolution SGM

Gopal Varma¹, Marcus Richard Makowski¹, Hannes Dahnke², Stephen Keevil³, Rene M. Botnar¹, Tobias Schaeffter¹

¹Division of Imaging Sciences, King's College London, London, UK; ²Philips Healthcare, Hamburg, Germany; ³Medical Physics, Guy's & St Thomas' NHS Foundation Trust, London, UK

Detection of atherosclerotic plaque with iron oxide contrast agents that induce a susceptibility effect in MRI is sometimes hindered by competing sources of negative contrast. Although positive contrast techniques aid in detection, hyper-intensities are often over-emphasised and may thus fail to accurately depict the plaque. We present a “true resolution” positive contrast method post-processing that provides a more accurate depiction of e.g. SPIO induced susceptibility effect with relation to its source.

13:30         4666.     A MRI-Based Virtual Cystoscopy System for Evaluation of the Entire Bladder

Jerome Zhengrong Liang¹, Lihong Li², Chaijie Duan¹, Su Wang¹, Mark Wagshul¹, Hongbing Lu³

¹Radiology, Stony Brook University, Stony Brook, NY, USA; ²Dept. of Engineering Science and Physics, The City University of New York, Staten Island, NY, USA; ³Biomedical Engineering, Fourth Military Medical University, Xi¡¯an, Shannxi, China

Bladder cancer is the fifth leading cause of cancer-related deaths in the US and is difficult to manage because of the high recurrence rate after resection of the tumors (as high as 80%). It is essential to detect bladder abnormalities in a non-invasive and convenience manner, especially for follow-ups on resection. This paper presents a MRI-virtual cystoscopy system, which extracts the bladder wall from T1 images where the wall is enhanced while the urine and surrounding fat are suppressed. It analyzes the extracted wall and detects abnormal features automatically. Test results are encouraging by FROC merit.

14:00         4667.     Deconvolution and QSI of Simulated Phase Images of the Human Brain: Applications to Assess Susceptibility

Guenther Grabner^1,2, Siegfried Trattnig¹, Markus Barth²

¹Department of Radiology, Medical University of Vienna, Vienna, Austria; ²Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen, Nijmegen, Netherlands

MR phase images reflect the magnetic susceptibility of tissues but the phase image is a convolution of the susceptibility map with the typical pattern of magnetic dipole. This leads to very complex phase patterns in areas of differing susceptibility. The challenge is to estimate the underlying tissue susceptibilities. We wanted to assess how accurate the underlying properties can be determined by performing filtered deconvolution and QSI on a simulated phase model.

14:30         4668.     Simulation of Nasal Air Flow from MRI Data – a Feasibility Study

Arthur Peter Wunderlich¹, Barbara Doll², Marc Scheithauer³, Wolfgang Freund¹, Karsten Urban⁴, Ulrich Simon²

¹Diagnostic and Interventional Radiology, Univ.-Clinic Ulm, Ulm, Baden-Württemberg, Germany; ²Scientific Computing Centre, University Ulm, Ulm, Baden-Württemberg, Germany; ³Otolaryngology, Univ.-Clinic Ulm, Ulm, Baden-Württemberg, Germany; ⁴Institue for Numerical Mathematics, University Ulm, Ulm, Baden-Württemberg, Germany

To study the feasibility of simulating the nasal air flow in inspiration and expiration with finite volume methods, we investigated one healthy subject with high-resolution MRI during breathing of a) room air and b) menthol. In a third experiment, the same sequence was used c) after inhalation of xylometazoline, a widely used decongestant drug. Nasal airways were segmented from MRI data and inspiratory as well as expiratory flow in all three conditions was simulated with mathematical methods. Simulations reveal pressure-flow-curves separately for both conchae as well as flow distribution in nasal airways and their dependence on medication.

15:00         4669.     MR Elastography Inversions Without Phase Unwrapping

Kevin John Glaser¹, Richard Lorne Ehman¹

¹Radiology, Mayo Clinic, Rochester, MN, USA

MR elastography (MRE) is a phase-based technique for noninvasively measuring the mechanical properties of tissue in vivo. MRE data can be adversely affected by unmanageable phase aliasing due to motion, SNR, and various magnetic field effects, which can degrade estimates of material properties like the shear stiffness. This work derives and demonstrates a method for processing MRE data without having to unwrap the phase data which yields elastograms comparable to those from conventional algorithms. It is expected that this will increase the flexibility and utility of MRE for applications including brain and abdominal imaging.

13:30         4670.     A Model-Based Fully Automatic Algorithm for Brain Extraction on 3D MRI Scans

Babak A. Ardekani¹, Alvin H. Bachman¹

¹Center for Advanced Brain Imaging, The Nathan S. Kline Institute for Psychiatric Research, Orangeburg, NY, USA

A model-based automatic algorithm is presented for brain extraction on 3D MRI scans. Given a structural MRI of the head, the objective is to classify each voxel within the image volume as either brain or non-brain. Brain extraction, as a preliminary step, often improves the results of inter-subject non-linear image registrations. It is also a very useful first step in automatic brain tissue classification algorithms. We have compared the algorithm presented in this paper with several previously published programs and have found it to be superior to those techniques. An implementation of this algorithm is freely available online at: www.nitrc.org/projects/art.

14:00         4671.     A Robust Parametric Method for Bias Field Estimation and Segmentation of MR Images

Chunming Li¹, Li Wang², J. Chris Gatenby¹, Adam Anderson¹, John C. Gore¹

¹Vanderbilt University, Nashville, TN, USA; ²School of Computer Science & Technology, Nanjing University of Science and Technology, China

In this work, we propose a novel parametric method for joint image segmentation and bias field estimation for MR images. The bias field is parameterized as a linear combination of smooth basis functions. Image segmentation and bias field estimation are performed by minimizing a cost function. A desirable advantage of the proposed method is its robustness to initialization, which thereby allows fully automatic applications. Comparisons with other methods show the advantage of our method in terms of accuracy and robustness.

14:30         4672.     Template-Based Bias Correction: Application to Paediatric Brain MRI

Maria Murgasova¹, Joseph V. Hajnal², Serena J. Counsell², A David Edwards², Daniel Rueckert¹

¹Department of Computing, Imperial College London, London, UK; ²Robert Steiner Magnetic Resonance Unit, Imaging Sciences Department, MRC Clinical Services Centre, Imperial College London, London, UK

We propose a novel template-based bias correction method highly effective for removal of stronger bias fields usually present in MR images acquired by modern high magnetic field MR scanners. The bias field is estimated by weighted least square B-spline fitting to a difference of log-transformed image and an aligned and intensity-adjusted template image. The method is applied 35 pediatric brain MR images acquired using a MP RAGE sequence on a 3T MR scanner. The bias correction with B-spline control point spacing between 50 and 75 mm substantially improves EM-based classification of brain tissues in these images.

15:00         4673.     Artificial Neural Network Analysis of Differences in Fiber Tracks Between Term and Preterm Children

Lutfi Tugan Muftuler¹, Ke Nie¹, Orhan Nalcioglu¹, Christine E. McLaren², Min Ying Su¹

¹Center for Functional Onco-Imaging, University of California, Irvine, CA, USA; ²Department of Epidemiology, University of California, Irvine, CA, USA

VBM techniques can be applied to DTI parameter maps to investigate local differences in white matter structures between patient and control groups. However, tractography is essential to investigate tract morphology and brain connectivity. Here, we developed an Artificial Neural Network based analysis to select a set of features that can achieve the highest differentiation power between the two groups. Once these features are found, inferences about the morphological differences in fiber tracts can be made. Compared to traditional statistical analysis methods, ANN was found to have higher prediction rates in complex and non-linear relationships among a large number of variables.

13:30         4674.     High Resolution Positive Contrast Imaging: Application to the Visualization of Localized Vascular Abnormalities

Thomas G. Perkins¹, Julien Sénégas², Hannes Dahnke², Elliot Sandberg^3,4, William W. Orrison^5,6

¹Philips Healthcare, Cleveland, OH, USA; ²Philips Research Europe, Hamburg, Germany; ³Department of Radiology, VAMC, Denver, CO, USA; ⁴Departments of Radiology and Neurology, University of Colorado Health Science Center, Aurora, CO, USA; ⁵Nevada Imaging Centers, Las Vegas, NV, USA; ⁶Touro University Nevada College of Osteopathic Medicine, Henderson, NV, USA

Positive contrast imaging based on susceptibility gradient mapping (SGM) is a relatively new technique that produces maps with high contrast-to-noise ratio for areas of local magnetic susceptibility variation. Applying high resolution SGM to assess localized abnormalities associated with hemosiderin deposition in conditions such as traumatic brain injury or cavernous angiomas could provide additional clinical information for detection and diagnosis. The purpose of this study was to explore the potential clinical utility of SGM to provide better characterization of cavernous angiomas, and shows the potential utility of positive contrast imaging for the visualization of localized vascular abnormalities.

14:00         4675.     Rapid Quantification of Stroke Infarct from Diffusion Weighted Images

Sumit K. Nath¹, Rakesh Mullick¹, Dattesh Shanbhag¹, Uday Patil²

¹Imaging Technologies Lab, GE Global Research, Bangalore, Karnataka, India; ²Conulting Radiologist, GE Global Research, Bangalore, Karnataka, India

A vital step in the evaluation of acute and chronic stroke involves volume quantification of the infarct as determined from diffusion-weighted images (DWI). The routine assessment of infarct size is by visual review of the approximate size reported along with the location of the defect with reference to the brain anatomy. Assessment of the infarct volume is critical to management decisions in acute stroke, especially for thrombolytic therapy, and calls for quantitative methods. There are multiple efforts to develop post-processing methods to segment and automatically measure the infarct from DWI images. Given T2 shine-through artifacts from pre-existing old infarcts in the DWI images make this task very challenging. In order to offer a robust workflow for this evaluation we present a hybrid (visualization and segmentation) method to address this. The core concept is to integrate the presentation of maximum intensity projections (MIP) of the DWI data to the user to select suggestive infarct regions and use an underlying level-set based (top-down) image segmentation approach for rapid quantification of suspect regions

14:30         4676.     Noise Performance of IDEAL Considering Relaxation and Multiple Spectral Peaks of Fat

Stephan van Zyl¹, Peter Börnert², Holger Eggers², Michael M. Morlock¹

¹Hamburg University of Technology, Hamburg, Germany; ²Philips Research Europe, Hamburg, Germany

In the past, water/fat separation algorithms like IDEAL have been further refined to account for finite signal lifetime (T2*) and the multi-peak nature of the fat spectrum. IDEAL is known to obtain good noise performance. Here, the noise performance of T2*-IDEAL and MP-IDEAL was evaluated using the Cramér-Rao bound (CRB). The results were independently verified with the effective number of signal averages (NSA) obtained by Monte Carlo simulations and phantom experiments. Investigations have shown that a reduction in NSA is to be expected when using T2*-IDEAL and MP-IDEAL. Short encoding times are therefore absolutely essential.

15:00         4677.     Efficient and Effective Anisotropic Smoothing of Diffusion Tensor Images in Log-Euclidean Framework

Qing Xu¹, Adam W. Anderson¹, John C. Gore¹, Zhaohua Ding¹

¹Vanderbilt University Institute of Imaging Science (VUIIS), Vanderbilt University, Nashville, TN, USA

We present a technique to denoise diffusion tensor images by performing non-iterative anisotropic smoothing in the Log-Euclidean framework. The diffusion tensors are first transformed to tensor logarithm space to perform Euclidean computing for vectors. Then an anisotropic smoothing algorithm for multi-channel image is implemented with an unconditionally stable and second order accurate semi-implicit scheme, which allows us to choose large step size and thus use one iteration to achieve optimal effect. The tests with in vivo DTI data have demonstrated that there is up to 50% improvement on the principal diffusion directions with one iteration of anisotropic filtering.

14:00         4678.     Leveraging Open Source/Freeware Tools and Design Methodologies in Translational Research: A Siemens-Based Case Study -- From Raw Diffusion and Structural Data to PACS and the Web

Rudolph Pienaar¹, Ruopeng Wang¹, Bruce Fischl¹, Van Wedeen¹, Patricia Ellen Grant¹

¹Radiology, Massachusetts General Hospital, Boston, MA, USA

This paper demonstrates a simple and robust method of assembling powerful image processing pipelines based on collections of largely independent image processing tools. These pipelines are open source in nature, and underlying components either open source or freeware. In this study, the pipelines are deployed to a Linux workstation and images collected on Siemens scanners. This Linux workstation can process data directly on the scanner itself, or receive DICOM data from a conventional hospital PACS. Final outputs are automatically parsed for access via the web, or are re-transmitted back to a PACS.

14:30         4679.     NeuroGrid Portal: A Web Interface Providing Processing Tools for Neuro-Imaging Research

Ali A. Khanban^1,2, Yu Chen³, Gianlorenzo Fagiolo², Nick C. Fox⁴, Derek L. G Hill⁵, Daniel Rueckert¹, Jo V. Hajnal²

¹Imperial College London, UK; ²Robert Steiner MRI Unit, Imaging Sciences Department, MRC Clinical Sciences Centre, Hammersmith Hospital, Imperial College London, London, UK; ³University College London, UK; ⁴Dementia Research Centre, Institute of Neurology, University College London, UK; ⁵IXICO Ltd, The London BioScience Innovation Centre, 2 Royal College Street, London, UK

NeuroGrid portal is a web portal designed to provide medical image processing applications in an easy to use web-based environment for the neuro-imaging community. Many applications which require high technical expertise to install and fine tune can be accessed with a few mouse clicks and will be run on remote servers. All the transactions use secure HTTPS protocols and users can share their data with other registered users if they want to. Some of the most frequent tasks faced in imaging research, such as image format conversion, correction of anatomical orientation, registration and segmentation are available on NeuroGrid Portal.

15:00         4680.     High Dynamic Range Processing Improves Global Contrast

Jerod Michael Rasmussen¹, Vinod Kaimal²

¹University of California, Irvine, Irvine, CA, USA; ²MIR Preclinical Services, Ann Arbor, MI, USA

High Dynamic Range (HDR) techniques are applied to images weighted by T1, T2 and proton density values to create a novel HDR image. ROIs were drawn around specific tissues including bone, muscle and ligament. Each ROI was evaluated to determine effects of the HDR technique on local contrast. The resulting image retained the best local contrast features from each respective weighting, resulting in an increase in global contrast.

15:30         4681.     Consistency in Automated Versus Manual Definition of MRI Scan Volume Orientations of the Human Heart

Harald Sepp Heese¹, Sebastian P. M. Dries², Daniel Bystrov², Jochen Peters³, Olivier Ecabert³, Juergen Weese³, Chiel J. den Harder⁴, Wendy de Kok⁴, Arianne M. van Muiswinkel⁴

¹Philips Research , Hamburg, Germany; ²Philips Research, Hamburg, Germany; ³Philips Research, Aachen, Germany; ⁴Philips Healthcare, Best, Netherlands

Definition of anatomically consistent and accurate scan volume orientations is a prerequisite for diagnostic reliability both for manual and automated MRI scan planning. The proposed method automatically recognizes major structures of the human heart from unseen breathing-gated, ECG-triggered, isotropic 3D MRI volumes, and calculates anatomical landmarks of these structures. Based on a set of training samples, for which automatically calculated landmarks and user-defined examples of scan volume orientations are available, the proposed method infers corresponding scan volume orientations for unseen images. Performance of the proposed recognition/planning method is compared to manually defined scan volume orientation for a four chamber view.

13:30         4682.     Simultaneous Mapping of Functional Connectivity and Activation Using Real-Time FMRI

Dae-Jin Kim¹, Bumhee Park¹, Hae-Jeong Park¹

¹Department of Radiology and Nuclear Medicine and Research Institute of Radiological Science, Yonsei University College of Medicine, Seoul, Korea

The aim of this study is to develop real-time platform for functional connectivity and activation using real-time fMRI. We performed the motion fMRI experiments, and visualized the functional connectivity and activation for the ongoing paradigm. On 3.0T MRI scanner, fMRI scanning was performed using EPI sequences with TR=3sec. Online correlation coefficients as regional functional connectivity and task-related activations were calculated and simultaneously visualized for 116 brain regions. We demonstrated the feasibility of real-time fMRI monitoring for the functional connectivity. It was possible to monitor the functional activity in real time environments, and to localize the subject¡¯s task performance.

14:00         4683.     Simple Subject-Specific Spatial Smoothing for BOLD FMRI

Robert L. Barry^1,2, John A. Sexton^1,3, John C. Gore^1,2

¹Vanderbilt University Institute of Imaging Science, Nashville, TN, USA; ²Department of Radiology and Radiological Sciences, Vanderbilt University, Nashville, TN, USA; ³Department of Biomedical Engineering, Vanderbilt University, Nashville, TN, USA

Spatial smoothing (SS) is a crucial processing step preceding statistical analyses of blood-oxygenation-level-dependent (BOLD) fMRI data. Although previous works have shown that the optimal SS kernel size varies both between subjects and active regions of interest (ROIs) within subjects, the use of a single kernel size for one or many subjects is virtually ubiquitous. A simple and computationally efficient automated algorithm based on matched filter theory is presented that determines the SS kernel size to maximize t-statistics for a given ROI. The results also emphasize the benefits of locally adaptive SS techniques to optimize BOLD contrast-to-noise in single-subject analyses.

14:30         4684.     Evaluation of the Image Intensity Non-Uniformity Correction Power in Voxel-Based Morphometry

Masami Goto^1,2, Osamu Abe³, Tosiaki Miyati⁴, Hiroyuki Kabasawa⁵, Takeshi Iwatsubo⁶, Hiroshi Matsuda⁷, Fumio Yamashita⁸, Shigeki Aoki⁹, Harushi Mori³, Naoto Hayashi¹⁰, Kenji Ino¹, Keiichi Yano¹, Kyouhito Iida¹, Kazuo Mima¹, Kuni Ohtomo³

¹Radiological Technology, University of Tokyo Hospital, Bunkyo-ku, Tokyo, Japan; ²Graduate School of Medical Science, Kanazawa University, odateno, kanazawa, Japan; ³Radiology, University of Tokyo, Bunkyo-ku, Tokyo; ⁴Graduate School of Medical Science, Kanazawa University, odateno, Kanazawa, Japan; ⁵Japan Applied Science Laboratory, GE Yokogawa Medical Systems, Ltd, Hino-shi, Tokyo, Japan; ⁶Neuropathology, University of Tokyo, Bunkyo-ku, Tokyo, Japan; ⁷Nuclear Medicine, Saitama Medical University International Medical Center, Hidaka, Saitama, Japan; ⁸Radiology, National Center Hospital of Neurology and Psychiatry, Kodaira, Tokyo, Japan; ⁹Radiology, Juntendo University, Bunkyo-ku, Tokyo, Japan; ¹⁰Computational Diagnostic Radiology and Preventive Medicine, University of Tokyo, Bunkyo-ku, Tokyo, Japan

Because we compared the same volunteer group, it was supposed to be no significant group difference. However, there are significant group differences. The cause will be due to intensity non-uniformity in the 3T image. If there is no significant group difference, such an intensity correction is believed to be efficient. In the present study, the intensity non-uniformity correction is more effective in SPM5 than N3. It is impossible to measure �greal brain volume�h and there is no gold standard of brain volume measurement. However, this result suggests that intensity non-uniformity correction is necessary to reduce inhomogeneity between MR scanners.

15:00         4685.     Effects of the Modulation in the White Matter Voxel Based Morphometry

Ozlem Ozmen Okur¹, Cengizhan Ozturk¹, Kubilay Aydin²

¹Bogazici University Biomedical Engineering Institute, Istanbul, Turkey; ²Istanbul Medical School, Istanbul University, Instanbul, Turkey

Voxel Based Morphometry (VBM) aims to reveal the differences between different populations&#8217; brains by making voxel based comparisons on MR brain images. SPM is quite a common tool utilized for VBM analyses. We computed the goodness of non-rigid registration metric (mutual information) in order to reveal the relationship between the registration quality and VBM results in SPM5 in case of with and without modulation. Additionally, we registered the images onto the groups&#8217; own template (group averages), too; and evaluated these templates&#8217; effects in the VBM analysis.

13:30         4686.     A Robust Breast Segmentation Method to Support Computer Aided Evaluation and Breast Density Assessment

Gokhan Ertas¹, Martin O. Leach¹

¹Cancer Research UK Clinical Magnetic Resonance Research Group, Institute of Cancer Research and Royal Marsden NHS Foundation Trust, Sutton, Surrey, UK

Breast anatomy on MR images is affected by intensity inhomogeneity and partial volume effects. In images acquired from different centres and scanners, there is considerable variation in the extent of these effects.

14:00         4687.     Development of a Computer Algorithm-Based Method for Identification of Blood Vessels on Dynamic Contrast Enhanced Breast MRI

Muqing Lin¹, Jeon-Hor Chen^1,2, Ke Nie¹, Orhan Nalcioglu¹, Min-Ying Lydia Su¹

¹Tu & Yuen Center for Functional Onco-Imaging, University of California, Irvine, CA, USA; ²Department of Radiology, China Medical University, Taichung, Taiwan

We presented an algorithm based method for identification of blood vessels on dynamic contrast enhanced breast MRI. The algorithm is consisted of global histogram equalization, a filter bank based on Wavelet Transform and Hessian Matrix, and high-pass filtering to enhance vessels, then thresholding. The procedure was first performed on 2D MIP then extended to 3D. The detected vessels were converted to a vasculature skeleton by thinning, and compared to radiologist’s drawing. The mean correct detection rate was 85%, the incorrect detection rate (detected by algorithm not by radiologist) was 15%, and the missed-rate (identified by radiologist not by algorithm) was 21%.

14:30         4688.     Design of an MR Phantom for Comparison of Frequency Based Texture Analysis Techniques

Robert Allan Brown^1,2, Richard Frayne^2,3

¹Biomedical Engineering, University of Calgary, Calgary, AB, Canada; ²Seaman Family MR Research Centre, University of Calgary, Calgary, AB, Canada; ³Radiology and Clinical Neurociences, University of Calgary, Calgary, AB, Canada

Texture analysis is an image analysis technique that complements standard pixel intensity methods. Textural features of MR images are associated with clinically important genetic differences in some brain tumors as well as being of interest in detection of other diseases. This abstract reports the design of an MR phantom to investigate and compare texture analysis techniques. Two techniques are compared and their repeatability measured using this phantom and imaging from a volunteer.

15:00         4689.     Demonstration of a Novel Edge Analysis Technique Using a Purpose Built MR Phantom

Robert Allan Brown^1,2, Ashley D. Harris^1,2, Richard Frayne^2,3

¹Biomedical Engineering, University of Calgary, Calgary, AB, Canada; ²Seaman Family MR Research Centre, University of Calgary, Calgary, AB, Canada; ³Radiology and Clinical Neurociences, University of Calgary, Calgary, AB, Canada

In many diseases, the interface between diseased and healthy tissue may contain a great deal of information. One example is the brain tumor oligodendroglioma, where characteristics of the tumor margin are significantly associated with 1p/19q co-deletion, which is a marker for good response to chemo- and radiation-therapy. In this abstract we report the design of a phantom to simulate “invasive” and “non-invasive” tissue border properties, and use a novel edge analysis technique to quantify these features.

13:30         4690.     Automatic Quality Assessment in Structural Brain Magnetic Resonance Imaging

Bénédicte Mortamet¹, Matt A. Bernstein², Clifford R. Jack Jr², Jeffrey L. Gunter², Chadwick Ward², Paula J. Britson², Reto Meuli³, Jean-Philippe Thiran⁴, Gunnar Krueger¹

¹Advanced Clinical Imaging Technology, Siemens Suisse SA, Healthcare Sector IM&WS - CIBM, Lausanne, Switzerland; ²Mayo Clinic, Rochester, MN, USA; ³Centre Hospitalier Universitaire Vaudois and University of Lausanne; ⁴Ecole Polytechnique Fédérale de Lausanne (EPFL), Signal Processing Laboratory (LTS5), Lausanne, Switzerland

Quality assessment of MRI is of great importance to derive reliable diagnostic information. As automated quantitative image analysis is being increasingly used in routine, automated measures of quality are needed. Based on a single magnitude image, we propose a procedure that automates the classification of data quality and allows detecting patient-/scanner-related artifacts. Validated on 750 datasets, the approach proofs to be a very promising candidate to perform quality assurance analysis for clinical practice and research. It could greatly improve clinical workflow through its ability to rule-out the need for a repeat-scan while the patient is still in the magnet bore.

14:00         4691.     Simultaneous Identification of Noise and Estimation of Noise Standard Deviation in MRI

Cheng Guan Koay¹, Evren Özarslan¹, Carlo Pierpaoli¹

¹National Institutes of Health, Bethesda, MD, USA

Analysis of MRI data usually entails a series of processing steps. One of these steps is noise assessment, which includes both the identification of noise and the estimation of noise variance (standard deviation). In MRI, the identification of noise has received less attention than has the estimation of noise variance. Here, we propose a novel approach to simultaneously identify noise and estimate the standard deviation of noise from a data structure commonly used in MRI. Experimental data acquired using an 8-channel phased array coil were used to investigate the feasibility and the stability of the proposed technique.

14:30         4692.     A Framework for Correcting the Noise-Induced Bias in Noisy Magnitude MR Signals

Cheng Guan Koay¹, Evren Özarslan¹, Peter J. Basser¹

¹National Institutes of Health, Bethesda, MD, USA

A long-standing problem in MRI is correcting noise-induced bias in magnitude signals. This problem is particularly pressing in diffusion MRI obtained at high diffusion-weighting. Here, we present a scheme to address this problem by transforming noisy Rician signals to noisy Gaussian signals. Real experimental data are used to illustrate the proposed framework and demonstrate its effectiveness in removing the noise-induced bias.

15:00         4693.     Characterization of Image Heterogeneity Using Minkowski Functionals

Holly C. Canuto^1,2, Mikko I. Kettunen^1,2, Charles I. McLachlan³, Anant S. Krishnan¹, Andre' A. Neves^1,2, De-en Hu¹, Michael P. Hobson^3,4, Kevin M. Brindle^1,2

¹Biochemistry, The University of Cambridge, Cambridge, Cambridgeshire, UK; ²Cancer Research UK, Cambridge, Cambridgeshire, UK; ³Metropolis Data Consultants, Cambridge, Cambridgeshire, UK; ⁴Astrophysics, Cavendish Laboratory, The University of Cambridge, Cambridge, Cambridgeshire, UK

Tumor cell death following treatment is a good prognostic indicator for treatment outcome. The spatially heterogeneous nature of cell death has recently been exploited, using the morphological descriptors Minkowski Functionals, to enhance the sensitivity of detection of a MR contrast agent targeted to bind to apoptotic cells. Since tissue morphology is a very sensitive indicator of underlying tissue biology we have used 2D Minkowski Functionals to parameterize the morphological heterogeneity present in T₂-weighted MR images of tumors before and after drug-treatment. We have shown that we can detect a treatment response in the absence of a contrast agent.

14:00         4694.     Landmark-Based Prostate MR Image Matching Using Incompressible Large Deformation Diffeomorphism

Xiaofeng Liu¹, Steven Roys², Jerry L. Prince^1,3, Rao Gullapalli²

¹Computer Science, Johns Hopkins University, Baltimore, MD, USA; ²Radiology, University of Maryland Baltimore, Baltimore, MD, USA; ³Electrical and Computer Engineering, Johns Hopkins University, Baltimore, MD, USA

Magnetic resonance (MR) imaging and spectroscopy with an endorectal coil is a standard clinical procedure for prostate cancer diagnosis. However images and spectra are obtained when the prostate is in a distorted state due to the use of the endorectal coil. For optimal delivery of radiation dose to the cancerous tissue during therapy, these deformed images need to be transformed back to their non-deformed state. We propose an incompressible large deformation diffeomorphism method for landmark-based prostate image matching, which preserves the incompressibility, an important physical property of the prostate. Preliminary results on 2D images demonstrate the effectiveness of this method.

14:30         4695.     Non-Rigid Registration of T1 MR Images Constrained with Fractional Anisotropic (FA) Maps

Xia Li¹, Benoit Dawant², Xin Hong¹, Zhaohua Ding¹, John Gore¹, Adam Anderson¹

¹Institute of Imaging Science, Vanderbilt University, Nashville, TN, USA; ²Electrical Engineering and Computer Science, Vanderbilt University, Nashville, TN, USA

Conventional T1 or T2 weighted MR images provide basic contrast information for brains. However, those images are limited because they provide little contrast between distinct white matter structures. Diffusion tensor imaging (DTI) is a complementary tool that provides information on the orientation and integrity of white matter fiber bundles. The co-registration of intra- or inter-subject DTI data in a standard coordinate space allows us to perform statistical analysis of DTI images, or compare data across subjects. In this study, a non-rigid registration algorithm, which combines both T1 weighted MR images and DTI information, is proposed, in order to retain both MR information and the significant micro-structural information about tissues.

15:00         4696.     Metrics for Quantifying the Quality of MR Images

Francisco Prieto^1,2, Marcelo Guarini^1,2, Cristian Tejos^1,2, Pablo Irarrazaval^1,2

¹Departamento de Ingeniería Eléctrica, Pontificia Universidad Católica de Chile, Santiago, Chile; ²Biomedical Imaging Center, Pontificia Universidad Católica de Chile, Santiago, Chile

Evaluating the quality of reconstruction techniques in MRI is a difficult task. The Root Mean Square Error (RMSE) is commonly used to compare reconstructed images, although it does not correlate well with subjective perception of quality. We present two new metrics which incorporate human perception factors. One is based on the null space of the transformation from one image to the compared one, and the other on the variation of the Just Noticeable Differences map for different contrasts. Visual experiments show that both indexes provide more information than the RMSE and we propose them as metrics for comparing images.

15:30         4697.     Improved Visualization of Iron-Rich Structures in the Brain with 3D Multiecho GRE Imaging

Yunhong Shu¹, Matt A. Bernstein¹, Marsha Swatosh¹, John Huston¹, Kiaran P. McGee¹, Kendall H. Lee²

¹Radiology, Mayo Clinic, Rochester, MN, USA; ²Neurologic Surgery, Mayo Clinic, Rochester, MN, USA

Reliable localization of subthalamic nucleus (STN) and small veins is a critical step in the surgical planning of deep brain stimulation (DBS) for Parkinson’s disease. 3D MR imaging has been used to guide placement of the lead tips to the STN. A multiple-echo gradient-echo (GRE) pulse sequence can be used to acquire a set of T2*-weighted imaging volumes. There remain, however, multiple options to post-processing the 3D image sets to enhance the visualization of the STN and the small veins. The purpose of this study is to evaluate the performances of several of those post-processing methods to find an optimal technique.

13:30         4698.     Use of Multi-Spectral MR Data to Generate an Attenuation Map for Application to PET/MR Hybrid Imaging

Harry Robert Marshall^1,2, Robert Z. Stodilka^1,2, Benoit Lewden¹, Jean Theberge^1,2, Eric Sabondjian^1,2, Alexandre Legros¹, Andrea Mitchell¹, Lela Dorrington¹, Jane Sykes¹, Frank S. Prato¹

¹Imaging, Lawson Health Research Institute, London, ON, Canada; ²Medical Biophysics, The University of Western Ontario, London, ON, Canada

Attenuation correction is necessary to generate quantitatively accurate PET images but it is not straightforward in PET/MR hybrid imaging. One method to generate an attenuation map under such circumstances is to segment the MR images into tissue types and to assign said tissue types their known attenuation coefficients. The aim of our study was to determine whether using multispectral MR is beneficial to this end, and if so, which combination of pulse sequences is most effective. We confirmed that multispectral MR yields superior attenuation maps than single MR data sets, but the best combination of sequences depends upon anatomical position.

14:00         4699.     Navigator Filtering Using Principal Component Analysis.

Sergio Andres Uribe Arancibia¹, Philipp Batchelor¹, Claudia Prieto¹, Reza Razavi¹, Tobias Schaeffter¹

¹Division of Imaging Sciences, King's College London, London, UK

Cardiac MR usually requires either navigator echoes or k0 profiles to monitor diaphragm position. Unwanted signals that interfere navigator projections may lead to an incorrect respiratory motion. In this abstract we propose a novel method for filtering navigator projections using a singular value decomposition. The method cleans the projections rather than a derived respiratory signal. Results showed unwanted noise was removed from the projections leading to a robust respiratory motion detection.

14:30         4700.     Augmenting Surgery Planning for Neoadjuvant Chemotherapy Patients by 3D Transformation of Prone Breast MR Images to the Supine Images in the Operating Room Setting

Ruparani Chittineni¹, Jeon-Hor Chen^1,2, Orhan Nalcioglu¹, Min-Ying Su¹

¹Tu & Yuen Center for Functional Onco-Imaging, Irvine, CA, USA; ²Department of Radiology, China Medical University, Taichung, Taiwan

15:00         4701.     Implicit Reference-Based Group Registration in Structural and Functional MRI Studies

Xiujuan Geng¹, Hong Gu¹, Thomas J. Ross¹, Yihong Yang¹

¹Neuroimaging Research Branch, National Institute on Drug Abuse, National Institutes of Health, Baltimore, MD, USA

We propose an implicit reference-based group (IRG) registration method with a high-dimensional elastic deformation model to align structural or functional images within groups. The performance of the IRG method was evaluated using T1-weighted MR images with pre-defined ROIs and compared with reference-based registration methods. Results show that IRG registration produces better registration performance in terms of higher relative overlaps. IRG registration was also applied to an fMRI study with a motor and visual task. The statistical detection power is improved by the elastic registration compared to the affine approach and IRG registration further increases the power compared to reference-based methods.

13:30         4702.     Effects of Patient Population Specific Atlases on Automatic Segmentation of Subcortical Structures in Freesurfer

Istvan Csapo¹, Jared Price¹, Troy Russell¹, Jeff Dewey¹, Ekta Sem¹, Daniel McCaffrey¹, Charles R. Guttmann¹, Bradford Navia², David F. Tate¹

¹Radiology, Brigham and Women's Hospital, Boston, MA, USA; ²Tufts New England Medical Center, Boston, MA, USA

We set out to investigate whether using manually labeled training data from a specific patient cohort to build probabilistic atlases for Freesurfer improves the automatic segmentation of novel images from the same cohort compared to the segmentations using the default atlas. 106 brain images were segmented and manually corrected. 64 of these images were used to build 6 different atlases. The other 42 images were segmented with Freesurfer using the new atlases. The results were compared to the manual segmentations. Out of the 8 subcortical structures investigated, the hippocampus segmentations were significantly improved with the new population specific atlases.

14:00         4703.     A Simple Semi-Automated Method for Caudate Volume and Shape Analysis

Karlene M. Fraser¹, Mitul A. Mehta², Ruth L. O'Gorman^1,3

¹Neuroimaging, Maudsley Hospital, London, UK; ²Centre for Neuroimaging Sciences, Institute of Psychiatry, London, UK; ³Neuroradiology, King's College Hospital, London, UK

Here we present a simple, semi-automated method for evaluating the volume and shape of the caudate nucleus. The shape is quantified by the surface area to volume (SA/V) ratio, calculated from contours defining the edge of the caudate on a segmented grey matter map. This method was applied to images from 17 healthy adults and the accuracy and inter-rater reproducibility was assessed. The volume measures showed good agreement with literature values. Significant laterality differences emerged, with the right caudate demonstrating higher volume and surface area and lower SA/V relative to the left (p<0.005). The method also showed good reproducibility, (ICC=0.83-0.9).

14:30         4704.     Automatic Segmentation of Deep Grey Matter Structures for the Assessment of DTI Images

Emil Malucelli¹, David Neil Manners¹, Claudia Testa¹, Caterina Tonon¹, Giovanni Rizzo¹, Valerio Carelli², Giuseppe Nicoletti³, Aldo Quattrone³, Bruno Barbiroli¹, Raffaele Lodi¹

¹Dipartimento di Medicina Interna dell’Invecchiamento e delle Malattie Nefrologiche, University of Bologna, Bologna, Italy; ²Dipartimento di Scienze Neurologiche , University of Bologna, Bologna, Italy; ³Institute of Neurological Sciences, National Research Council, Cosenza, Italy

Time-consuming manual selection of regions of interest is currently one limiting factor in the clinical use of DTI. Methods exist to automatically identify deep brain structures using high resolution images. We evaluated the possibility of automatically segmenting structures of deep grey matter in diffusion tensor maps using high resolution T1-weighted images. We compared manual and automatic segmentation in ten controls. Results were indistinguishable for quantification of mean diffusivity in thalamus, pallidus and putamen, but not caudate. Using automatic segmentation, three patients with well-characterized neurological disorders showed differences compared to controls, in line with those expected from the known pathologies.

15:00         4705.     Reliability and Validity of MRI-Based Automated Volumetry Software Relative to Manual Measurement of Subcortical Structures in HIV-Infected Patients from a Multisite Prospective Study

Jeffrey Dewey¹, George Han², Troy Russell¹, Jared Price¹, Daniel McCaffrey¹, Ekta Sem¹, Joy Chinazom Anyanwu¹, Charles Guttmann¹, Bradford Navia³, Ronald Cohen², David Tate¹

¹Center for Neurological Imaging, Brigham and Women's Hospital, Boston, MA, USA; ²Warren Alpert School of Medicine at Brown University, Providence, RI, USA; ³Tufts New England Medical Center, Boston, MA, USA

While software packages designed to automatically identify subcortical structures can greatly increase the efficiency of volumetric studies, notable differences in the performance of these tools have previously been reported. The purpose of this study was to compare the accuracy and consistency of volumetric data for several clinically relevant subcortical structures obtained using three methods: manual tracing, Freesurfer, and SPM. Freesurfer output was more accurate and consistent relative to manual tracings than that of SPM in the majority of structures. However, the inaccuracy and variability of both tools suggest that researchers supplement their use with visual inspection to ensure reliable data.

13:30         4706.     Calculation of Optic Nerve Dimensions Using Gaussian Fitting of MR Images

Li Sze Chow¹, Lauren Wallis¹, Nigel Hoggard¹, Simon J. Hickman², Martyn N.J. Paley¹

¹Academic Unit of Radiology, University of Sheffield, Sheffield, UK; ²Department of Neurology, The Sheffield Royal Hallamshire Hospital, Sheffield, UK

The aim of this study was to develop a program to accurately calculate the optic nerve diameter, area and volume using Gaussian fitting of the acquired MR images. The program will be used in a new study of optic neuritis on 1.5T and 3.0T MR systems with both T1- and T2-weighted 3D Spectral Presaturation Inversion Recovery (SPIR) sequences and much shorter scan times. This study has also shown that T2W images are suitable for measuring the actual optic nerve’s dimension; whereas T1W images can be used to measure the optic nerve and the sheath and CSF surrounding the nerve.

14:00         4707.     Multilevel Segmentation and Classication of MS Lesions

Alvina Goh¹, Christophe Lenglet², Mariappan Nadar³

¹Biomedical Engineering, Johns Hopkins University, Baltimore, MD, USA; ²Center for Magnetic Resonance Research & Department of Electrical and Computer Engineering, University of Minnesota, Minneapolis, MN, USA; ³Siemens Corporate Research, Princeton, NJ, USA

This paper focuses on the segmentation and classification of multiple sclerosis lesions in magnetic resonance images. As MRI is the primary tool used in the diagnosis of multiple sclerosis, there is substantial interest in developing an algorithm that will detect lesions from such images. We present a flexible framework in which segmentation and classification are integrated. We assume that we are given a training set of MRI images which contains manually labeled regions of MS lesions. The algorithm we used combines two effective techniques from the computer vision literature: graph-based bottom-up methods and top-down generative models.

14:30         4708.     Computer-Aided GBM Tumor ROI Contouring and Mapping Framework for Structural and Physiological MR Images in a Multi-Center Clinical Trial

Jing Huo¹, Matthew S. Brown¹, Jeffry R. Alger^2,3, Whitney B. Pope³, Hyun J. Kim³, Kazunori Okada⁴, Jonathan G. Goldin³

¹Department of Radiological Sciences , UCLA, Los Angeles, CA, USA; ²Department of Neurology, UCLA, Los Angeles, CA, USA; ³Department of Radiological Sciences, UCLA, Los Angeles, CA, USA; ⁴Computer Science Department, San Francisco State University, San Francisco, CA, USA

The main contribution is to develop a computer-aided 3D multi-modal tool for clinical trials with GBM brain tumor. First we developed a 3D semi-automated segmentation tool to contour the tumor on T1-weighted contrast-enhanced volumetric images and then a 3D tumor ROI mapping tool to map the tumor contours from T1w images onto all the physiological images including dynamic contrast enhance (DCE) images, dynamic susceptibility contrast-enhanced (DSC) and diffusion-weighted (DW) images. We applied the tool to about 650 scans and did the quantitative evaluation with a sub-group of 31 patients. 20 out of 31(64.5%) have over 90% overlap ratio.

15:00         4709.     Model Error Maps as Robust Features for T1/T2-Weighted Data Segmentation

Umesh Suryanarayana Rudrapatna¹, Annette van der Toorn¹, Ivo A. Tiebosch¹, Josien P.W Pluim¹, Rick M. Dijkhuizen¹

¹University Medical Center Utrecht, Utrecht, Netherlands

T1- and T2-weighted data segmentation has hitherto relied upon parametric maps and intensities as primary features. The parametric maps contain the necessary information explicitly, but become unreliable at low SNR. To overcome this hurdle, building on estimation theory results, we propose using functions based on fitting-errors when specific relaxivities are assumed. This approach leads us to a simpler linear estimation problem and provides for incorporation of prior knowledge about relaxivities, the reliability of which does not critically affect the outcome. Feature selection filters, the results of which generalize to a broad class of supervised and unsupervised learning scenarios testify the merits of this strategy.

14:00         4710.     Motion Correction for Perfusion Weighted Images in Stroke: Approach and Impact on Quantification

Rakesh Mullick¹, Dattesh Shanbhag¹, Uday Patil¹, Xiaodong Tao²

¹Imaging Technologies Lab, GE Research, Bangalore, Karnataka, India; ²Imaging Technologies Lab, GE Research, Niskayuna, NY, USA

Head motion while imaging stroke, neoplasic and neuro-degenerative disease is often observed during routine clinical imaging. Even with head stabilization, involuntary head movement can cause image artifacts leading to incorrect diagnosis. These artifacts have acute effects in select (extended duration) image sequences targeted to acquire spatio-temporal data like perfusion-weighted imaging and functional imaging scans. It is therefore imperative to address motion artifacts either early in the imaging chain or through a post-processing step. In cases of acute stroke, where time to image is of utmost importance, the extended duration of scan due to use of navigators precludes their use and places more importance on retrospective motion correction through image segmentation/registration techniques. In this work we present an approach based on image registration to align the spatio-temporal data, evaluate the impact on quantification of stroke related perfusion maps. The underlying challenge to correct and evaluate motion artifacts is further discussed.

14:30         4711.     Model-Based Automatic Detection of the Anterior and Posterior Commissures on MRI Scans

Babak A. Ardekani^1,2, Alvin H. Bachman^1,3

¹Center for Advanced Brain Imaging, The Nathan S. Kline Institute for Psychiatric Research, Orangeburg, NY, USA; ²Department of Psychiatry, New York University School of Medicine, New York, USA; ³Emeritus Professor, City College of CUNY, New York, USA

A model-based fully automatic algorithm is presented for detection of the anterior and posterior commissures (AC/PC) on 3D MRI scans. The algorithm has several advantages over previously published methods for solving this problem. It is fast, accurate, and robust in the presence of image artifacts. It does not rely on any specific image contrast or location of the corpus callosum. It is flexible in its definition of these landmarks (e.g., center versus edge of the AC/PC cross-section with the mid-sagittal plane). It relies on a relatively small number of parameters whose default values rarely need to be changed.

15:00         4712.     MARM: Multiscale Adaptive Regression Models for Neuroimaging Data

Yimei Li¹, Hongtu Zhu², Dinggang Shen³, Songyuan Tang⁴, Weili Lin³

¹Biostatistics, UNC-CH, USA; ²Biostatistics and Biomedical Research Imaging Center, UNC-CH, Chapel Hill, NC, USA; ³Radiology and BRIC, UNC-CH, USA; ⁴UNC-CH,, Radiology and BRIC, USA

We develop a multiscale adaptive regression model for spatial

15:30         4713.     Statistic Model of Respiratory Motion by Using Dynamic MRI

Gang Gao¹, Jamie McClelland¹, Dave Hawkes¹

¹Centre for Medical Image Computing, University College London, London, England, UK

The accurate delivery of medical treatments to internal organs (e.g. MR guided intervention) that are subject to considerable respiratory motion has proved to be difficult. A motion model that can predict the internal motion from respiratory surrogate signals (e.g. the displacement of points on the skin surface) can help improve this situation. we are presenting a method of building respiratory motion models using the statistical technique called CCA and dynamic MRI. In this abstract, we have demonstrated that the MR-based CCA models can accurately predict the internal organ motion from respiratory surrogates measured from the skin surface. Additionally, we have compared the CCA model with a previously-reported PCA-based modelling technique by using the same MR data. The experimental results show the CCA model is superior to the PCA model in 3 out of 4 cases and appears to be more stable than the PCA model.

13:30         4714.     Automatic Macro-Vessel-Minimization Using Independent Component Analysis and Local Scaling Functions

Gernot Reishofer¹, Stephen Keeling², Robert Merwa³, Christian Enzinger⁴, Stefan Ropele⁴, Rudolf Stollberger³, Franz Ebner⁵

¹Department of Radiology / MR-Physics, Medical University of Graz, Graz, Austria; ²Institute for Mathematics and Scientific Computing, University of Graz, Graz, Austria; ³Institute of Medical Engineering, Graz University of Technology , Graz, Austria; ⁴Department of Neurology, Medical University of Graz, Graz, Austria; ⁵Department of Radiology / Division of Neuroradiology, Medical University of Graz, Graz, Austria

Independent component analysis has the potential to separate the signal assigned to macro vessels from tissue signal in DCE-MRI data. This enables an algorithm for minimizing macro-vessel signal in perfusion imaging to avoid overestimation of hemodynamic parameters. We set out to eliminate user dependency by automating the selection of appropriate ICs using a local scaling function technique. Furthermore we preserve the static tissue signal of the macro vessel representing IC by editing the time dependent mixing matrix in the ICA model before the back transformation is performed. The implemented algorithm is fast and stable and therefore applicable for clinical use.

14:00         4715.     Hemodynamic Characterization of the Trans-Thoracic Circulation by Contrast-Enhanced Magnetic Resonance Imaging

Massimo Mischi¹, Harrie CM van den Bosch², Jacques A. den Boer, Jan Verwoerd³, Rene JE Grouls², Hendrikus HM Korsten²

¹Eindhoven University of Technology, Eindhoven, Netherlands; ²Catharina Hospital; ³Philips Healthcare

A minimally-invasive characterization of the trans-thoracic circulation (TTC) by contrast magnetic resonance imaging (MRI) is presented and validated in vitro. The clinical feasibility of the method is also tested with five healthy volunteers. A bolus of a paramagnetic agent is intravenously injected and detected by MRI in the right and left ventricles. A parametric deconvolution is designed for the estimation and hemodynamic interpretation of the TTC dilution impulse response. This permits in the assessment of the intra-thoracic blood volume as well as the ratio between contrast convection and dispersion in the TTC. The results are accurate and motivate further investigation.

14:30         4716.     Volume Tracking - A Novel Method for Visualization and Quantification of Cardiac Blood Flow

Johannes Töger¹, Håkan Arheden¹, Gustaf Söderlind², Einar Heiberg¹

¹Clinical Physiology, Lund, Skåne, Sweden; ²Mathematics, Numerical Analysis, Lund, Skåne, Sweden

The detailed interactions between blood, myocardium, valves and vessels are still incompletely understood. Although three-dimensional, time-resolved Phase Contrast MRI promises greater insights into cardiac blood flow, no single solution to the visualization problem exists. In this study Volume Tracking, a novel method for visualization and quantification of cardiac blood flow is developed and validated.

15:00         4717.     Estimation of Left Ventricle Parameters from Cine MR Data: Fully Automatic Versus Manual Analysis.

Thomas Ki-Yong Broussaud¹, Alain Nchimi¹

¹Medical Imaging Dept., CHC Saint-Joseph, Liège, Belgium

We present a study consisting in comparing the performances of automatic versus manual segmentation for estimating functional parameters of the left ventricle. The automatic analysis tested here consists in global detection of the left ventricle followed by local deformation of an active contours. Our study shows limitations of automatic analysis and supports quantitative evaluation of automatic analysis approaches.

13:30         4718.     In Vivo Cardiac NMR Diffusion Weighted Imaging(DWI) for the Human Heart: Tackling Motion Issue with Temporal Maximum Intensity Projection(TMIP)-DWI and First Results in Humans

Stanislas Rapacchi¹, Pierre Croisille², Magalie Viallon³, Denis Grenier¹, Peter Kellman⁴, Han Wen⁴

¹Creatis-LRMN, Université Lyon 1, INSA Lyon, Lyon, France; ²Hopital Cardiologique et Pneumologique L. Pradel, Lyon, France; ³Hopital Cantonal Universitaire de Genève, Genève, Switzerland; ⁴National Heart, Lung and Blood Institute, NIH, Bethesda, MD, USA

Diffusion weighted imaging in the heart is greatly affected by contractile motion and remains challenging to date. Stimulated-echo approaches have relatively low signal level and require complete beat-to-beat positional repeatability which may not be met in patients. With spin-echo techniques, Gamper et al. recently proposed a robust motion-compensated sequence. However, through-slice diffusion was still difficult to measure. From their work, we developed an alternative approach for DWI, where a number of single-shot images of the same diffusion weighting are acquired at a series of different time points in diastole, and these are projected along the temporal axis by maximum-intensity-projection (tMIP) to form a true DWI. Through quantifying the motion-induced signal loss in DWI of a single slice based on experimental myocardial strain, we show the validity of tMIP and preliminary results in free-breathing scans of volunteers.

14:00         4719.     Adaptive Trigger Delay Using a Predictive Model Applied to Black Blood Fast Spin Echo Cardiac Imaging in Systole

Brice Fernandez^1,2, Julien Oster^2,3, Maelene Lohezic^1,2, Damien Mandry^2,4, Olivier Pietquin^2,5, Pierre-André Vuissoz^2,3, Jacques Felblinger^2,3

¹Global Applied Science Lab., GE Healthcare, Nancy, France; ²IADI, Nancy-Université, Nancy, France; ³U947, INSERM, Nancy, France; ⁴CHU de Nancy, Nancy, France; ⁵IMS Research Group, Supelec Metz Campus, Metz, France

Cardiac-gated Fast Spin Echo (FSE) sequences with black blood preparation provide detailed anatomical information to clinical cardiac MR applications. However, the black blood preparation time is too long to allow this acquisition scheme to run in systole, hence black blood FSE are usually performed in diastole. Overcoming this limitation requires to predict the R-R interval to launch the preparation properly and to handle instantaneous heart rate variability. This approach has been implemented and tested on four healthy volunteers, and compared with fixed trigger delay (TD). Results demonstrate accurate depiction of the myocardium at end-systolic phase in black blood.

14:30         4720.     Doubled Scan Efficiency in Multi-Slice Real-Time Cine TrueFISP Cardiac Imaging by Pseudo-Continuous Sliding Slice Acquisition

Andreas Greiser¹, Sven Zuehlsdorff², Edgar Mueller¹

¹Siemens AG Healthcare Sector, Erlangen, Bavaria, Germany; ²Siemens Medical Solutions, Chicago, IL, USA

Real-time cine TrueFisp cardiac imaging offers multiple slice acquisitions within a single breathhold. Typically, the steady state preparation is established utilizing a full heartbeat per slice. Hence, only every other heartbeat is used to collect data. A simple elimination of the preparation heartbeats reduces the contrast of blood and myocardium in the first few phases. This work presents a novel method that shifts the slice position between consecutive phases by a distance small compared to slice thickness. High contrast is maintained throughout all cardiac phases providing whole heart coverage within a single breathhold.

15:00         4721.     3D Dynamic Contrast-Enhanced Using Undersampled Golden-Radial Phase Encoding

Claudia Prieto¹, Sergio Uribe¹, Reza Razavi¹, David Atkinson², Philipp Beerbaum¹, Tobias Schaeffter¹

¹Division of Imaging Sciences, King's College London, London, UK; ²Centre for Medical Imaging Computing, University College of London, London, UK

A new acquisition and reconstruction strategy for 3D isotropic DCE-MRI has been proposed. The proposed method is based in a new undersampled dynamic acquisition which combines a 3D radial phase encoding trajectory with the golden angle profile order, providing explicit regularization for iterative reconstructions. The main advantages of this approach are: (i) it allows retrospective reconstructions with different temporal resolutions, (ii) it does not require bolus track previous scan and (iii) it reduces strike artefacts and noise in comparison with the non-explicit regularized reconstructions. The method was tested, using a 32-channel coil and iterative SENSE reconstructions, in phantoms and patients.

13:30         4722.     Time-Resolved Contrast-Enhanced Radial Ghost MRA

Hyun Jong Jeong¹, Ioannis Koktzoglou², Christopher S. Eddleman³, John Sheehan⁴, James Carr⁴, Timothy John Carroll^1,4, Robert Edelman⁵

¹Biomedical Engineering, Northwestern University, Chicago, IL, USA; ²Radiology, Northshore University Hospital, Evanston, IL, USA; ³Neurological Surgery, Northwestern University, Chicago, IL, USA; ⁴Radiology, Northwestern University, Chicago, IL, USA; ⁵Radiology, North Shore University Hospital, Evanston, IL, USA

A novel time-resolved contrast-enhanced MR angiography technique is introduced. The technique does not require subtraction and uses ghost images to separate the vessels from background tissues. The SNR loss resulting from subtraction of pre-contrast images from the post-contrast frames is prevented. Based on inducing ghost artifacts arising from inconsistency in k-space data, this technique is combined with radial sliding window reconstruction for time-resolved imaging.

14:00         4723.     Time-Resolved MRA Using Radial Multi-Echo Sequence and Sliding Window Reconstruction

Hyun Jong Jeong¹, Saurabh Shah², Christopher S. Eddleman³, John Sheehan⁴, James Carr⁴, Timothy John Carroll^1,4

¹Biomedical Engineering, Northwestern University, Chicago, IL, USA; ²Siemens Medical Solutions, Chicago, IL; ³Neurological Surgery, Northwestern University, Chicago, IL, USA; ⁴Radiology, Northwestern University, Chicago, IL, USA

We introduce a technique to decrease the actual acquisition time to decrease the temporal footprint and achieve true acceleration for time-resolved MRA by taking multiple echoes per TR, acquiring a full 3D volume in 7 seconds without parallel imaging. Combined with radial sliding window reconstruction, we obtained time resolved MRA with frame rates comparable to clinical X-Ray DSA, resulting in complete separation of arterial and venous phases. The technique has been verified in flow phantoms as well as volunteers and patients in vivo.

14:30         4724.     Fast Imaging Sequence for Free-Breathing Planning in Cardiac Imaging

Tamer A. Basha¹, Monda L. Shehata², Nael F. Osman^1,2

¹ECE Dept, Johns Hopkins University, Baltimore, MD, USA; ²Dept of Radiology, Johns Hopkins University, Baltimore, MD, USA

In this work, we propose a fast black-blood pulse sequence which requires scan duration as short as one heartbeat. This can be used for free-breathing cardiac imaging planning, which helps to reduce the overall scan time and the number of breath-holds during the exam.

15:00         4725.     T2 Weighted Preparation with Selective Response: Contrast Mechanism for Dark Blood Imaging and Arterial-Venous Discrimination

Peter Kellman¹, Anthony Z. Faranesh¹, J. Andrew Derbyshire¹, Logi Vidarsson²

¹National Institutes of Health, Bethesda, MD, USA; ²The Hospital for Sick Children, Toronto, Canada

T2-selective imaging may be used to provide contrast between tissues with differing T2 values and to provide suppression of long T2 species. Conventional T2 mapping or linear combination filtering requires multiple T2 weighted acquisitions. It is possible to design a selective T2-weighted RF preparation which has the benefit of a single acquisition. Selective T2-preparations have been demonstrated for imaging the meniscus of the knee with short T2. Initial results for cardio-vascular application are presented for a new design approach based on adiabatic pulses and phase sensitive reconstruction. This method has potential applications to oximetry and other tissue classification.

14:00         4726.     Comparison of Large Phased Arrays Using Stacked Segments

Wolfgang Driesel¹, André Pampel¹, Toralf Mildner¹, Stefan Hetzer¹, Harald E. Möller¹

¹Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany

Medium-sized stacked combinations of loop coils with a microstrip transmission-line element (MTL) as basic elements for large coil arrays are compared in extensive computer simulations. The calculations indicate a good SNR performance and a potential for imaging with acceleration in all three directions.

14:30         4727.     Multi Arm Archimedean Coil for High Field MRI

Sofia Sanchez¹, Silvia Sandra Hidalgo¹, Sergio Enrique Solis¹, Mario Rosas¹, Alfredo Odon Rodriguez¹

¹CI3M, UAM Iztapalapa, Mexico, City, Mexico

Spiral coils have been used for various MRI applications as single-channel coils and array coils. Archimedes introduced a special type of spiral consisting of two arms or more. The Archimedean spiral geometry has not been studied for MRI applications according to the literature reported. Computer simulations of the magnetic field for different Archimedean spiral coil configurations were computed using a Finite Element Method. For comparison, the magnetic field of circular-shaped coil with similar dimensions was also performed. Archimedean coils showed higher magnetic field intensity than the popular circular coil at 300 MHz.

15:00         4728.     Noise Figure Limits for Circular Loop MR Coils

Ananda Kumar^1,2, William Alan Edelstein¹, Paul Arthur Bottomley^1,2

¹Radiology, Johns Hopkins University, Baltimore, MD, USA; ²Electrical and Computer Engineering, Johns Hopkins University, Baltimore, MD, USA

Loop coils are fundamental building blocks of phased arrays. For phased arrays using large numbers of loops, loop size is critical in determining detection efficiency. As loop size decreases, coil noise becomes dominant and adversely affects realizable SNR by increasing system noise figure. There are consequent limits to SNR gains as array element number increases and element size shrinks. 1 dB noise figure contributions are reached for coils with radii 23 mm at 1.5T, 16 mm at 3T and 10 mm at 7T. Constructing arrays out of smaller coils than these at such fields is therefore of questionable value.

15:30         4729.     A High Temperature Superconducting (HTS) Coil for Imaging of Human Extremities

Hoon-Sin Cheong¹, Ivo Volkov², Neil Alford³, Chris Randell⁴, Jim Wild¹, Martyn Paley¹

¹Academic Radiology, University of Sheffield, Sheffield, South Yorkshire, UK; ²Materials, London South Bank University, London, Middlesex, UK; ³Materials, Imperial College London, London, Middlesex, UK; ⁴MRI, Pulseteq Limited, Gloucester, Gloucestershire, UK

A 10-turn, 70mm YBCO spiral surface coil was designed for clinical imaging of human extremities at 0.2T. Liquid nitrogen was added to cool the coil to 81K during image acquisitions. The phantom imaging test results show the YBCO coil has SNR advantages of 270% and 80% when compared to an equivalent copper coil at room temperature and 81K respectively. The in-vivo human hand images show good details of the bones and tissue structures. This means the YBCO coil was capable of producing useful anatomical images of human extremity even in a low-field environment.

13:30         4730.     Hybrid Decoupling for RF Coils

dashen chu¹, ricardo matias, Scott Lindsay, robert stormont, eddy boskamk, saban kurucay

¹MR, ge healthcare, waukesha, wi, USA

The work presented in this abstract is the study of a new decoupling circuit for RF surface coils for MR imaging. The new decoupling circuit is designed to operate in two modes, one is for normal mode in which the appropriate DC biasing is present and the other one is abnormal mode in which the DC biasing is absent. The study was focused on the comparison between the characteristics of traditional active decoupling circuit and the proposed hybrid circuit. We provided experimental testing results including impedance data and temperature rise measurement due to RF heating.

14:00         4731.     Model-Based MRI Coil Tuning Algorithm

Liang Liu¹, Charles Bouman¹, Thomas Talavage¹, Victor Taracila², YunJeong Stickle², Limin Feng², Pei H Chan², Fraser Robb²

¹Purdue University, West Lafayette, IN, USA; ²GE Healthcare, Aurora, OH, USA

Model-based tuning algorithm for MRI coil array

14:30         4732.     Frequency Division Multiplex for Phased Array Receive Coils

Joseph Iannotti¹, Steven Go¹, Kevin Dufel¹, Bill Platt¹, Eric Fiveland¹

¹General Electric Global Research Center, Niskayuna, NY, USA

A novel way of combining multiple receive channels into a single coax cable reducing bulk and simplifying cable routing constraints while allowing for parallel image acquisition

15:00         4733.     Coil Performance Vs. Preamplifier Location in the MR Phased Array

Randy O. Giaquinto¹, Joseph E. Piel¹

¹GE Global Research Center, Niskayuna, NY, USA

A study is presented comparing the performance of two coil arrays, with preamplifiers located on the coil, and remotely. In our particular setup, it was found that remote preamps may result in a slight increased image SNR, at the expense of increased cable bulkiness and decreased patient ergonomics.

13:30         4734.     A High Throughput 8 Channel Mouse Probehead for 9.4 T

Titus Lanz¹, Matthias Müller¹, Hannah Barnes², Jürgen E. Schneider²

¹RND, Rapid Biomedical, Rimpar, Germany; ²Cardiovascular Medicine, University of Oxford, Oxford, UK

Rodent MRI is well established but still puts substantial demands on combining highly sophisticated experimental setups with careful animal handling. Rodents, in particular disease models, may be very sensitive to the anaesthetic burden, and, therefore, minimal scan time is essential. In this work we present an 8 channel 9.4 T mouse probehead optimized for high throughput providing simple and speedy positioning and connecting up. The total setup time for the probehead including positioning and wiring is reduced to below 1 min. Probehead and animal can be handled completely independently.

14:00         4735.     A 16-Channel Rat-Body Array Coil with an Integrated Birdcage Transmitter at 7T

Martin Tabbert¹, Marco Motz¹, Marcos Lopez², Erhard Pfrommer¹, Sven Junge¹

¹Bruker BioSpin MRI GmbH, Ettlingen, Germany; ²Universität Würzburg, Germany

With this study we present a novel 16 channel array-coil with a transmitter coil for investigations on rats which is optimized for rat abdominal imaging. Decoupling electronics, phase shifters and preamplifiers are included in the coil. The receive-signals are amplified with 16 built-in low-impedance preamplifiers. An insulating network that reduces the coil-coupling by additional 6dB was designed. Compared to a quadrature resonator, the coil shows an SNR improvement of the 16-channel coil in the center of a transversal slice by a factor of 1.1 up to 2.5 regarding the periphery of the VOI. All advantages of parallel imaging techniques on small animals can be utilised, e.g. rats can be examined in shorter times by using accelerated 2D/3D imaging.

14:30         4736.     An Array RF Coil for Rabbit Thoracic Arteries Imaging at 200MHz

Raimo PJ Joensuu¹

¹AstraZeneca, Molndal, Sweden

A 4-channel array coil for rabbit’s thoracic arteries imaging at 4.7T has been constructed. The coil elements were made of a semi-rigid coaxial cable where the outer shield constitutes the receiver loop. In the Tx mode the center conductor is coupled to the circuit which changes the total inductance and detunes the coil. l/4 –cables are used to transmit the signal to the preamplifier and they also to transmit Tx/Rx control current to PIN-diodes. The l/4 –cables are covered with a resistive coating to curb sheet currents and Tx radiation scattering. Very good shielding from external disturbances has been demonstrated.

15:00         4737.     A Four Channel SENSE Microcoil Array with Integrated Physiology Support System for Imaging Four Mouse Brains

Marcelino Bernardo¹, Peter Choyke²

¹Molecular Imaging Program, NCI, SAIC-Frederick, Inc., Bethesda, MD, USA; ²Molecular Imaging Program, National Cancer Institute, Bethesda, MD, USA

A four channel SENSE mouse brain microcoil array with integrated physiology support and monitoring for imaging four mice simultaneously is presented. Phantom and in vivo images are have been acquired which illustrates the utility of the system for screening a large cohort of orthotopic and genetically engineered mouse brain tumor models.

13:30         4738.     A Semi-Flexible 32-Ch RF Coil Array for Clinical 3T Brain Imaging

Yu Li¹, Matthias Gyori¹, KD McClellan¹, Charlie Saylor¹, Arne Reykowski¹

¹Advanced Concept Development, Invivo Diagnostic Imaging, Gainesville, FL, USA

In this study, a semi-flexible 32-channel RF coil array was developed for clinical brain imaging. In this coil, a semi-flexible mechanical structure was used to minimize the gap between the head and coil elements for different patients. The alignment of 32 coil elements was optimized to minimize the coil loss and overall coupling. It was demonstrated that this coil gives lower SENSE g-factor and higher SNR in depth compared with a standard eight-channel clinical RF coil in clinical brain imaging applications.

14:00         4739.     An Improved 28 Channel Coil Array for Optic Nerve Imaging

Robb Merrill¹, J. Rock Hadley¹, Emilee Minalga¹, Dennis Parker¹, Sathya Vijayakumar¹, Seong-Eun Kim Choi¹, Laura Bell¹, John Rose^2,3

¹Dept. of Radiology (UCAIR), Salt Lake City, UT, USA; ²Neurovirology Laboratory VASLCHCS; ³Brain Institute

This work describes an RF coil optimized for imaging the Optic Nerve on a Siemens 3T MRI scanner. The coil was constructed using two fiberglass formers that fit together to maintain close coupling near the eyes for any arbitrary head size. The 28 coil elements were circular, constructed from 14-gauge wire, and arranged in a soccer ball layout to surround the entire head. Phantom studies showed that the coil provided ~35% greater rSNR at the optic chiasm and ~300% near the orbits compared to the 12-channel commercial coil. The improvement in rSNR allows performance of high resolution DTI.

14:30         4740.     Coil System for Optimal MR Mammography and MR-Guided Intervention

Bernadette Marie Kaufman¹, Yoshinori Hamamura¹, Ashok Menon¹, Jovan Jevtic²

¹MR, Invivo, Pewaukee, WI, USA; ²Milwaukee School of Engineering, Milwaukee, WI, USA

MR Mammography is a rapidly growing area of diagnostic imaging. It has been found useful for screening, diagnosis, staging, therapy-monitoring and pre-surgical planning. With the advent of high-channel count MR systems and parallel imaging techniques, RF technology must keep pace. Based on the different needs for excellent diagnostic MR imaging and interventional capability, a coil system was developed to meet both sets of requirements. A diagnostic coil was developed to exploit RF technology to achieve high signal and parallel imaging capability. An interventional coil was paired with it to achieve high SNR imaging and maximum interventional access.

15:00         4741.     Simplified RF Modeling of Coil Arrays Composed of Cylindrically Arranged Modules

Andre Pampel¹, Wolfgang Driesel¹, Harald E. Moeller¹

¹Nuclear Magnetic Resonance Unit, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany

A simplified coil-modeling method for phased-array coils composed of cylindrically arranged elements was investigated. Provided a good decoupling between the coil elements is achieved, it is shown to be sufficient to calculate the electromagnetic fields of only one stacked element by full-wave electromagnetic simulation. Considering the boundary conditions describing the whole coil in these calculations, the transmit properties of the remaining elements can be obtained using matrix rotations thus considerably saving computation time and data storage space during the coil optimization process.

14:00         4742.     SNR Benefits of Surface Coil Lift-Off at High Magnetic Field Strength

Qi Duan¹, Graham Wiggins¹, Bei Zhang¹, Riccardo Lattanzi¹, Bernd Stoeckel², Daniel K. Sodickson¹

¹Radiology, Center for Biomedical Imaging, NYU School of Medicine, New York, NY, USA; ²Siemens Medical Solutions USA Inc., New York, NY, USA

This work demonstrates that SNR gains deep in an object can be achieved with a single loop coil by lifting the coil off the object surface. Experimental SNR measurements on a 3T Siemens Tim system were compared with full-wave electrodynamic simulations incorporating a realistic noise model. Close agreement between experiment and simulation was observed. Lift-off-related SNR benefits had a notably asymmetric distribution, with the greatest gains occurring on the far side of the object, opposite to the position of the coil.

14:30         4743.     A Novel Metamaterial Transmit/receive Coil Element for 7 T MRI – Design and Numerical Results

Jochen Mosig¹, Achim Bahr¹, Thomas Bolz¹, Mark E. Ladd²

¹RF&Dosimetry, IMST GmbH, Kamp-Lintfort, Nordrhein-Westfalen, Germany; ²Erwin L. Hahn Institute for Magnetic Resonance Imaging, Essen, Nordrhein-Westfalen, Germany

In this work numerical results for the novel usage of a metamaterial transmission line resonator as a transmit/receive coil are presented. On the basis of resonant antennas a coil element is designed for 7 Tesla. The metamaterial element consists of 4 unit cells, which are optimized for a series mode zeroth-order resonance. This resonance induces a constant current distribution along the element, resulting in a homogenous magnetic field in a larger volume compared to a conventional microstrip transmission line element. The metamaterial element can be easily extended by adding additional unit cells.

15:00         4744.     Improving SNR and RF by Adding a Stripline to Endorectal Coil for the 7T

Catalina S. Arteaga¹, Bob van den Bergen¹, Peter L. Luijten¹, Uulke A. van der Heide¹, Dennis W.J. Klomp¹

¹Radiology and radiotherapy, University Medical Center Utrecht, Utrecht, Netherlands

Theorically adding tranceive/receive elements will increase the SNR and optimize the RF power deposition. We added a stripline into an endorectal loop coil system for the 7T MR. The obtained results proved that indeed the second element increased the SNR and homogeneity of the transmit field. Simulation results for the transmit field in quadrature and anti-quadrature and receive fields for the loop coil and the stripline were obtained showing similar results to the MR measurements. Simulations were also obtained for the calculation of the SAR for only the loop and the combined coils system obtaining a SAR max of 42% less for the combined system than for the loop only.

15:30         4745.     Trap Design Performance for RF Coils for Dual-Frequency MRI & MRS

Arash Dabirzadeh¹, Mary Preston McDougall²

¹Electrical Engineering, Texas A&M University, College Station, TX, USA; ²Biomedical Engineering, Texas A&M University, College Station, TX, USA

The performance of a trap design for multi-nuclear coils is studied, with particular emphasis on its potential application in designing second-nucleus coils that are “insertable” into existing proton coils without modification. Modeling, imaging, and spectroscopy at 4.7T demonstrate the insertability of a P-31 coil into surface, volume, and array proton coils. This work has particular applicability at high fields, where spectroscopy holds great promise and proton coil design becomes more complex.

13:30         4746.     Design of a Patch Antenna for Creating Traveling Waves at 7 Tesla

Bei Zhang¹, Graham Wiggins¹, Qi Duan¹, Daniel K. Sodickson¹

¹Radiology, Center for Biomedical Imaging, NYU School of Medicine, New York, NY, USA

Recently the use of traveling waves for high field MRI has been suggested as a promising method for large FOV imaging and comparatively uniform excitation. In the traveling wave system, conventional volume coils and surface coils used as the transmitter/receiver are replaced by a patch antenna, a kind of directional electromagnetic antenna, placed a considerable distance from the sample. Here, we provide details of our experience in the practical design of a patch antenna, and document its properties for traveling wave generation in a 7 Tesla scanner bore.