Chang-Sheng Mei¹, Cheng-Chieh Cheng^1,2, Lawrence P. Panych¹, and Bruno Madore^1
1Radiology, Harvard Medical School, Brigham and Women’s Hospital, Boston, MA, United States, ²National Taiwan University, Taipei, Taiwan

The most widely-accepted way of detecting tissue damage, with MR-guidance, involves calculating the thermal dose (TD) that was delivered. Such calculations require accurate knowledge of the entire thermal history of every piece of tissue throughout the procedure, which may be difficult to obtain especially when motion is present. We propose an approach capable of performing thermometry and T2 mapping at the same time, with the hope that damage-induced T2 changes will prove a valuable complement to TD calculations toward detecting tissue damage, especially when motion might make thermal maps unreliable.

Adam C. Waspe¹, Charles Mougenot², Samuel Pichardo³, Thomas Looi¹, Joao Amaral⁴, Michael Temple⁴, Hai-Ling Margaret Cheng⁵, and James M. Drake^1
1Division of Neurosurgery, Hospital for Sick Children, Toronto, Ontario, Canada, ²Philips Healthcare Canada, Markham, Ontario, Canada, ³Thunder Bay Regional Research Institute, Thunder Bay, Ontario, Canada, ⁴Department of Medical Imaging, Hospital for Sick Children, Toronto, Ontario, Canada, ⁵Department of Physiology & Experimental Medicine, The Hospital for Sick Children, Toronto, Ontario, Canada

MR-guided focused ultrasound is currently used for palliation of bone metastasis pain. Temperature is monitored using the proton resonance frequency (PRF) shift technique. PRF-thermometry is limited to water-based tissues and temperature is not measured in bone and marrow, limiting its use for primary tumor treatment. We hypothesize that marrow temperature can be quantified by measuring T₁ change during heating, and that the temperature distribution across the bone can be modeled from the surrounding temperatures at the bone-muscle interface and marrow. This abstract introduces a MATLAB-based research tool to simultaneously measure temperature in muscle and fat using a hybrid PRF-T₁technique.

Mengyue He¹, Chao Zou¹, Changjun Tie¹, Wensha Guo¹, Yiu-Cho Chung¹, and Xin Liu^1
1Paul C Lauterbur Research Centre for Biomedical Imaging, Shenzhen Key Laboratory for MRI, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, Guangdong, China

In referenceless MR thermometry, large susceptibility change can affect polynomial model fitting and result in large temperature error. Here, we propose a method to improve the accuracy of referenceless method by excluding the regions with large susceptibility artifact automatically based on the local field map derived from the projection onto dipole fields (PDF) method. Phantom and ex-vivo bovine liver study showed that the method can estimate the baseline image with an error of less than 3oC even when the ROR is corrupted by large susceptibility artifact. The method will be very useful in real time monitoring of temperature changes.

Manivannan Jayapalan^1
1MR Pulse Sequence Applications and System Tools Engineering, GE Healthcare, Bangalore, Karnataka, India

Thermal monitoring in Magnetic Resonance guided Focused Ultrasound (MRgFUS) treatments is a crucial step where the phase images from MR images are used to get thermal maps. One of the widely used techniques is proton resonance frequency (PRF) shift technique that involves some form of image subtraction using a baseline pre-treatment image. Subject motion and tissue deformation due to coagulation can severely distort these techniques. Self-referenced methods help to overcome this hurdle which requires some area of tissue around the area of treatment, hot zone, for polynomial fitting to estimate the baseline phase. The accuracy of the temperature map from this method mainly depends on how close the estimated phase is with baseline phase. In this work, a new technique is described, where the equivalent-baseline phase values are generated/ estimated by applying the current phase values to a model that is based out of Bloch-Siegert phase Shift obtained in run-time along the same location. The phase estimated using Bloch-Siegert phase shift matches well with the baseline phase and so the temperature variations. This method not only eliminates the need for baseline subtraction but also produces better results as the model is generated in runtime against the known values. In clinical scenario, temperature measurement at any location, at any point of treatment could be obtained without the baseline information by using the Bloch-Siegert phase shift from a known RF pulse and the current phase values. This new technique would be amenable for the MRgFUS treatments, in particular moving organs where subtraction methods fail.

Florian Maier¹, David T. A. Fuentes¹, Jeffrey S. Weinberg², John D. Hazle¹, and R. Jason Stafford^1
1Department of Imaging Physics, The University of Texas M. D. Anderson Cancer Center, Houston, TX, United States, ²Department of Neurosurgery, The University of Texas M. D. Anderson Cancer Center, Houston, TX, United States

Referenceless proton resonance frequency shift thermometry algorithms require unwrapped phase maps to estimate temperature changes. Fast, robust, and fully automated unwrapping is mandatory for online monitoring. A novel phase unwrapping algorithm is proposed. Pixels are sorted according to their magnitude values and processed in descending order. Pixel clusters are merged as soon as they are connected. A region of interest for referenceless thermometry was processed in less than 40 ms. The method was successfully tested on data acquired during thermal therapy treatments of patients. In conclusion, the algorithm works robustly in vivo and allows for online temperature monitoring without noticeable delay.

Guanshu Liu^1,2, Qin Qin², Kannie W.Y. Chan^3,4, Jeff W.M. Bulte^3,4, Michael T. McMahon^1,2, Peter C.M. van Zijl^1,2, and Assaf A. Gilad^3,4
1F.M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, United States, ²Department of Radiology, Johns Hopkins University School of Medicine, Baltimore, MD, United States, ³Department of Radiology, Johns Hopkins University, Baltimore, MD, United States, ⁴Cellular Imaging Section, Institute for Cell Engineering, Johns Hopkins University School of Medicine, Baltimore, MD, United States

Monitoring of temperature changes in deep-seated tissues is a necessity during the course of cancer thermotherapy. Among currently available MR temperature mapping methods, water proton resonance frequency (PRF) imaging is the most widely used technology, but it is often hampered by interference from lipid signals. We developed a new approach for assessing the water PRF through direct water saturation to overcome problems with fat resonance interference. Compared to phase mapping, the currently most popular PRF measurement method, the temperature-responsive water saturation shift referencing (T-WASSR) method was demonstrated with an improved performance in fat-containing tissue.

Christoph Birkl¹, Christian Langkammer¹, Johannes Haybaeck², Christina Ernst², Rudolf Stollberger³, Franz Fazekas¹, and Stefan Ropele^1
1Department of Neurology, Medical University of Graz, Graz, Austria, ²Department of Neuropathology, Institute of Pathology, Medical University of Graz, Graz, Austria, ³Institute of Medical Engineering, Graz University of Technology, Graz, Austria

In this work we investigated the temperature dependency of T1 and T2 relaxation times in the human brain. Our results confirmed the predicted temperature dependency of T1 but showed a strong difference between the brain structures. The temperature dependency of T2 is to small to measure because it is masked out by different effects. This work provides for the first time detailed temperature dependency values of the human brain and can be used to improve MR thermometry or to translate findings from postmortem MR to in-vivo data.

Michael Marx¹, Juan C. Plata², and Kim Butts Pauly^3
1Electrical Engineering, Stanford University, Stanford, California, United States, ²Bioengineering, Stanford University, Stanford, CA, United States, ³Radiology, Stanford University, Stanford, CA, United States

The MASTER sequence is a new approach to multi-slice thermometry with longer TEs and better measurement accuracy than comparable interleaved SPGR sequences. MR-guided Focused Ultrasound (MRgFUS) has been used to treat brain disorders including essential tremor and neuropathic pain. Current treatments are monitored using a single slice thermometry. MASTER will allow for larger volumes of temperature monitoring during treatment, helping ensure patient safety by warning clinicians if undesired heating occurs in healthy brain tissue.

Joshua de Bever^1,2, Nick Todd², Allison Payne^2,3, and Robert Roemer^4
1School of Computing, University of Utah, Salt Lake City, Utah, United States, ²Utah Center for Advanced Imaging Research, Salt Lake City, Utah, United States, ³Department of Radiology, University of Utah, Salt Lake City, Utah, United States, ⁴Department of Mechanical Engineering, University of Utah, Salt Lake City, Utah, United States

Treatment of large, complex, 3D tumor volumes using MR guided high intensity focused ultrasound (MRgHIFU) would be difficult for a human to perform efficiently and safely. The highly configurable adaptive model-predictive controller (AMPC) presented here addresses this challenge by leveraging MR thermometry as a feedback mechanism to automate MRgHIFU treatments. The AMPC dynamically identifies and adapts the heating model during treatment, obviating lengthy pre-treatment model identification, and making the controller’s predictions robust to changes in tissue properties. Tests performed in vivo show the AMPC successfully optimized the MRgHIFU treatment of 3D volumes while protecting healthy tissue.

Ashvin Kurian George¹, Nelly A. Volland^1,2, and Nassir F. Marrouche^1
1CARMA Center, University of Utah, Salt Lake City, Utah, United States, ²UCAIR, University of Utah, Salt Lake City, Utah, United States

A novel method PRF thermometry method combining spatially selective 2DRF excitation with EPI projection readouts for rapid image acquisition, and estimation of temperature using a parametric model.

Lawrence P. Panych¹, Renxin Chu¹, Chang-Sheng Mei¹, Guangyi Wang^1,2, W. Scott Hoge¹, Matthew Toews¹, and Bruno Madore^1
1Department of Radiology, Harvard Medical School, Brigham and Women's Hospital, Boston, MA, United States, ²Department of Radiology, Guangdong General Hospital, Guangdong Academy of Medical Sciences, Guangzhou, Guangdong, China

In thermometry, a number of advantages have been claimed in terms of temperature-to-noise ratio (TNR) and image contrast for sampling a spin-echo like signal in addition to the usual gradient-echo signal. The present work looked at the TNR claims more closely, in the context of abdominal imaging. While TNR advantages were observed and/or simulated for almost any combination of TR and flip angle (FA) settings for kidney imaging, due to differences in relaxation times such advantages were found in the liver only for TR < 20 ms and FA > 20 degrees or so.

David T. A. Fuentes¹, Samuel J. Fahrenholtz², Anil Shetty³, Roger J. McNichols³, Jeffrey S. Weinberg², John D. Hazle², and Jason Stafford^2
1The University of Texas M.D. Anderson Cancer Center, Houston, TX, United States, ²MD Anderson, Houston, TX, United States, ³Visualase Inc., Houston, Tx, United States

Significant efforts are ongoing to incorporate predictive prospective computer simulation into MRgLITT procedures. Truly predictive prospective computer modeling requires substantial validation efforts and novel computer modeling techniques that incorporate the uncertainty of the input of computer model parameters. Statistical methods provide novel methodologies for modeling the complex bioheat transfer phenomena. Within the probabilistic setting of uncertainty quantification (UQ), the range of constitutive nonlinearities may be modeled through the uncertainty within the linear UQ problem. This novel modeling techniques facilitates a substantial increase in computational efficiency while maintaining the predictability in the computer modeling by incorporating the advanced bioheat transfer phenomena.

Joshua P. Yung^1,2, Christopher J. MacLellan^1,2, Anil Shetty³, Roger J. McNichols³, John D. Hazle^1,2, R. Jason Stafford^1,2, and David T. A. Fuentes^1,2
1Imaging Physics, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, United States, ²The University of Texas Graduate School of Biomedical Sciences, Houston, Texas, United States, ³Visualase, Inc, Houston, Texas, United States

In previous studies, the Pennes bioheat transfer equation was used to predict temperature heating during thermal therapy. In this work, a non-physical model that takes a priori temperature measurements to predict future heating is used. The method was tested on in vivo data of laser induced interstitial thermal therapy of human brain. The prediction was run using two a priori time steps and three a priori time steps. In conclusion, the proposed method predicted future temperature values with uncertainty values allowing for confidence intervals. The mean and standard deviation values can offer additional information for the procedure.

Florian Maier¹, David T. A. Fuentes¹, Joshua P. Yung^1,2, Kamran Ahrar³, John D. Hazle¹, and R. Jason Stafford^1
1Department of Imaging Physics, The University of Texas M. D. Anderson Cancer Center, Houston, TX, United States, ²The University of Texas Graduate School of Biomedical Sciences, Houston, TX, United States, ³Department of Interventional Radiology, The University of Texas M. D. Anderson Cancer Center, Houston, TX, United States

Recently, a new physics-based method for referenceless proton resonance frequency shift thermometry was presented, which estimates background phase by solving the Dirichlet problem in a circular region of interest. In this work, we modified this algorithm to enable arbitrary regions of interest. The algorithm was tested with in vivo data of laser induced interstitial thermal therapy of human liver. For common ROIs in a 2D image a processing time of less than 0.1 s was found. In conclusion, the proposed algorithm enables definition of arbitrary ROIs and does not restrict users to circular ROIs.

Michael A. Boss¹, Jameel A. Feshitan², and Mark A. Borden^2
1National Instiute of Standards and Technology, Boulder, CO, United States, ²Mechanical Engineering, University of Colorado, Boulder, Boulder, CO, United States

Gas-filled microbubbles can be a useful theranostic agent in MRI-guided focused ultrasound (MRIgFUS). We have investigated the relaxation properties of microbubbles to which were conjugated Gd-DOTA, and found that relaxivity is a good indicator of the microbubble's integrity. We also elucidate the possible relaxation mechanisms of these microbubbles in the intact and fragmented states.

Bu S. Park¹, Martin J. Lizak², Leonardo M. Angelone¹, and Sunder S. Rajan^1
1CDRH/DP, FDA, Silver Spring, MD, United States, ²NIH, Bethesda, MD, United States

We present experimental results of MR thermometry using a novel lanthanide complex, Tm-DOTMA. Results from imaging the water protons and the methyl protons of Tm-DOTMA were compared, using both chemical shift imaging (CSI) and conventional water proton reference frequency (PRF) methods in a Tm-DOTMA gel phantom. In addition, the experimental results were compared with the numerical simulation results. Because Tm-DOTMA has much higher temperature sensitivity (~0.57- 0.7 ppm/°C) compared to water proton (~ 0.01 ppm/°C), the thermometry using Tm-DOTMA can detect small temperature changes with high temporal resolution using CSI method.

Michael A. Boss^1
1National Instiute of Standards and Technology, Boulder, CO, United States

The temperature-dependence of the chemical shift agents Tm-DOTA and Tm-DOTMA was determined from 5-45 °C in vitro. Previous measurements of the chemical shift of proton residues on lanthanide-chelates had indicated a linear dependence on temperature over small temperature ranges. However, it was found that linear calibrations can lead to errors of several degrees when using the -CH°₃ of Tm-DOTMA chemical shift as an MR thermometer. Our measurements over a larger temperature range indicate a non-linear dependence, and greatly improve the accuracy of MR thermometry performed with Tm-chelates.

Peter T. Fwu^1,2, Jeon-Hor Chen³, Yuting Lin¹, Wei-Ching Lin⁴, Po-Jung Tseng¹, Eddie Lin¹, and Min-Ying Su^1,2
1Center for Functional Onco-Imaging, Department of Radiological Sciences, University of California, Irvine, CA, United States, ²Department of Physics and Astronomy, University of California, Irvine, CA, United States, ³Center for Functional Onco-Imaging, Department of Radiological Sciences, University of California Irvine, Irvine, CA, United States,⁴Department of Radiology, China Medical University Hospital, Taichung, Taiwan

Regional hypothermia through endorectal cooling balloon (ECB) is shown capable of minimizing the long term urinary incontinence and sexual dysfunction. However, there is no tool to optimize ECB setting to reach the desired protective cooling effect. A bioheat simulation equation based on each individual patient’s own 3D anatomy is developed to model the cooling process by using the finite element method. Using this model, we evaluated the impact of geometrical distance from ECB, the balloon temperature, and the presence of blood. In addition, the effect of arterial ligation to decrease the blood perfusion to NBV was also evaluated.

Sherif G. Nour^1,2, David A. Kooby^3,4, Shishir K. Maithel^3,4, Charles A. Staley^3,4, Hiroumi D. Kitajims^1,2, William C. Small^1,4, and William E. Torres^1,4
1Radiology and Imaging Sciences, Emory University, Atlanta, GA, United States, ²Interventional MRI Program, Emory University, Atlanta, GA, United States, ³Surgical Oncology, Emory University, Atlanta, GA, United States, ⁴School of Medicine, Emory University, Atlanta, GA, United States

Adequate quantification of liver metastases and reliable targeting of subtle lesions for percutaneous ablation are current challenges leading to hepatic resections/open ablations that maybe avoidable particularly in poor surgical candidates. This investigation reports the utilization of intra-procedural gadoxetate disodium with controlled breath suspension under anesthesia for a)detecting subtle metastases not seen on pre-procedure scans in a manner analogous to using intra-operative ultrasound for metastatic mapping prior to metastatectomy; b)facilitating precise percutaneous targeting of subtle previously unapproachable lesions; c)allowing proper tailoring of ablations and inclusion of adequate safety margins around infiltrative lesions and d) enhancing ablation safety near central bile ducts.

Joost W. Wijlemans¹, Mario Ries², Martijn de Greef², Gerald Schubert³, Max Köhler³, Mika Ylihautala³, Lambertus W. Bartels², Chrit T.W. Moonen², and Maurice A.A.J. van den Bosch^1
1Department of Radiology, University Medical Center Utrecht, Utrecht, Netherlands, ²Image Sciences Institute, University Medical Center Utrecht, Utrecht, Netherlands, ³Philips Medical Systems MR, Vantaa, Finland

The heat sink effect, a major challenge in MR-HIFU ablation of liver tumors, can be reduced by performing intra-arterial embolization prior to MR-HIFU ablation. In this study, we investigated the synergistic effect of embolization and MR-HIFU. Six treatment cells were sonicated in a porcine liver model using PRFS thermometry. Subsequently, the liver was embolized and the treatment cells were re-sonicated. Mean temperature increase after embolization was 1.7 times higher than in perfused liver tissue. These results indicate that combination treatment with intra-arterial embolization and MR-HIFU could be an attractive treatment option for liver cancer patients in the palliative phase.

Matthew Bucknor¹, Viola Rieke¹, Youngho Seo¹, Randall Hawkins¹, Mark Wilson¹, Sharmila Majumdar¹, Thomas M. Link¹, and Maythem Saeed^1
1Radiology & Biomedical Imaging, University of California, San Francisco, San Francisco, CA, United States

MR-guided high intensity focused ultrasound (MRgHIFU) is a powerful technique for thermally ablating focal lesions. The purpose of this study was to clearly delineate post-treatment changes in bone following HIFU on MR, PET and MDCT imaging in a swine model. Two discrete ovoid lesions were created at each proximal diaphysis and distal metadiaphysis of the right femur. Follow-up imaging studies at five days demonstrated focal ovoid hypoenhancement and decreased radiotracer uptake (dynamic 18NaF-PET) which correlated precisely to the prescribed targets. CT failed to demonstrate the lesions. This study supports the usage of MRgHIFU in treating focal lesions in bone.

Joost W. Wijlemans¹, Mario Ries², Martijn de Greef², Gerald Schubert³, Max Köhler³, Mika Ylihautala³, Lambertus W. Bartels², Maurice A.A.J. van den Bosch¹, and Chrit T.W. Moonen^2
1Department of Radiology, University Medical Center Utrecht, Utrecht, Netherlands, ²Image Sciences Institute, University Medical Center Utrecht, Utrecht, Netherlands, ³Philips Medical Systems MR, Vantaa, Finland

MR-HIFU ablation of highly perfused liver tissue requires high powers with sufficient cool down times. Therefore, ablation of larger volumes will be time consuming and challenging in the clinical setting. In this study, we evaluated a clinically feasible treatment protocol using a porcine liver model. A clinical MR-HIFU system was used to sonicate seven 4mm treatment cells in the liver, which yielded a non-perfused volume of ~2.2ml. Post mortem examination revealed no near field damage. Ablation time was 95 minutes, total treatment time 135 minutes. These results demonstrate that it is clinically feasible to ablate substantial volumes of liver tissue.

Ying Zhu¹, Juan Wei², Bilgin Keserci³, Xuedong Yang¹, Rong Rong¹, Jing Liu¹, and Xiaoying Wang^1
1Radiology, Peking University First Hospital, Beijing, China, ²Philips Research Asia, Shanghai, China, ³Philips Healthcare South Korea, Seoul, Korea

Our study showed the tendency that the T2* values of type3 fibroids were obviously higher than type1 and type2. That implied the higher oxygenation of type3 fibroids, maybe associated with the mechanism that poor effect of the MR-HIFU in type3 fibroids. The oxygenation may increase in the residual non-necrotic area of the fibroids in the short-term after MR-HIFU

Benu Sethi¹, Andriy Shmatukha¹, Mohammed Shurrab², Jennifer Barry³, and Eugene Crystal^2
1Sunnybrook Research Institute, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada, ²Arrhythmia Services, Schulich Heart Centre, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada, ³Sunnybrook Research Institute, Sunnybrook Research Institute, Toronto, Ontario, Canada

3D High-Resolution MRI is used for analyzing the internal compositions of thermal ablation lesions and their evolution after the ablation. The usefulness of different MRI techniques and their correlation to histology at different post-ablation periods is discussed. T1w MRI is shown to discriminate between coagulation necrosis and hemorrhage in acute lesions. T2w MRI is demonstrated to depict the contraction band necrosis rim as accurately as delayed enhancement at a certain period after ablation, when the ablation lesion borders become well defined, leading to satisfactory ablation lesions border delineation without Gd injection.