XueHua Peng¹, LiFei Ma², Rui Zhang², and XiaoMeng Wu^2
1Radiology Department, WuHan Children’s Hosptial, WuHan, China, ²philips, shanghai, China

Keywords: Parkinson's Disease, Quantitative Susceptibility mapping, Rapid QSM measurement, EPI acquisition, Neuro system

Quantitative susceptibility mapping（QSM） can be used for quantitatively measurement of neuro system magnetic susceptibility information caused by various diseases, such as Alzheimer's disease (AD), Parkinson's disease (PD) and neuro vascular diseases. In this study, a rapid QSM acquisition based on EPI sequence was proposed to improve clinical feasibility of QSM. The results demonstrated the rapid QSM were very close to those obtained by standard QSM.

View Presentation Video

Shengyong Li¹, Xiaonan Wang^2,3, Yida Wang¹, Yang Song⁴, and Guang Yang^1
1Shanghai Key Laboratory of Magnetic Resonance, East China Normal University, Shanghai, China, ²Department of Radiology, Beijing Hospital, National Center of Gerontology, Institute of Geriatric Medicine, Beijing, China, ³Graduate School of Peking Union Medical College, Chinese Academy of Medical Sciences, Beijing, China, ⁴MR Scientific Marketing, Siemens Healthcare, Shanghai, China

Keywords: Alzheimer's Disease, Arterial spin labelling

To investigate the added value of T1-mapping to arterial spin labeling (ASL) for computer-aided early diagnosis of Alzheimer’s disease (AD). A total of 97 (45 AD/24 mild cognitive impairment (MCI)/38 normal control (NC)) people were enrolled retrospectively. We extracted features from 24 automatically segmented brain regions based on T1-mapping and ASL MR images and constructed three radiomics models to differentiate AD-NC/MIC-NC/AD-MCI, for which the radiomics models achieved a favorable prediction performance with the AUCs of 0.921/0.764/0.727, respectively.

View Presentation Video

J. Kevin DeMarco^1,2, H. Douglas Morris², Robert Y. Shih^1,2, Daniel W. Bess³, Luca Marinelli⁴, Vincent B. Ho^1,2, and Thomas K. F. Foo^4
1Walter Reed National Military Medical Center, Bethesda, MD, United States, ²Uniformed Services University of the Health Sciences, Bethesda, MD, United States, ³Fort Belvoir Community Hospital, Fort Belvoir, VA, United States, ⁴GE Global Research Center, Niskayuna, NY, United States

Keywords: Gradients, Brain

Ultra-high-performance MAGNUS gradient coils optimized for advanced diffusion imaging were used to obtain routine anatomical brain imaging in a cohort of 31 consecutive subjects enrolled in a mild TBI study. In a blinded review, three neuroradiologists reviewed 3D MPRAGE and 3D T2 FLAIR obtained with MAGNUS and a whole-body 3T scanner (WB3T). Substantial agreement of better overall image quality for MAGNUS vs. WB3T was seen for T2 FLAIR, and mild agreement of better overall image quality for MPRAGE. MAGNUS supports rapid gradient switching and higher gradient strength for additional SNR with 3D T2 FLAIR.

View Presentation Video

Bahman Tahayori¹, Robert E. Smith¹, David N. Vaughan¹, Chris Tailby¹, Eric Y. Pierre¹, Graeme D. Jackson^1,2, and David F. Abbott^1,2
1The Florey Institute of Neuroscience and Mental Health, Heidelberg, Australia, ²Florey Department of Neuroscience and Mental Health, University of Melbourne, Melbourne, Australia

Keywords: Data Analysis, fMRI (task based)

Multi-Echo fMRI data acquisition has multiple advantages over single-echo acquisition. Principal amongst them is that multiple echoes can distinguish neural activity from artefacts. TE Dependent ANAlysis (TEDANA) is an existing software tool designed to denoise multi-echo fMRI datasets. We evaluated the performance of TEDANA to denoise fMRI data of 120 subjects. Our results demonstrated that TEDANA improved the activation detection at a group level. However, for a subset of subjects TEDANA degraded their individual result substantially. We identified potential causes and proposed a modified framework for multi-echo data analysis that provides reasonable results at an individual subject level. 

View Presentation Video

Anika Strittmatter^1,2, Lothar R. Schad^1,2, and Frank G. Zöllner^1,2
1Computer Assisted Clinical Medicine, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany, ²Mannheim Institute for Intelligent Systems in Medicine, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany

Keywords: Machine Learning/Artificial Intelligence, Data Processing, Image Registration

In this research we investigated the performance of published neural networks for an affine registration of multimodal medical images and examined the networks' generalizability to new datasets. The neural networks were trained and evaluated using a synthetic multimodal dataset of three-dimensional CT and MRI volumes of the liver. We compared the Normalised Mutual Information, Dice coefficient and the Hausdorff distance across the neural networks described in the papers, using our CNN as a benchmark and the conventional affine registration method as a baseline. Seven networks improved the pre-registration Dice coefficient and are therefore able to generalise to new datasets.

View Presentation Video

Yasutaka Fushimi¹, Satoshi Nakajima¹, Sachi Okuchi¹, Akihiko Sakata¹, Takuya Hinoda¹, Sayo Otani¹, Azusa Sakurama¹, Krishna Pandu Wicaksono¹, Hiroshi Tagawa¹, Yang Wang¹, Satoshi Ikeda¹, Shuichi Ito¹, Miyuki Takiya¹, and Yuji Nakamoto^1
1Department of Diagnostic Imaging and Nuclear Medicine, Kyoto University Graduate School of Medicine, Kyoto, Japan

Keywords: Quantitative Imaging, Brain

Recently, strategically acquired gradient echo (STAGE) imaging has been developed as a potential standardized brain imaging protocol. STAGE is designed to use two 3D fully flow compensated multi-echo SWI sequences with 2 different flip angles (FAs). We evaluated the inter-scanner consistency of T1 values and proton spin density (PSD) values derived from STAGE imaging for elderly healthy volunteers. We conducted VOI analysis both in Native space and “Mutual space” using AAL3 VOIs. Parcellation was automatically performed with few manipulations, and good ICC and minimal biases were demonstrated between the two scanners.

View Presentation Video

Aidin Arbabi¹, Irina De Alba Alvarez¹, Vitaliy Khlebnikov¹, Jose P. Marques¹, and David. G. Norris^1
1Radboud University, Nijmegen, Netherlands

Keywords: Quantitative Imaging, Brain

Frequency Offset maps are widely used to correct image distortion or estimate signal loss in gradient echo signal due to field inhomogeneities. Routinely, multiecho gradient echo sequences are used to collect this information. In this work, we introduce an innovative method which enables frequency offset mapping via one single-shot acquisition. The proposed method offers a significant reduction in the acquisition time without loss of accuracy.

View Presentation Video

Ziyu Meng¹, Tianyao Wang², Bin Bo¹, Yibo Zhao^3,4, Yudu Li^3,5, Rong Guo^3,6, Wen Jin^3,4, Xin Yu⁷, Zhi-Pei Liang^3,4, and Yao Li^1
1School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China, ²Radiology Department, The Fifth People's Hospital of Shanghai, Fudan University, Shanghai, China, ³Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ⁴Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ⁵National Center for Supercomputing Applications, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ⁶Siemens Medical Solutions USA, Inc, Urbana, IL, United States, ⁷Department of Biomedical Engineering, Case Western Reserve University, Cleveland, OH, United States

Keywords: Stroke, Stroke

Accurate assessment of tissue viability is of great importance for the design of therapeutic interventions in acute stroke. ¹H-MRSI and quantitative T₂ imaging can provide neurochemical biomarkers sensitive to the pathological changes in stroke but their applications have been limited by long scan times. This study introduces a novel approach that combines fast high-resolution 3D-MRSI with quantitative T₂ imaging to assess the viability of ischemic tissues in acute stroke. This approach achieved improved discrimination between ischemic brain tissues that ultimately infarcted and those recovered using joint neurometabolite concentrations and quantitative T₂ values, in comparison with classical DWI and PWI methods.

View Presentation Video

James Grover^1,2, Paul Liu^1,2, Bin Dong², Shanshan Shan^1,2, Brendan Whelan^1,2, Paul Keall^1,2, and David Waddington^1,2
1Image X Institute, University of Sydney, Sydney, Australia, ²Ingham Institute for Applied Medical Research, Sydney, Australia

Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence, MRI guided radiation therapy

Real-time MRI is limited in its spatiotemporal resolution due to imaging time being proportional to the spatial resolution. Super-resolution imaging was integrated into an MRI-linac to improve the spatiotemporal resolution of images used in real-time adaptive MRI guided radiation therapy. Real-time up-sampling techniques included conventional bicubic interpolation and deep learning-based super-resolution. Up-sampling increased the spatial resolution as characterised by healthy volunteer brain and thorax MRIs with negligible impact on the temporal resolution as measured in a motion phantom tracking experiment.

View Presentation Video

Quan Chen^1,2, Xiaozhi Cao¹, Congyu Liao¹, Siddharth Srinivasan Iyer³, Sophie Schauman¹, Nan Wang¹, and Kawin Setsompop^1
1Department of Radiology, Stanford University, Stanford, CA, United States, ²standford university, palo alto, CA, United States, ³Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Stanford, CA, United States

Keywords: Quantitative Imaging, Brain

A rotation-fitting based PV reduction pipeline with B0, B1+ and frequency-based flip angle (FA) scaling corrections are proposed to improve the accuracy and repeatability of magnetic resonance fingerprinting (MRF) in the existence of motion. 5 repeated scans from both stable baseline and big motion positions were acquired and analyzed to compare the repeatability of MRF under different rotation-fitting schemes and B0, B1+ corrections. Rotation before the fitting can substantially reduce the PV effects. The B0/ B1+ and frequency-based FA scaling corrections improve the estimation accuracy. The 2min and 1min 1mm3 whole brain MRF quantitative results also indicate a high repeatability. 

View Presentation Video

Xiaohan Guo¹, Pingping Wang¹, Gengming Mu¹, Tianju Jia¹, Kaiguo Zhu¹, Kai Ai², and Baoying Chen^1
1XI'AN INTERNATIONAL MEDICAL CENTER HOSPITAL, Xi'an, China, ²Philips Healthcare, Xi'an, China

Keywords: Blood vessels, Blood vessels

Basi-parallel anatomic scanning magnetic resonance imaging (BPAS-MRI) is an MRI technique that can reveal the outer contour of the vertebrobasilar artery, even in the presence of occlusion. A total of 53 patients with 106 segments of blood vessels were scanned by BPAS+MRA, and the results were verified by comparing with VW-MRI. It was found that BPAS+MRA could significantly improve the detection rate of vertebrobasilar lesions, which was of great value for the diagnosis and differential diagnosis of arterial dysplasia, arteriosclerosis and arterial occlusion, and could improve the diagnostic confidence of radiologists.

Xin Chen^1,2, Tao Gong¹, Tong Chen³, Changyuan Xu¹, Yuchao Li¹, Qingxu Song⁴, Liangjie lin⁵, Georg Oeltzschner^6,7, Richard A. E. Edden⁶, Guangbin Wang¹, and Zhangyong Xia^2
1Shandong Provincial Hospital Affiliated to Shandong First Medical University, Jinan, China, ²Liaocheng People’s Hospital, Liaocheng, China, ³Shandong Provincial Hospital, Cheeloo College of Medicine, Shandong University, Jinan, China, ⁴Qilu Hospital of Shandong University, Jinan, China, ⁵Philips Healthcare, Beijing, China, ⁶The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD, United States, ⁷F.M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, MD, United States

Keywords: Alzheimer's Disease, Brain, Amnestic mild cognitive impairment; hippocampus; MEGA-PRESS; amide proton transfer-weighted imaging; imaging diagnostic marker

Amnestic mild cognitive impairment (aMCI) is considered as a prodromal stage of Alzheimer's disease (AD). Growing evidence supports the hypothesis that unbalanced excitatory/inhibitory (glutamate and gamma-aminobutyric acid) neurotransmitters and consequent abnormal protein deposition in hippocampus contribute to the pathological process of MCI and AD. We explored the changes of hippocampal Glx/GABA+ levels and APTw in aMCI patients using MEGA-PRESS and APTw imaging. Patients with aMCI exhibited decreased Glx levels (and Glx/GABA+ ratios) and increased APTw values in hippocampus. The combination of Glx and APTw values improved the diagnostic performance for aMCI, suggesting it was a potential imaging diagnostic marker.

View Presentation Video

Zijing Dong^1,2, Fuyixue Wang^1,2, Amelia K. Strom^1,3, Korbinian Eckstein⁴, Beata Bachrata⁴, Simon D. Robinson⁴, Bruce R. Rosen^1,2,3, Lawrence L. Wald^1,2,3, Laura D. Lewis⁵, and Jonathan R. Polimeni^1,2,3
1Athinoula A. Martinos Center for Biomedical Imaging, MGH, Charlestown, MA, United States, ²Department of Radiology, Harvard Medical School, Charlestown, MA, United States, ³Harvard-MIT Health Sciences and Technology, MIT, Cambridge, MA, United States, ⁴High Field MR Centre, Medical University of Vienna, Vienna, Austria, ⁵Department of Biomedical Engineering, Boston University, Cambridge, MA, United States

Keywords: Neurofluids, Brain

Cerebrospinal fluid (CSF) flow is a key component of the brain’s waste clearance system. However, little is known about the brain-wide CSF flow in subarachnoid space (SAS), due to the low sensitivity of MRI-based flow imaging methods for measuring slow flow. Here, we proposed a phase-contrast 4D CSF flowmetry method using a pulsed-gradient-spin-echo EPI sequence, providing high sensitivity and efficiency to measure brain-wide slow CSF flow dynamics in SAS. Whole-brain CSF flow dynamics including velocity and direction changes were measured and our preliminary data suggest that both cardiac pulsation and respiration can drive CSF flow in both ventricles and SAS.

View Presentation Video

Magnus Svensen¹, Christian Dölle², Njål Brekke³, Heidi Espeland⁴, Nora Tvedten⁴, John Georg Seland⁴, Frank Riemer⁵, and Charalampos Tzoulis^6
1Neuro-SysMed & Mohn Medical Imaging- and Visualization Center (MMIV), Bergen, Norway, ²Neuro-SysMed & University of Bergen, Bergen, Norway, ³Haukeland University Hospital, Bergen, Norway, ⁴University of Bergen, Bergen, Norway, ⁵Mohn Medical Imaging- and Visualization Center (MMIV), Bergen, Norway, ⁶Neuro-SysMed & University of Bergen & Haukeland University Hospital, Bergen, Norway

Keywords: Data Analysis, Spectroscopy, 31P

1H-decoupled NOE-enhanced 31P-MRS has promising applications in the diagnosis and stratification of neurodegenerative diseases such as Alzheimer's disease, Parkinson's disease and amytrophic lateral sclerosis (ALS) due to its ability to longitudinally and noninvasively measure changes to cell metabolism and energy conversion, which is heavily involved in these diseases. To aid clinical translation, this study focuses on assessing the reproducibility of 1H-decoupled NOE-enhanced 31P-MRS by measuring the resonances of metabolites involved in neural energy metabolism. 

View Presentation Video

Mauro Zucchelli¹, Christos Papageorgakis¹, and Stefano Casagranda^1
1Department of R&D Advanced Applications, Olea Medical, La Ciotat, France

Keywords: Brain Connectivity, Tractography & Fibre Modelling, Colormap, Diffusion

We propose a new tractography colormap based on the HSL model to complement the classical RGB colormap used in the field. Our colormap is based on a new feature, called "streamlines normal", an oriented vector that is computed from the tractography. The streamlines normal in combination with the HSL colormap improves tractography data visualization and can be an asset for better and faster bundle segmentation both manually and automatically. 

View Presentation Video

Sina Amirrajab¹, Rien Boonstoppel¹, Aymen Ayaz¹, and Marcel Breeuwer^1,2
1Biomedical Engineering, Eindhoven University of Technology, Eindhoven, Netherlands, ²MR R&D - Clinical Science, Philips Healthcare, Eindhoven, Netherlands

Keywords: Machine Learning/Artificial Intelligence, Brain, Super resolution

In many medical applications, high resolution images are required to facilitate early and accurate diagnosis. In this paper, we investigate the potential of data augmentation using simulated brain magnetic resonance (MR) images for training a deep-learning (DL) super resolution model that can generalize to brain data from different publicly available sources. Our qualitative visual evaluation results suggest that data augmentation with simulated images can improve the robustness and generalization of the model and decrease the artifacts of the super-resolved images.

View Presentation Video

Alireza Sharifzadeh-Kermani¹, Finbar Argus¹, Jiantao Shen¹, Sarah-Jane Guild², David Dubowitz², Miriam Scadeng^2,3, Paul Condron³, Eryn Kwon^1,2,3, Samantha Holdsworth^2,3, Gonzalo Daniel Maso Talou¹, and Soroush Safaei^1
1Auckland Bioengineering Institute, The University of Auckland, Auckland, New Zealand, ²Faculty of Medical and Health Sciences & Centre for Brain Research, The University of Auckland, Auckland, New Zealand, ³Mātai Medical Research Institute, Tairāwhiti-Gisborne, New Zealand

Keywords: Blood vessels, Modelling, Arterial Spin Labeling, ASL, cerebral perfusion, CBF, 4D flow, blood flow

Computed flows via haemodynamic models highly depend on peripheral resistances that affect the blood perfusion. To this end, finding a way to estimate these resistances is crucial for having a subject-specific model. Perfusion (CBF) maps produced by Arterial Spin Labeling (ASL)-MRI provide valuable information via which the model can be personalized. This new approach may be promising for providing blood flow measurements of the brain from ASL-CBF data in the absence of a full brain 4D flow sequence.

Rong Wang¹, Yuxin Li¹, Jie Lin¹, Yong Zhang², and Jiankun Dai^3
1Huashan Hospital, Shanghai, China, ²GE Healthcare, Shanghai, China, Shanghai, China, ³MR Research, GE Healthcare, Beijing, China, Shanghai, China

Keywords: Stroke, fMRI (resting state)

This is the first study that used a combined arterial spin labeling (ASL) and resting-state fMRI approach to assess the neurovascular coupling in patients with mitochondrial myopathy, encephalopathy, lactic acidosis and stroke-like episodes (MELAS), which may provide a new mechanistic perspective into understanding numerous brain diseases.

View Presentation Video

Alexandra Grace Roberts^1,2, Yi Wang^1,2, Pascal Spincemaille², and Thanh Nguyen^2
1Electrical Engineering, Cornell University, Ithaca, NY, United States, ²Radiology, Weill Cornell Medicine, New York, NY, United States

Keywords: Machine Learning/Artificial Intelligence, Brain

3D super-resolution of QSM is feasible using the VDSR and U-net architectures using a fraction of the required number of epochs as previous 2D super-resolution networks. Additionally, this method both reduces whole brain MSE and ROI MSE and increases the apparent resolution as compared to the interpolated input

View Presentation Video

Aocai Yang^1,2, Hangwei Zhuang^3,4, Guolin Ma², and Yi Wang^3,5
1Peking Union Medical College, Beijing, China, ²Department of Radiology, China–Japan Friendship Hospital, Beijing, China, ³Department of Biomedical Engineering, Cornell University, New York, NY, United States, ⁴Weill Cornell Medical College, Department of Radiology, New York, NY, United States, ⁵Department of Radiology, Weill Cornell Medical College, New York, NY, United States

Keywords: Alzheimer's Disease, Alzheimer's Disease

 We assessed the whole-brain oxygen metabolism perturbations in AD using QSM and quantitative BOLD. mGRE data and QQ model was employed to calculate OEF and ASL data was used to reconstruct CBF map. The CMRO2 can be estimated from CBF and OEF based on Fick’s principle. Our results demonstrated a characteristic whole-brain hypoperfusion and hypometabolism pattern in AD, predominantly located within default-mode network. Additionally, decreased CBF and CMRO2 in substructures of bilateral hippocampus strongly correlated with global cognition. QQ model-based noninvasively quantitative measurements have a great potential to be complementary biomarkers for evaluating cognitive impairment in AD.