ISMRM25 - Image Recon / Motion Correction / AI

ISMRM & ISMRT Annual Meeting & Exhibition • 10-15 May 2025 • Honolulu, Hawai'i

Image Recon / Motion Correction / AI

Traditional Poster

Tuesday, 13 May 2025

Building: Room: Exhibition Hall

16:45 - 17:45

Session Number: T-02

No CME/CE Credit

		5121. Rapid Quantitative Susceptibility Mapping for Intracranial Hemorrhage using Deep Learning-based 2.5D Diffusion Models PDF Poster Z. Xiong, Y. Gao, W. Jiang, K. Butcher, A. Wilman, H. Sun University of Queensland, Brisbane, Australia Impact: This reliability across imaging conditions highlights QSMDiff’s potential as a versatile and accurate tool for clinical susceptibility mapping for ICH patients.
		5122. Rank-One-Approximated Decomposition for N-dimensional NMR Spectroscopy Reconstruction with Physical Intelligent Neural Network PDF Poster Y. Huang, Y. Gao, D. Guo, V. Orekhov, T. Agback, X. Qu Xiamen University, Xiamen, China Impact: Designed by fast low-rank approximation, neural network correction and peak retrieval, ROAD shows the robust reconstruction benefited from optimization and fast computation from deep learning. Instead of reconstructing high-dimensional signals directly, ROAD reconstructs signals in each dimension with high efficiency.
		5123. Evaluation of Deep learning-based reconstruction synthetic magnetic resonance imaging MAGIC PDF Poster J. Wang, H. Xu, Y. Hou, B. Xu, J. Lu Xuanwu Hospital Capital Medical University, Beijing, China Impact: The utilization of deep learning-based image reconstruction techniques in MAGIC image not only provide high quality images, but also underscore the potential of the MAGIC DL methodology as an alternative tool to detect lesions in pathology studies.
		5124. Predictive Imaging Methodology for Real-Time MRI-guided Radiotherapy PDF Poster G. Han, M. Wright, E. Yip, B. G. Fallone, J. Yun, K. Wachowicz University of Alberta, Edmonton, Canada Impact: Prediction of images beyond the current real-time image will allow a simple and accurate means of controlling beam position, overcoming issues with equipment lag. Because this work does not rely on prior characterizations, it is potentially robust against non-cyclic motions.
		5125. Accelerated Magnetic Resonance Imaging via Quantum Computing and Deep Neural Network PDF Poster S. Zhou, Y. Zhou, C. Liu, Y. Zhu, H. Zheng, D. Liang, H. Wang Shenzhen Institute of Advanced Technology，Chinese Academy of Sciences, Shenzhen, China Impact: This fully demonstrates the potential and broad development prospects of quantum computing in accelerating magnetic resonance imaging.
		5126. Accurate, markerless optical tracking of head pose for motion correction: A dual-camera simulation with deep learning PDF Poster M. Silic, S. Graham University of Toronto, Toronto, Canada Impact: Accurate, markerless OT is feasible in simulations with two in-bore cameras and deep learning. Pre-training of a twin neural network was successful (mean RMSE = 0.13 mm/degrees) motivating additional development in the real world, towards motion correction in MRI.
		5127. Retrospective Motion Correction at Ultra-High Field using SAMER and Reflected Power PDF Poster Y. HUANG, D. Polak, T. Yu, A. Klauser, N. Montemayor, J. Philippe, E. Sleight, T. Kober, D. Popp, D. N. Splitthoff, H. Tan, G. F. Piredda, T. Hilbert  Siemens Shenzhen Magnetic Resonance Ltd, Shenzhen, China Impact: Retrospectively reducing motion artifacts in ultra-high field scans enhances diagnostic usefulness, especially for high-resolution acquisitions to detect small focal pathology, a typical clinical use case of ultra-high field scanners.
		5128. Rigid Body Head Tracking with Scattering: A Comparison of Linear, Gaussian Process and Deep Learning Models PDF Poster H. Simmons, J. Kent, M. Nadar, M. Mostapha, J. Tanner, A. Hess Wellcome Centre for Integrative Neuroimaging, FMRIB, Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, United Kingdom Impact: We have shown that Gaussian Process and simple deep learning models can outperform the existing linear models for tracking head pose with scattering on a single subject. This could be valuable in the future for applications to motion correction.
		5129. Clinical evaluation of motion-corrected 2D axial T2-weighted turbo spin echo 3T brain MRI in the neurological intensive care unit PDF Poster A. Hajati, C-H Chiang, S. Yee, A. Tabari, D. Polak, D. Splithoff, B. Clifford, W-C Lo, Y. T. Huang, S. Cauley, J. Conklin, S. Huang Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Harvard Medical School, Charlestown, United States Impact: SAMER effectively corrects for motion and enhances the diagnostic quality of axial T2 TSE brain MRI in motion-prone, critically ill patients, supporting its utility for more accurate, timely assessments in neurologically compromised individuals.
		5130. Rigid and multirigid motion-mitigated MRI with self-gated Cartesian acquisition PDF Poster Z. Zhou, W. Yang, W. Liu, H. Qi, P. Hu ShanghaiTech University, Shanghai, China Impact: The self-gating technique can detect rigid or multirigid motions with high sensitivity. It is also robust for different readout directions. With a minimal acquisition time overhead, this technique enables motion-mitigated reconstruction and the image quality is significantly improved.
		5131. Up to four times accelerated musculoskeletal MRI at 0.4T using the CIRIM-network PDF Poster D. van den Berg, R. Varriale, F. Ferrando, P. Traverso, L. Balbi, G. Strijkers, M. Caan Amsterdam UMC, Amsterdam, Netherlands Impact: Low-field MRI is cost-effective and can therefore increase the worldwide accessibility of MRI. By accelerating imaging using deep learning reconstruction on undersampled data, we realize time-efficient scanning.
		5132. WITHDRAWN
		5133. An Algorithm for Detecting Short-Lived Motion-Induced Corruption in k-Space PDF Poster T. Höpfner, F. Werner, T. Wech, H. Köstler University of Würzburg, Würzburg, Germany Impact: We propose a method to automatically detect k-space lines, corrupted by patient motion, using a multiscale scanning statistic, allowing to replace these data in optimized reconstructions.
		5134. Reducing aliasing-induced artifacts in BLADE sequence using extended-FOV reconstruction PDF Poster K. Zhou, N. Xiao, L. Yang Siemens Shenzhen Magnetic Resonance Ltd., Shenzhen, China Impact: The extended-FOV method can be integrated seamlessly into the existing scanning process without requiring additional scan time, to effectively eliminate aliasing-induced artifacts.
		5135. A Rapid Free-Breathing Whole-Abdomen Scan for Simultaneous T1, T2, and T2* Mapping Across Abdominal Organs PDF Poster M. K. Manhard, A. Kilpattu Ramaniharan, J. Greer, J. Tkach, A. Trout, J. Dillman, A. Pednekar Cincinnati Children's Hospital Medical Center, Cincinnati, United States Impact: This study demonstrates that a single one-minute, free-breathing MI-SAGE scan can achieve T₁, T₂, and T₂^* estimates across abdominal organs, comparable to published values and consistent with liver and muscle values from conventional mapping techniques from multiple breath-hold scans.
		5136. Evaluating how far we can minimize MP2RAGE acquisition time with compressed sensing PDF Poster I. KIDA National Institute of Information and Communications Technology, Suita, Japan Impact: The significant scan time reduction obtained by CS-MP2RAGE sequence could improve clinical and experimental workflow efficiency, reduce patient discomfort, and expand research possibilities in ultra-high-field MRI.
		5137. Surface Tracking-assisted Real-time 4D Lung MRI PDF Poster X. Liang, L. Pan, E. Nevo, H. Soomro, S. Roys, R. Gullapalli, A. Sawant, T. Ernst, J. Zhuo University of Maryland School of Medicine, Baltimore, United States Impact: Continuous 4D lung MRI allows accurate modeling of respiration-induced internal target motion to surface motion prior to radiotherapy, and thus enables real-time target tracking during external beam radiation treatment without the need of MR linac.
		5138. Motion-Robust T1/T2 Mapping of the Abdomen using Pilot-Tone Navigation PDF Poster C. Ariyurek, B. Bilgic, S. Fujita, Y. Jun, S. Kurugol, B. Gagoski, O. Afacan Boston Children's Hospital and Harvard Medical School, Boston, United States Impact: We introduce a motion-robust technique for high-resolution T1/T2 mapping of the kidneys using 3D-QALAS with pilot-tone-navigation. By overcoming respiratory motion challenges, this method could improve the accuracy and clinical applicability of quantitative renal MRI, enabling earlier detection of kidney diseases.

Similar Session(s)