Cape Town - 2026 ISMRM-ISMRT Annual Meeting and Exhibition

Registration, Atlases, and Motion

Oral

Analysis Methods

Wednesday, 13 May 2026

Meeting Room 2.60

08:20 - 10:10

Moderators: Yiming Dong & Xiaodong Zhong

Session Number: 530-02

No CME/CE Credit

Similar Sessions

This session will highlight recent advances in image registration, with an emphasis on AI-based methods. Contributions will span atlas construction and motion-related applications, reflecting current trends and open challenges in the field.

Skill Level: Intermediate

08:20		530-02-001. RealKeyMorph: A Resolution-Agnostic Keypoint-Based Image Registration Framework for MRI Mina Chookhachizadeh Moghadam, Alan Wang, Omer Taub, Martin Prince, Mert Sabuncu Weill Cornell Medicine, Cornell University, New York, United States of America Impact: This research revolutionizes medical image registration by enabling resolution-agnostic alignment of MR volumes without resampling. RKM serves as a powerful preprocessing tool for multi-stack reconstruction and longitudinal analysis, enhancing MR-based diagnostics by eliminating interpolation noise and preserving anatomical fidelity.
08:31		530-02-002. Test Time Adapted Generalized AI-based Medical Image Registration Method Sneha Sree C, Dattesh Dayanand Shanbhag, SUDHANYA Chatterjee GE HealthCare, Bengaluru, India Impact: This study enables fast, generalizable image registration across modalities, improving clinical workflows and diagnostic accuracy. It empowers clinicians with reliable motion correction, inspires scalable AI solutions, and opens new research into cross-modality registration and adaptive learning strategies.
08:42		530-02-003. Wavelet-Guided Deep Residual Network for Unsupervised Medical Image Registration Zhengyong Huang, Ning Jiang, Yao Sui Peking University, Beijing, China Impact: We develop a methodology designed to enable the precision and robustness of medical image registration. This approach highlights the potential of frequency-aware deep learning for modeling various deformations and offers a new strategy for improving registration performance.
08:53		530-02-004. Learning Non-Rigid Motion From MIMO RF Navigators Rinni Bhansali, Thomas Olausson, Gustavo Perez, Philippa Krahn, Alessandro Sbrizzi, Michael Lustig University of California, Berkeley, United States of America Impact: We integrate MIMO BPT and PT signals into a low-rank motion model, establishing a direct link between these external tones and complex 3D motion states. This demonstrates their potential for tracking non-rigid motion with per-readout temporal resolution in MRI.
09:04		530-02-005. Landmark Matching and B-spline Implicit Neural Representations for Diffusion-Weighted Imaging Distortion Correction Yunxiang Li, Yen-peng Liao, Yan Dai, Jie Deng, You Zhang University of Texas Southwestern Medical Center, Dallas, United States of America Impact: By combining the regularization properties of B-spline parameterization with the cross-modal matching capabilities of foundation models, our method achieves more accurate correction of geometric distortions in DWI, with the potential to enhance precision in intra/post-radiotherapy assessment.
09:15		530-02-006. Population-Representative Brain Templates and Morphometric Aging Signatures from Ultra-Low-Field (64 mT) MRI Kh Tohidul Islam, Parisa Zakavi, Shenjun Zhong, Sanuwani Dayarathna, Himashi Peiris, Helen Kavnoudias, Yi Chao Foong, Anneke Van Der Walt, Juan Domínguez D., Karen Caeyenberghs, Gary Egan, Meng Law, Zhaolin Chen Monash University, Clayton, Victoria, Clayton, Australia Impact: This study establishes age-specific brain templates and quantitative morphometry at 64 mT, demonstrating that ultra-low-field MRI can sensitively capture aging-related brain changes. It enables reliable population studies and extends quantitative neuroimaging to resource-limited, point-of-care environments.
09:26		530-02-007. A Novel Joint Synthesis and Registration Framework for Registering Diffusion MRI and T1-weighted Images Xiaofan Wang, Junyi Wang, Yuqian Chen, Lauren O’Donnell, Fan Zhang University of Electronic Science and Technology of China, Chengdu, China Impact: The proposed framework enables effective fusion of dMRI and T1w information, allowing dMRI-derived features, such as tractography, to be spatially aligned within a standard anatomical space. This capability facilitates population-level analyses, brain atlas construction, and clinical planning.
09:37		530-02-008. One Touch Patient Registration and Smart MR Scan Planning Using Deep Learning SAGNIK GHOSH, SANKET MALI, HARIKRISHNA RAI GE HealthCare (Bengaluru, India), Bengaluru, India Impact: This 3D vision–based, contactless morphometric and BMI estimation framework with automatic 3D patient contour detection transforms MR/PET-MR workflows—empowering radiologists, technologists, and patients through SAR/SUV-aware intelligent scan planning, enabling personalized imaging, adaptive protocolling, and precision dosing while advancing safety and automation.
09:48		530-02-009. Atlas-based Personalized Aorta Topological Heatmaps for Improved Characterization of Thoracic Aortic Disease William Dong, Ethan Johnson, Bradley Allen, Haben Berhane, Charilaos Apostolidis, Michael Markl Northwestern University, Chicago, United States of America Impact: The personalized aortic topological heatmap approach has the potential to fully leverage the 3D information of MRA in risk assessment of thoracic aorta disease by incorporating geometric features not represented by standard aorta diameter measurements.
09:59		530-02-010. Improving Cross-Field MRI Alignment with Nonlinear ANTs Registration for Reliable Connectivity Mapping Jamini Bhagu, Gail Harmata, Adriana Rivera-Dompenciel, Michelle Voss, Jenny Richards, Sarah Smith, Spencer Smith, Aislinn Williams, Jerome Maller, John Wemmie, Vincent Magnotta University of Iowa, Iowa City, United States of America Impact: Improved alignment of 7T functional and 3T anatomical MRI using nonlinear registration enables more accurate mapping of brain connectivity. This refinement enhances reproducibility in multimodal studies and supports better investigation of neural circuits relevant to mood and neuropsychiatric disorders.