Cape Town - 2026 ISMRM-ISMRT Annual Meeting and Exhibition - Artifact Correction Strategies in Brain and Body

Artifact Correction Strategies in Brain and Body

Digital Poster

Acquisition & Reconstruction

Tuesday, 12 May 2026

Digital Posters Row B

13:40 - 14:35

Session Number: 461-03

No CME/CE Credit

Similar Sessions

Artifact correction strategies for brain and body.

Skill Level: Advanced

		461-03-001. Correcting Concomitant Gradient Field in Multi-Echo GRE Tissue Field Map Measurement Haodong Zhong, Yang Song, Yi Wang, Jianqi Li East China Normal University, Shanghai, China Impact: Phantom-based calibration of echo-dependent concomitant-field bias improves multi-echo GRE tissue-field maps, enhancing spatial uniformity and yielding more homogeneous susceptibility, thereby increasing the accuracy and reliability of quantitative susceptibility mapping in neuroimaging studies.
		461-03-002. Smooth Total variation Regularization for Interference Detection and Elimination (STRIDE) for MRI Alexander Mertens, Diego Martinez, Amgad Louka, Ying Yang, Chad Harris, Ian Connell University of Toronto, Toronto, Canada Impact: Our algorithm improves our understanding of electromagnetic interference (EMI) removal from EMI corrupted data. Minimizing spatial total-variation in EMI removal is a simple, yet effective technique that improves results and can be incorporated into any subtraction-based EMI removal scheme.
		461-03-003. Towards elucidating ripple artifacts encountered in accelerated 3D GRE acquisitions at UHF Joseph Obriot, Franck Mauconduit, Vincent Gras, Chaithya Giliyar Radhakrishna, Philipp Ehses, Rüdiger Stirnberg, Caroline LeSter, Nicolas Boulant Université Paris-Saclay, CEA, CNRS, BAOBAB, Neurospin, Gif-sur-Yvette, France Impact: Ripple artifacts in accelerated 3D GRE stem from ACS–imaging inconsistencies under strong $B_0$ gradients. External short-TE ACS acquisitions mitigate these effects, enabling accurate GRAPPA reconstruction and improving reconstruction reliability while other sources of mismatch can make the artifact re-emerge.
		461-03-004. Motion Parameter Estimation for Brain MRI Using Physics-Based K-Space Simulation and Deep Learning Sena Azamat, Saritha Unnikrishnan, Esin Ozturk Isik Basaksehir Cam and Sakura City Hospital, Istanbul, Turkey Impact: A multi-head CNN was developed to estimate brain MRI motion parameters with submillimeter translation accuracy (0.05–1.25mm) and strong correlation for k-space corruption (r=0.88), potentially replacing subjective visual assessment and reducing repeat scans, costs, and interpretation variability.
		461-03-005. Shear correction of in-plane off-resonance artifacts in Cartesian spin-echo imaging Alexander Toews, Brian Hargreaves Stanford University, Stanford, United States of America Impact: The prospective distortion correction of View-Angle Tilting (VAT) can now be applied retrospectively with the flexibility to suppress blurring in on-resonant pixels. Clinical metal protocols can be shortened by synthesizing both VAT on/off images from one acquisition instead of two.
		461-03-006. Rapid comprehensive 3D volumetric MN-MRI using accelerated non-Cartesian techniques with a single-channel dual-tuned MN coil Minako Azuma, Masahiro Enzaki, Masami Yoneyama Miyazaki University, Miyazaki, Japan Impact: Challenges in MN-MRI (long scan times, low SNR) are addressed by the Spiral/FLORET technique. It achieved shorter scan times and acceptable SNR, CNR, and image quality for both 3D volumetric $^{1}$H and $^{23}$Na MRI compared to conventional methods.
		461-03-007. Model-based distortion correction for EPI MR images using learnt priors SUDHANYA Chatterjee, Florian Wiesinger, Dattesh Dayanand Shanbhag, Patricia Lan, Rajagopalan Sundaresan, Marc Lebel, Arnaud Guidon GE HealthCare, Bangalore, India Impact: The proposed AI-informed, model-based approach effectively corrects distortion in EPI-based MR imaging, thus improving anatomical accuracy and supporting reliable diagnoses. This advancement may reduce misinterpretation risks for radiologists, particularly in high-susceptibility regions, enhancing clinical confidence and diagnostic quality.
		461-03-008. Deep learning-accelerated ultra-fast Isotropic 3D FSE shoulder MRI outperforms Conventional 2D FSE in diagnostic assessment Yufan Gao, Weiyin Vivian Liu, Yunfei Zha Renmin Hospital of Wuhan University, Wuhan 430060,China, Wuhan, China Impact: DL-accelerated 3D FSE enables rapid, high-resolution isotropic shoulder imaging, while maintaining robust diagnostic performance. This technology promises enhanced workflow efficiency and improved patient comfort.
		461-03-009. Fast and interpretable motion and eddy-current correction in dMRI Kinda Muqari, Samo Lasic, Nicola Spotorno, Markus Nilsson Lund University, Lund, Sweden Impact: ECHO enables fast and accurate correction of motion and eddy-current distortions in diffusion MRI, thereby supporting large-scale studies of neurological disorders.
		461-03-010. Deep Learning Reconstruction with Chemical Shift Correction Enhances Craniofacial Bone Imaging in ZTE MRI Shotaro Fuchibe, Naoko Takagawa, Yuka Uchimoto, Yuka Uchiyama, Sagar Mandava, Kanae Moriyama, maggie fung, Nozomu Uetake, Shumei Murakami GE HealthCare, Hino, Japan Impact: Deep learning-based chemical shift correction significantly improves ZTE MRI quality in craniofacial imaging, enabling clearer visualization of bone and dental structures. This approach may reduce radiation exposure and expand MRI applications in orthodontics, TMJ evaluation, and pediatric dentistry.
		461-03-011. Deep Learning-Based Motion Artifact Correction in DSC MRI using PINN-Based Perfusion-Aware Loss Sojeong Kim, Asaduddin Muhammad, Sung-Hong Park Korea Advanced Institute of Science & Technology, Daejeon, Korea, Republic of Impact: This work emphasizes superiority, robustness and advantage of perfusion-aware loss in both U-Net and Restormer for artifact correction in DSC MRI, uniquely enabling the generation of accurate perfusion maps from motion corrupted DSC MRI data.
		461-03-012. Harnessing Open Innovation for Integrated Retrospective Motion Correction Daniel Nicolas Splitthoff, Patrick Hucker, Jesús Díaz Pereira, Jakob Slipsager, Michael Koenig, Marcel Dominik Nickel, Tobias Würfl, Maxim Zaitsev, Stefan Glimberg, Thomas Gaass Research & Clinical Translation, Magnetic Resonance, Siemens Healthineers AG, Erlangen, Germany Impact: We demonstrate the benefits of enabling Open Innovation interfaces for retrospective motion correction approaches.
		461-03-013. High-Resolution 7T Imaging of the In Vivo Human Cerebellum: A Synergistic Approach by Universal Pulse and Motion Correction Yuancheng Jiang, Jianxun Qu, Franck Mauconduit, Daniel Gallichan, Caohui Duan MR Research Collaboration, Siemens Healthineers Ltd., Beijing, China Impact: The proposed 7T cerebellum imaging protocol by UP and FatNav enables clear visualization of fine structures like small veins and folia. This may improve the lesion identification and allow more precise measurement of cerebellar atrophy.
		461-03-014. Post-processing techniques for correcting fat ghosting in multi-shot echo-planar diffusion weighted imaging Kalina Jordanova, Bruce Daniel, Brian Hargreaves Stanford University, Stanford, United States of America Impact: Multi-shot echo planar diffusion-weighted imaging may enable breast cancer detection without contrast agents, but can be limited by artifacts from residual fat. Exploiting low diffusivity of fat allows reduction of artifacts in post-processing of any diffusion-weighted data, improving clinical utility.
		461-03-015. Sampling Density Compensation using Fast Fourier Deconvolution Rui Luo, Peng Hu, Haikun Qi ShanghaiTech University, Shanghai, China Impact: This method reduces the computation time of a 3D density compensation function (DCF) from tens of minutes to tens of seconds, substantially reduces the time cost to obtain the DCF in non-Cartesian reconstruction, such as NUIFFT, iterative and deep-learning reconstruction.