ISMRM25 - Advanced Regularizers & AI-based Image Reconstruction

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

Advanced Regularizers & AI-based Image Reconstruction

Digital Poster

Acquisition & Reconstruction

Tuesday, 13 May 2025

Exhibition Hall

13:30 - 14:30

Session Number: D-10

No CME/CE Credit

		Computer Number: 17 2594. Deep subspace unrolling network for accelerating non-Cartesian sampled CMR Multitasking imaging View Presentation Video J. Zhao, S. Jia, Q. Liu, J. Hu, J. Cheng, Y. Wang, J. Xu, D. Liang, Y. Li Medical AI Research Center, Shenzhen Institutes of Advanced Technology, Shenzhen, China Impact: The proposed method reduces the imaging time of CMR multitasking from 71 minutes (MATLAB) to 5 minutes (GPU), providing possible clinical applications.
		Computer Number: 18 2595. Sensitivity analysis of self-supervised variational manifold learning based accelerated dynamic upper-airway collapse MRI View Presentation Video M. S. Ali, W. Alam, M. Jacob, S. Lingala University of Iowa, Iowa City, United States Impact: Enables robust sleep MRI methods for patient specific imaging diagnostics and therapeutic planning.
		Computer Number: 19 2596. The Effect of Axial Reorientation on Deep Learning-Based Susceptibility Mapping View Presentation Video F. Salman, T. Jochmann, I. Benslimane, S. Inglis, N. Bergsland, M. Dwyer, J. Haueisen, R. Zivadinov, F. Schweser Buffalo Neuroimaging Analysis Center, Department of Neurology at the Jacobs School of Medicine and Biomedical Sciences, University at Buffalo, The State University of New York, Buffalo, United States Impact: Using spline-based interpolation before background correction in QSM improves the visibility of clinical features, aiding the accuracy and effectiveness of QSM applications in diagnosing brain disorders.
		Computer Number: 20 2597. Improved Simultaneous Multislice EPI Reconstruction for Diffusion MRI Using Iterative Phase Error-Corrected SENSE (iPEC-SENSE) View Presentation Video M. Wang, Y. Zhao, H. Wang, Y. Liu Jinan University, Guangzhou, China Impact: Iterative phase error corrected SENSE (iPEC-SENSE) effectively suppresses noise and artifacts in diffusion MR images, significantly improves image quality and DTI analysis.
		Computer Number: 21 2598. Does it matter which subspectra are modelled in GABA-edited MRS? View Presentation Video Z. Shams, S. Murali-Manohar, H. Zöllner, G. Oeltzschner, R. Edden The Johns Hopkins University School of Medicine, Baltimore, United States Impact: New, more flexible MRS modeling software allows modeling of multiple sub-spectra of edited experiments, and it is important to appreciate the implications of choices of modeling approach.
		Computer Number: 22 2599. Implicit-GRAPPA for Data-Efficient, real-time free-breathing cardiac imaging View Presentation Video T. Zhao, M. Nishimura, D. Abraham, N. Seiberlich, K. Setsompop Stanford University, Stanford, United States Impact: The proposed implicit-GRAPPA improves the image reconstruction quality and markedly reduces the calibration data requirement of non-Cartesian GRAPPA, and should prove useful for applications requiring robust real-time non-Cartesian reconstructions.
		Computer Number: 23 2600. Maximizing the speed of Synthetic MRI acquisition while maintaining quantitative values: Validation study View Presentation Video M. Nakashima, T. Kawai, K. Matsumoto, T. Kawaguchi, T. Aoki, M. Yamada, M. Obara, A. Hiwatashi Graduate School of Medical Sciences, Nagoya City University, Nagoya, Japan Impact: Even with fast scanning of Synthetic MRI, T2 and R2 values are maintained. There is a potential for utilizing these values as new quantitative indicators for severity assessment and prognosis estimation in acute brain disorders.
		Computer Number: 24 2601. Robust subspace-based parallel imaging (SPAN): Bridging sensitivity maps and convolution kernels View Presentation Video T. Zu, Y. Zhang Zhejiang University, Hangzhou, China Impact: Transforming sensitivity maps into convolution kernels for parallel imaging can enhance image quality and robustness, and provide a novel perspective on leveraging sensitivity maps for researchers.
		Computer Number: 25 2602. Rapid, High-resolution and Distortion-free R2* Mapping of Fetal Brain using Multi-echo Radial FLASH and Model-based Reconstruction View Presentation Video X. Wang, H. Fan, S. Vasylechko, O. Afacan, E. Yang, M. Uecker, S. Warfield, A. Gholipour Boston Children's Hospital, Harvard Medical School, Boston, United States Impact: Our technique enables high-resolution, distortion-free R₂* mapping of the fetal brain using radial multi-echo acquisition and model-based reconstruction. This approach overcomes the low resolution and field distortion artifacts common in EPI-based methods.
		Computer Number: 26 2603. Unadjusted Langevin Sampling for Uncertainty Estimation in MRI Reconstruction - Theory and Numerical Validation View Presentation Video T. Holliber, M. Blumenthal, M. Uecker Institute of Biomedical Imaging, Graz University of Technology, Graz, Austria Impact: Uncertainty estimation via posterior sampling is an important tool to understand the reliability of reconstruction using generative machine-learning models.
		Computer Number: 27 2604. Computational Feasibility of MR Image Reconstruction via Explicit Construction and Inversion of the Encoding Matrix View Presentation Video K. Yeung, F. Gleeson, R. Schulte, D. Tyler, J. Grist, F. Wiesinger University of Oxford, Oxford, United Kingdom Impact: Demonstration of the current computational feasibility of generalized MR image reconstruction via direct pseudoinversion of the encoding matrix (Pinv-Recon), which is as a simple and versatile reconstruction approach able to incorporate a variety of encoding mechanisms and distortions.
		Computer Number: 28 2605. Accelerated Deep Learning Reconstructed Turbo Spin-Echo MRI for Temporomandibular Joint: A Prospective Study X. Lyu, Z. Ye, R. Zhao, C. Xia, T. Yin, D. Nickel, X. Xiong, Z. Li West China Hospital, Sichuan University, Chengdu, China Impact: This study demonstrated that accelerated TSE DL MRI reduced acquisition time of TMJ and held great image quality and comparable diagnostic confidence, which has great potential in optimizing clinical protocols and improving the comfort level of patients with TMJ disorders.
		Computer Number: 29 2606. Improved Reconstruction Speed for 5D Free Running using a Variable Projection Augmented Lagrangian (VPAL) Method with Targeted Descent Step View Presentation Video Y. Yang, M. Naeem, J. Yerly, D. Piccini, M. Stuber, J. Oshinski, M. Chung Georgia Institute of Technology and Emory University, Atlanta, United States Impact: This study shows that using an advanced numerical algorithm for highly under-sampled MR reconstruction both reduces computational time and results in better image quality for diagnostics, bringing 5D free-running imaging is closer to clinical usage.
		Computer Number: 30 2607. Pseudo real-time multi-parametric quantitative 3D MRI using Parallel non-Cartesian stDLNN for interventional procedures View Presentation Video Z. Wang, S. Wu, W. Feng, Z. Ding, H. She, Y. Du Shanghai Jiao Tong University, Shanghai, China Impact: This work demonstrates the potential for pseudo real-time multi-parametric 3D MRI during surgeries, supporting procedures like DBS by providing quantitative maps with a 11.2s acquisition and 4s computation, which could enhance intraoperative decision-making and improve patient outcomes.
		Computer Number: 31 2608. Accelerating MR Fingerprinting with Self-Supervised Learning Driven by Bloch Equations View Presentation Video Y. Liu, Y. Pang, Y. Chen, P-T Yap University of North Carolina at Chapel hill, Chapel Hill, United States Impact: Our method achieves better results than conventional dictionary matching and is faster than current self-supervised MRF methods.
		Computer Number: 32 2609. Zero-Shot Unsupervised Denoiser Learning for Plug-and-Play MRI Reconstruction View Presentation Video S. H. Kang, J. Cho, B. Bilgic, T. H. Kim Hongik University, Seoul, Korea, Republic of Impact: The proposed zero-shot method offers improved MRI reconstruction without requiring any training data. This framework not only provides enhanced image quality and adaptability across diverse MR applications but also is able to extend to other imaging tasks beyond MRI.

Similar Session(s)