ISMRM25 - Promising AI Applications in Body MRI

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

Promising AI Applications in Body MRI

Power Pitch

Body

Monday, 12 May 2025

Power Pitch Theatre 2

08:15 - 10:15

Moderators: Amitkumar Choudhari, Filiz Yetisir & Amitkumar Choudhari

Session Number: PP-07

No CME/CE Credit

08:15		Screen Number: 26 0091. Multicenter, Multivendor Development and Evaluation of Automated Liver MR Image Prescription View Presentation Video G. Fullerton, J. Starekova, C. Buelo, D. Harris, R. do Vale Souza, A. Faacks, A. Anagnostopoulos, A. Murphy, N. Duritsa, E. Agarwal, L. Müller, D. Kadi, T. Yokoo, M. Bashir, S. Reeder, D. Hernando University of Wisconsin-Madison, Madison, United States Impact: The validated multicenter, multivendor liver prescription model may improve liver MRI workflow efficiency and reproducibility. Further, the trained models will be made available upon publication, providing an important resource for research and clinical MR applications in the liver.
08:17		Screen Number: 27 0092. Joint Denoising and Reconstruction of T2-Weighted PROPELLER MRI of the Lung at 0.55T Using Self-Supervised Deep Learning View Presentation Video J. Chen, H. Pei, M. Bruno, Q. Wen, C. Maier, D. Sodickson, H. Chandarana, L. Feng New York University Grossman School of Medicine, New York, United States Impact: We developed a self-supervised learning-based joint reconstruction and denoising scheme for lung MRI at 0.55T. The proposed self-supervised model enhances image quality by reducing noise and improving structural clarity.
08:19		Screen Number: 28 0093. Feasibility of Automated Liver Metastasis Detection and Report Generation in Gadoxeotic-enhanced Abdominal MRI using Vision-Language Models View Presentation Video Y. Cai, Z. Zhu, P. Bassi, Z. Zhou, K. Wang, Y. Yang Northeastern University, Berkeley, United States Impact: VLMs offer a promising approach to enhancing both the accuracy and efficiency of liver metastasis evaluation in abdominal MRI interpretation.
08:21		Screen Number: 29 0094. Multi-contrast MR-driven deep learning for abdominal multi-organ segmentation (McDAMOS) View Presentation Video P. Wang, D. Ruan, J. Chen, J. Xiao, D. Ling, L. Ma, W. Yang, Z. Fan University of Southern California, Los Angeles, United States Impact: Our work demonstrates the utility of multi-contrast MR in achieving abdominal auto-segmentation and presents a methodology to address limited data available from a novel research MR sequence. The approach benefits clinicians and propelling automated segmentation techniques forward.
08:23		Screen Number: 30 0095. Neural Network Methods Enhance Test-Retest Repeatability of IVIM Parameter Estimation in Pancreatic Imaging View Presentation Video N. Avidan Pearl, D. Link-Sourani, R. Weiss, M. Freiman Technion, Haifa, Israel Impact: This study demonstrates that neural network-based methods, specifically the motion-robust IVIM-Morph and SUPER-IVIM-DC for limited b-values, improve test-retest repeatability in IVIM parameter estimation for pancreatic imaging, supporting better assessment of pancreatic conditions and enhancing diagnosis, treatment planning, and patient outcomes.
08:25		Screen Number: 31 0096. Deep Learning Synthesized Hepatobiliary Phase Images for Optimizing Clinical Workflow of Gd-EOB-DTPA-enhanced Liver MRI: A Multicenter Study View Presentation Video K. Zhao, Y. Liu, Y. Zhang, Y. Xu, Y. Wang, J. Qin, F. Yang, J. Liu, T. Niendorf, Z. Liu, G. Wang Guangdong Provincial People‘s Hospital, Guangzhou, China Impact: Our study demonstrated the importance of an early HBP scan of 5 minutes for high-quality HBP synthesis. Our findings help to shorten examination times and support the optimization of scanning protocols of Gd-EOB-DTPA-enhanced liver MRI.
08:27		Screen Number: 32 0097. Shortening 3D T2-Weighted Breast MRI scan time using deep learning based reconstruction: A phantom and patient reader study View Presentation Video P. Wang, C. Lu, K. Keen, L. Wilmes, S-H Chou, M. Chung, A. Lee, X. Zhu, A. Guidon, N. Hylton, B. Joe GE Healthcare, Menlo Park, United States Impact: This work demonstrated the feasibility of using a pseudo-random under-sampled acquisition coupled with deep learning-based reconstruction (Sonic DL) to reduce 3D T2w breast MRI scan time by 50% while improving image quality.
08:29		Screen Number: 33 0098. Unsupervised learning based on clinical factors and MRI radiomic features to predict 5-year progression-free survival in prostate cancer G. Hu, X. Liu Fudan University Shanghai Cancer Center, Shanghai, China Impact: Unsupervised learning-based bpMRI radiomics features and clinical factors have high predictive prognostic value, and these features have the potential to help to identify high-risk patients at an early stage, adjust the treatment regimen, and improve the prognosis of patients.
08:31		Screen Number: 34 0099. WITHDRAWN
08:33		Screen Number: 35 0100. Accelerating Free-Breathing Liver MRI at 7T using Recurrent Inference Machines View Presentation Video M. Tavakkoli, D. van den Berg, B. A. Runderkamp, W. van der Zwaag, M. D. Noseworthy, M. W.A. Caan McMaster University, Hamilton, Canada Impact: This work advances ultra-high field, free-breathing liver MRI with deep-learning reconstruction, offering improved motion robustness over CS. It paves the way for prospectively undersampled, submillimeter resolution free-breathing acquisitions in future studies while maintaining short scan durations and minimizing patient discomfort.
08:35		Screen Number: 36 0101. Impact of Deep Learning denoising and ultra-high density coil array on prostate diffusion imaging View Presentation Video S. Huang, X. Wang, M. Medved, C. K. Follante, Y-J Stickle, P. Lan, A. Yousuf, R. Engelmann, F. Robb, A. Guidon, G. Lee, A. Oto GE HealthCare, Royal Oak, United States Impact: This study shows a clinically feasible approach by combining a novel 50-channel pelvic coil with perineal coverage with DL based reconstruction for an improved DWI imaging in prostate MR.
08:37		Screen Number: 37 0102. Perspectives on Data Sharing and Artificial Intelligence Among Participants in Renal Clinical Studies View Presentation Video V. Aramendía-Vidaurreta, L. García-Ruiz, M. Aznarez, M. Aastrup, M. Bozzetto, P. Brambilla, R. Echeverria-Chasco, E. S. Hansen, L. Micu, J. M. Mora-Gutierrez, S. Pasini, A. Raj, S. Ringgaard, N. M. Selby, A. Strittmatter, T. Vendelboe, G. Villa, I. Urdea, N. Henrik Buus, N. Garcia-Fernandez, M. Trillini, S. Francis, L-M Itu, C. Laustsen, F. G. Zoellner, A. Caroli, M. A. Fernández-Seara Clínica Universidad de Navarra, Pamplona, Spain Impact: Current findings highlight the importance of improving institutional trust and education on AI to foster patient engagement in data sharing practices, and the use of AI in medical imaging.
08:39		Screen Number: 38 0103. Harmonization of AI/DL accelerated quantitative bi-parametric prostate MRI: demonstration in multi-parametric phantom and patients View Presentation Video D. Malyarenko, S. Swanson, J. Richardson, S. Lowe, J. O'Connor, J. Fajardo, Y. Jiang, S. Wells, T. Chenevert University of Michigan, Ann Arbor, United States Impact: AI/DL-accelerated acquisition implemented on two clinical 3T MRI vendor systems allows six-minute quantitative bpMRI for prostate patients. The developed QA workflow enables harmonization of quantitative T₂and ADC mapping in multi-vendor clinical settings.
08:41		Screen Number: 39 0104. Value of Deep Learning-Accelerated T1-w Dixon MRI for Upper Abdominal Imaging View Presentation Video J. Fingerhut, T. Scheef, H. Engel, A. Rau, L. Kolbe, C. Wilpert, R. Strecker, M. Nickel, F. Bamberg, J. Weiss, N. Verloh Medical Center – University of Freiburg, Freiburg, Germany Impact: DL-accelerated MRI optimizes scan time and image quality, especially benefiting patients with breath-holding difficulties.
08:43		Screen Number: 40 0105. Deep Supervision Attention U-net for segmentation of uterine zones: a multi-center study View Presentation Video S. Tripathy, N. Castro, M. May, L. Siegler, L. Story, M. Uder, J. Hutter Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany Impact: This study enabled the deep learning-based automatic segmentation of the uterine zones, that would in future provide deeper insights into uterine layer changes at different menstrual cycle points and facilitate the study of Adenomyosis and other uterine abnormalities.
08:45		Screen Number: 41 0106. The potential of quantitative synthetic MRI for predicting rectal cancer stage and pathological characteristics K. Wang, L. Zhu, W. Feng, Y. Xia, J. Dai, B. Shi, H. Shen, B. Ding, H. Zhang Ruijin Hospital, Shanghai Jiao Tong University of Medicine, shanghai, China Impact: The standard MRI performs poorly in N staging. Our study showed qsMRI was significantly superior to subjective assessment for differentiating advanced N stage RC patients. Therefore, qsMRI may be used as non-invasive imaging technique for evaluating and managing RC patients.
08:47		Screen Number: 42 0107. Preliminary Results of Prospective Clinical Study Evaluating Machine Learning Software for Detecting Significant Prostate Cancer on bpMRI View Presentation Video M. R. S. Sunoqrot, R. Segre, T. A. Sjøbakk, G. A. Nketiah, P. Davik, S. Langørgen, M. Elschot, T. F. Bathen NTNU - Norwegian University of Science and Technology, Trondheim, Norway Impact: his prospective clinical study suggests that PROVIZ software can enhance diagnostic accuracy in prostate cancer care, improve clinically significant lesion targeting, and potentially reduce unnecessary biopsies, thereby minimizing overdiagnosis and improving patient outcomes in prostate cancer management.
08:49		Screen Number: 43 0108. Diagnostic value of deep learning–based renal virtual ASL sequences in CKD View Presentation Video Y. Chen, P. Luo, R. Lai, J. Mo, W. Liu, R. Qi, J. Li, Q. Chen, Q. Liang, F. Meng, H. Qin, B. Kuehn, Y. Zeng, B. Huang Shenzhen Traditional Chinese Medicine Hospital (The Fourth Clinical Medical College of Guangzhou University of Chinese Medicine), Shenzhen, China Impact: This virtual ASL technique enables non-invasive renal perfusion assessment based on conventional MRI sequences, potentially expanding access to perfusion imaging in CKD diagnosis and monitoring. Future studies can explore its application in treatment response prediction.
08:51		Screen Number: 44 0109. A Deep Learning 3D Super-Resolution Radiomics Model based on Gd-Enhanced MRI for Improving Pre-operative Prediction of HCC Pathological Grading F. JIA, X. ZHAO, Y. XIONG, J. ZHANG Department of Magnetic Resonance, The Second Hospital & Clinical Medical School, Lanzhou University, Lanzhou, China., LANZHOU, China Impact: The study’s findings suggest that super-resolution imaging can significantly enhance radiomics models’ accuracy in predicting hepatocellular carcinoma grades, potentially leading to better patient stratification and personalized treatment plans, thus improving clinical outcomes and guiding future research in oncologic imaging.
08:53		Screen Number: 45 0110. qMRI enhances standard-of-care for hepatocellular carcinoma detection: proof-of-concept using Bayesian networks View Presentation Video Y-C Wang, Y. J. Wong, J. W. S. Choo, S. B. Ithnin, G. H. Lee, M. A. De Roza, K. K. ONG, P. C. Y. Chia, K. T. Tan, N. C. Tan, H. C. Toh, O. F. Chong, J. P. E. Chang, W. L. Yang, X. Y. Yeap, C. M. E. Chua, J. S. S. Chua, J. S. Q. Goh, Y. K. Sim, C. C. Y. Lim, D. Bulte, M. Brady, P. Chow University of Oxford, Oxford, United Kingdom Impact: qMRI information increases a clinician’s confidence in ruling out HCC using either GLMs or BNs. Additionally, Bayesian networks enable incremental assessment of new metrics before the completion of data collection.
08:55		Screen Number: 46 0111. Machine Learning-Based Automating Breast Cancer Detection and Classification using DWI View Presentation Video M. Iima, R. Mizuno, M. Kataoka, A. Minami, M. Honda, K. Imanishi, Y. Zhang, H. Satake, R. Ito, S. Naganawa, Y. Nakamoto Nagoya University Graduate School of Medicine, Nagoya, Japan Impact: Potential to boost screening efficiency, minimize false positives, and improve patient care via more precise, swift diagnoses.
08:57		Screen Number: 47 0112. Automated Measurement of Membranous Urethral Length (MUL) on MRI Images Using Deep Learning View Presentation Video A. Hadjivasiliou, K. Hong, I. Zaffar, L. Dickinson, Z. Tandogdu, I. Drobnjak UCL, London, United Kingdom Impact: By eliminating the need for specialized radiologist expertise, this automated system could enable widespread adoption of MUL-based surgical planning in resource-limited settings, helping surgeons optimize their approach to preserve urinary continence for prostate cancer patients.
08:59		Screen Number: 48 0113. Fast T2WI with DLR-Enhanced SSFSE: A Reliable Solution for diagnosing Acute Abdominal Pain in Emergency Settings View Presentation Video J. Xu, L. Zhu, W. Liu, W. Liu, Y. Lu, J. Liu, C. Ma, Y. Zhang, X. Wang, F. Feng Peking Union Medical College Hospital, Beijing, China Impact: This study highlights the potential of combining deep learning reconstruction with SSFSE T2WI to significantly improve image quality and diagnostic performance compared to conventional T2WI, offering a rapid and reliable diagnostic option for evaluating acute abdominal conditions in emergency settings.
09:01		Screen Number: 49 0114. 3D T1, T2, and M0 Mapping of the Bladder Wall in Healthy Subjects Using Deep Image Prior MR Fingerprinting View Presentation Video J. Hamilton, G. Ippolito, S. Wells University of Michigan, Ann Arbor, United States Impact: High-resolution 3D bladder MRF may offer potential future diagnostic utility in various conditions, such as overactive bladder, where inflammatory and fibrotic changes in the detrusor muscle are believed to drive symptoms.
09:03		Screen Number: 50 0115. WITHDRAWN

Similar Session(s)