Cape Town - 2026 ISMRM-ISMRT Annual Meeting and Exhibition

AI Applications in Body MRI

Traditional Poster

Body

Wednesday, 13 May 2026

Traditional Posters | Exhibition Hall

08:20 - 09:15

Session Number: 570-01

No CME/CE Credit

Similar Sessions

This session highlights clinically applicable AI solution for different oncological problems in the body

Skill Level: Intermediate

		570-01-165. MULTI-CENTER RADIOMICS-DEEP LEARNING FUSION MODEL FOR STRATIFYING IPMN MALIGNANCY RISK ON MRI Andrea Bejar, María Jaramillo Gonzalez, Ziliang Hong, Gorkem Durak, Elif Keles, Halil Ertugrul Aktas, Fergan Bol, Lili Zhao, Chao Chen, Concetto Spampinato, Alpay Medetalibeyoglu, Sukru Mehmet Erturk, Yury Velichko, Emil Agarunov, Ziyue Xu, Sachin Jambawalikar, Ivo Schoots, Muhammed Enes Tasci, Marco Bruno, Chenchang Huang, Tamas Gonda, Candice Bolan, Frank Miller, Michael Wallace, Rajesh Keswani, Pallavi Tiwari, Ulas Bagci Northwestern University Feinberg School of Medicine, Chicago, United States of America Impact: Our approach provides objective risk stratification of IPMNs that could reduce unnecessary, invasive interventions for low-risk lesions. Integration with radiology workstations would allow real-time assessment without disrupting workflow. Cost-effectiveness analyses and prospective validation are needed to advance clinical translation.
		570-01-166. An interpretable machine learning model in predicting neoadjuvant chemoradiotherapy response using multiparametric MRI xuemeng Li, Fei Gao The First Affiliated Hospital of USTC, Hefei, China Impact: The nomogram based on multiparametric MRI radiomics and clinical-radiological features may be used as an accurate and non-invasive method to predict the efficacy of nCRT in rectal cancer patients, and the Shapley algorithm could provide interpretability of the radiomics model.
		570-01-167. 2.5D DL with Multi-Instance Learning Predicts Axillary Lymph Node Metastasis in Breast Cancer Lingsong Meng, Xiaoan Zhang, Xin Zhao, Lin Lu, Xiang Meng, Fuming Shao, Yuxia Zhang Shangqiu Medical College, Shangqiu, China Impact: This multi-center study developed a novel 2.5D deep learning model that non-invasively predicts axillary lymph node metastasis in breast cancer, demonstrating high accuracy and robust generalizability to aid personalized surgical planning.
		570-01-168. Mapping of Prostate Biological Age from MRI: Advancing Early Detection and Preventive Care in Prostate Health Roozbeh Bazargani, Saqib Basar, Mahya Mobinikhaledi, Sadiya Hussain Chatham Parayil, Rodrigo Solis Pompa, Mojan Izadkhah, Daniel Daly-Grafstein, Javad Khaghani, Duc Nguyen, Saurabh Garg, Siavash Khallaghi, Yuntong Ma, Sam Hashemi Prenuvo, Inc, San Francisco, United States of America Impact: AI‑driven estimation of prostate biological age using axial T2 -weighted pelvic MRI introduces a non‑invasive biomarker for age‑related prostate changes. Its association with benign prostatic hyperplasia suggests its potential for early detection and risk assessment in prostate health management.
		570-01-169. "Contrast or Not? Leveraging DINOv2 for Robust Detection of Contrast Uptake in Dynamic MRI" Dattesh Dayanand Shanbhag, Gurunath Reddy Madhumani, Seyed Iman Zare Estakhraji, Marc Lebel, Sajith Rajamani, Uday Patil GE HealthCare, Bangalore, India Impact: This work enables automated detection of contrast administration failures in DCE-MRI using foundation models, overcoming unreliable DICOM metadata. It enhances clinical reliability across anatomies and scanners, supporting robust quality control and protocol validation in dynamic imaging workflows
		570-01-170. Comparative Analysis of AI and Radiologist Assessments for Breast Cancer Detection on MRI Nika Rasoolzadeh, Koen Eppenhof, Ritse Mann Radboud University Medical Center, Nijmegen, Netherlands Impact: Evidence of an AI model’s diagnostic behavior demonstrates its feasibility for integration into breast MRI workflows and assisting radiologists. It enables informed deployment decisions and workflow optimization leading to potential workload reduction and patient care improvement.