ISMRM25 - AI: Prognostic & Predictive Models

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

AI: Prognostic & Predictive Models

Oral

AI & Machine Learning

Wednesday, 14 May 2025

310 (Lili-u Theater)

15:45 - 17:45

Moderators: Sola Adeleke & Wolfgang Bogner

Session Number: O-15

No CME/CE Credit

15:45		Introduction Sola Adeleke
15:57		0987. Radiomic profiling for IDH-mutant astrocytoma stratification with distinct biologic pathway activities PDF Poster View Presentation Video C. Zhang, J. Yan, K. Wang The First Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan, China Impact: Our study introduces a prognostic Radscore for non-invasive stratification of IDH-mutant astrocytomas. This score is informed by biological pathways associated with immunity, proliferation, cell function, and treatment response, thereby supporting targeted therapies and personalized management.
16:09		0988. Estimation of Time-to-Total Knee Replacement Surgery with Multimodal Modeling View Presentation Video O. Cigdem, H. Rajamohan, K. Cho, R. Kijowski, C. Deniz New York University Grossman School of Medicine, New York, United States Impact: Our model, using AI, survival analysis, and multimodal approaches, enhances TKR decision precision by accurately predicting time-to-TKR within 9 years, supporting personalized osteoarthritis treatment. It enables biomarker exploration and promotes early intervention strategies in knee osteoarthritis.
16:21		0989. Comparison of Imaging-Derived Features and Multimodal Models for Prognosis Prediction in Motor Neuron Disease F. Townend, A. Ijishakin, E. Spinelli, S. Basaia, Y. Falzone, P. Schito, M. Filippi, J. Grosskreutz, F. Agosta, R. Steinbach, A. Malaspina, J. Cole University College London, London, United Kingdom Impact: This study demonstrates that integrating routinely collected MRI, typically not used for prognosis, with clinical data can enhance prognostic predictions in motor neuron disease without additional patient data collection, as found through a comprehensive evaluation of models and imaging features.
16:33		0990. Deep learning-based biological age estimation from magnetic resonance imaging predicts cardiometabolic outcomes in the general population View Presentation Video M. Jung, M. Reisert, S. Rospleszcz, C. Schlett, M. Lu, F. Bamberg, V. Raghu, J. Weiss Massachusetts General Hospital, Boston, United States Impact: Individuals at high MRI-Age could benefit from personalized prevention strategies, lifestyle interventions, and treatment planning.
16:45		0991. Semi-automated Biparametric MRI Peritumoral Radiomics for Predicting the Risk of Positive Surgical Margin in Patients with Prostate Cancer View Presentation Video H. Xu, Q. Du, L. Xie, B. Liu, X. Bai, H. Ye, H. Wang 1st Medical Center of Chinese PLA General Hospital, Beijing, China Impact: A semi-automated bpMRI-based peritumoral radiomics can efficiently predict the risk of PSM.
16:57		0992. Meta-Learning-Driven Few-Shot Contrastive Learning for Stroke Prognosis Prediction across Multimodal Datasets View Presentation Video H. Peng, R. Zheng, Y. Zhang, C. Wang, H. Wang Institute of Science and Technology for Brain-Inspired Intelligence, Fudan University, Shanghai, China Impact: This research advances stroke recovery prediction by enhancing model robustness and generalization with limited data. The framework's ability to integrate diverse datasets could improve clinical decision-making in stroke rehabilitation, addressing a critical gap for accurate predictive modeling in this domain.
17:09		0993. Enhancing Brain Age Estimation with a Multimodal 3D CNN Approach Combining Structural MRI and AI-Synthesized Cerebral Blood Volume Data View Presentation Video J. Jomsky, Z. Li, Y. Zhang, A. Cao, J. Guo Columbia University, New York, United States Impact: By integrating functional AICBV data with structural T1w MRI, this study enhances brain age estimation, offering a non-invasive, cost-effective tool for early diagnosis of cognitive decline. It opens new research avenues in neurodegenerative disease detection and personalized brain health assessments.
17:21		0994. A Spatiotemporal Explainable Model for Predicting Pathological Complete Response to Neoadjuvant Chemotherapy Using Breast DCE-MRI View Presentation Video H. Yang, Y. Ren, M. Wang, D. Luo, W. Cui, Z. Hu, Z. Liu, N. Zhang Paul C. Lauterbur Research Center for Biomedical Imaging, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, ShenZhen, China Impact: This proposed method provides a more comprehensive understanding of the dynamic changes within the tumor, thereby improving the effectiveness of response assessment and offering valuable technical support for predicting neoadjuvant chemotherapy efficacy in breast cancer.
17:33		0995. Prediction of Early Neoadjuvant Chemotherapy Response of Breast Cancer through Deep Learning-Based Pharmacokinetic Quantification of DCE-MRI View Presentation Video C. Wu, L. Wang, N. Wang, S. Shiao, T. Dou, Y-C Hsu, A. Christodoulou, Y. Xie, D. Li Cedars-Sinai Medical Center, Los Angeles, United States Impact: This study introduces a quantitative, generalizable approach to early prediction of neoadjuvant chemotherapy (NAC) response in breast cancer, unlocking noninvasive imaging biomarkers with enhanced predictive accuracy and generalizability, thereby facilitating personalized treatment decisions without modifying clinical imaging protocols.

Similar Session(s)