Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2508.19325 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915760954146816 |
|---|---|
| author | Su, Haoyang Xiang, Jin-Yi Rui, Shaohao Gao, Yifan Chen, Xingyu Yin, Tingxuan Zhang, Shaoting Wang, Xiaosong Wu, Lian-Ming |
| author_facet | Su, Haoyang Xiang, Jin-Yi Rui, Shaohao Gao, Yifan Chen, Xingyu Yin, Tingxuan Zhang, Shaoting Wang, Xiaosong Wu, Lian-Ming |
| contents | Accurate prediction of major adverse cardiac events (MACE) remains a central challenge in cardiovascular prognosis. We present PRISM (Prompt-guided Representation Integration for Survival Modeling), a self-supervised framework that integrates visual representations from non-contrast cardiac cine magnetic resonance imaging with structured electronic health records (EHRs) for survival analysis. PRISM extracts temporally synchronized imaging features through motion-aware multi-view distillation and modulates them using medically informed textual prompts to enable fine-grained risk prediction. Across four independent clinical cohorts, PRISM consistently surpasses classical survival prediction models and state-of-the-art (SOTA) deep learning baselines under internal and external validation. Further clinical findings demonstrate that the combined imaging and EHR representations derived from PRISM provide valuable insights into cardiac risk across diverse cohorts. Three distinct imaging signatures associated with elevated MACE risk are uncovered, including lateral wall dyssynchrony, inferior wall hypersensitivity, and anterior elevated focus during diastole. Prompt-guided attribution further identifies hypertension, diabetes, and smoking as dominant contributors among clinical and physiological EHR factors. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_19325 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | PRISM: A Framework Harnessing Unsupervised Visual Representations and Textual Prompts for Explainable MACE Survival Prediction from Cardiac Cine MRI Su, Haoyang Xiang, Jin-Yi Rui, Shaohao Gao, Yifan Chen, Xingyu Yin, Tingxuan Zhang, Shaoting Wang, Xiaosong Wu, Lian-Ming Computer Vision and Pattern Recognition Accurate prediction of major adverse cardiac events (MACE) remains a central challenge in cardiovascular prognosis. We present PRISM (Prompt-guided Representation Integration for Survival Modeling), a self-supervised framework that integrates visual representations from non-contrast cardiac cine magnetic resonance imaging with structured electronic health records (EHRs) for survival analysis. PRISM extracts temporally synchronized imaging features through motion-aware multi-view distillation and modulates them using medically informed textual prompts to enable fine-grained risk prediction. Across four independent clinical cohorts, PRISM consistently surpasses classical survival prediction models and state-of-the-art (SOTA) deep learning baselines under internal and external validation. Further clinical findings demonstrate that the combined imaging and EHR representations derived from PRISM provide valuable insights into cardiac risk across diverse cohorts. Three distinct imaging signatures associated with elevated MACE risk are uncovered, including lateral wall dyssynchrony, inferior wall hypersensitivity, and anterior elevated focus during diastole. Prompt-guided attribution further identifies hypertension, diabetes, and smoking as dominant contributors among clinical and physiological EHR factors. |
| title | PRISM: A Framework Harnessing Unsupervised Visual Representations and Textual Prompts for Explainable MACE Survival Prediction from Cardiac Cine MRI |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2508.19325 |