{"ID":2856179,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.11112","arxiv_id":"2510.11112","title":"Multimodal Disease Progression Modeling via Spatiotemporal Disentanglement and Multiscale Alignment","abstract":"Longitudinal multimodal data, including electronic health records (EHR) and sequential chest X-rays (CXRs), is critical for modeling disease progression, yet remains underutilized due to two key challenges: (1) redundancy in consecutive CXR sequences, where static anatomical regions dominate over clinically-meaningful dynamics, and (2) temporal misalignment between sparse, irregular imaging and continuous EHR data. We introduce $\\texttt{DiPro}$, a novel framework that addresses these challenges through region-aware disentanglement and multi-timescale alignment. First, we disentangle static (anatomy) and dynamic (pathology progression) features in sequential CXRs, prioritizing disease-relevant changes. Second, we hierarchically align these static and dynamic CXR features with asynchronous EHR data via local (pairwise interval-level) and global (full-sequence) synchronization to model coherent progression pathways. Extensive experiments on the MIMIC dataset demonstrate that $\\texttt{DiPro}$ could effectively extract temporal clinical dynamics and achieve state-of-the-art performance on both disease progression identification and general ICU prediction tasks.","short_abstract":"Longitudinal multimodal data, including electronic health records (EHR) and sequential chest X-rays (CXRs), is critical for modeling disease progression, yet remains underutilized due to two key challenges: (1) redundancy in consecutive CXR sequences, where static anatomical regions dominate over clinically-meaningful...","url_abs":"https://arxiv.org/abs/2510.11112","url_pdf":"https://arxiv.org/pdf/2510.11112v1","authors":"[\"Chen Liu\",\"Wenfang Yao\",\"Kejing Yin\",\"William K. Cheung\",\"Jing Qin\"]","published":"2025-10-13T08:02:36Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","has_code":false}