发表机构
Southern Methodist University（南卫理公会大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对EHR数据多模态预测中现有方法缺乏可解释性的问题，提出显式多模态路由框架，通过构建不同模态路由及引入推理时路由掩码实现可解释、稳健且可审计的推理，经实验评估揭示了临床状况组模态依赖差异。
AI 中文摘要
电子健康记录（EHR）数据本质上是多模态的，利用多种模态可提高预测性能。但现有多数方法依赖深度融合，影响了对各模态预测贡献的解释及多模态推理的可解释性。本文提出一个用于临床预测的显式多模态路由框架，能跨结构化纵向变量（L）、临床笔记（N）和胸部X光（I）三种EHR模态进行可解释、稳健且可审计的推理。模型构建离散单模态、定向双模态和三模态路由，还引入推理时路由掩码来审计多模态推理和评估稳健性。通过MIMIC-IV中的三模态患者住院数据进行多标签表型预测（K = 25）和二元ICU死亡率预测评估，揭示了不同临床状况组对模态依赖的系统差异。总体而言，该框架为多模态临床预测提供了透明、可审计且实用的方法，具备可解释性、稳健性，并能洞察不同数据源如何驱动模型决策。
英文摘要
Electronic health record (EHR) data are inherently multimodal, and leveraging multiple modalities can improve predictive performance. However, most existing approaches rely on deep fusion, which obscures how individual modalities contribute to predictions and limits the interpretability of multimodal reasoning. We propose an explicit multimodal routing framework for clinical prediction that enables interpretable, robust, and auditable reasoning across three EHR modalities: structured longitudinal variables (L), clinical notes (N), and chest X-rays (I). Our model constructs discrete unimodal, directional bimodal, and trimodal routes to capture both individual modality signals and asymmetric cross-modal interactions. To audit multimodal reasoning and assess robustness, we introduce inference-time route masking, which simulates missing modalities and reweights the remaining routes without retraining. We analyze changes in performance and routing weights under these scenarios to understand model decision-making. We evaluate our framework on multi-label phenotype prediction (K = 25) and binary ICU mortality prediction using trimodal patient stays from MIMIC-IV, revealing systematic differences in modality reliance across clinical condition groups. Overall, our framework offers a transparent, auditable, and practical approach to multimodal clinical prediction, providing interpretability, robustness, and insights into how different data sources drive model decisions.
CommentsAccepted at CHASE 2026. 12 pages, 6 figures, 4 tables
Journal ref2026 IEEE/ACM Conference on Connected Health: Applications, Systems and Engineering Technologies (CHASE), 2026
DOI：10.1109/CHASE69719.2026.00010
