<ul data-eligibleForWebStory="true"><li>Integrating multimodal Electronic Health Records (EHR) data has potential for predicting clinical outcomes.</li><li>Previous work focused on temporal interactions within samples and fusion of information, overlooking critical temporal patterns across patients.</li><li>Identifying temporal patterns like abnormal vital signs and corresponding textual descriptions is crucial.</li><li>A Cross-Modal Temporal Pattern Discovery (CTPD) framework is introduced to extract cross-modal temporal patterns efficiently.</li><li>CTPD uses shared initial temporal pattern representations and slot attention to generate temporal semantic embeddings.</li><li>A contrastive-based TPNCE loss is introduced for cross-modal alignment in learned patterns, along with two reconstruction losses.</li><li>Evaluations on 48-hour in-hospital mortality and 24-hour phenotype classification tasks using the MIMIC-III database highlight the superiority of the method.</li></ul>

CTPD: Cross-Modal Temporal Pattern Discovery for Enhanced Multimodal Electronic Health Records Analysis

Discover more