Global Contrastive Training for Multimodal Electronic Health Records with Language Supervision

arXiv:2404.06723v1 Announce Type: new
Abstract: Modern electronic health records (EHRs) hold immense promise in tracking personalized patient health trajectories through sequential deep learning, owing to their extensive breadth, scale, and temporal granularity. Nonetheless, how to effectively leverage multiple modalities from EHRs poses significant challenges, given its complex characteristics such as high dimensionality, multimodality, sparsity, varied recording frequencies, and temporal irregularities. To this end, this paper introduces a novel multimodal contrastive learning framework, specifically focusing on medical time series and clinical …

abstract arxiv challenges cs.cl cs.lg deep learning electronic electronic health records global health language modern multimodal multiple patient personalized records scale supervision temporal through tracking training type