Self-supervised Representation Learning on Electronic Health Records with Graph Kernel Infomax

arXiv:2209.00655v2 Announce Type: replace
Abstract: Learning Electronic Health Records (EHRs) representation is a preeminent yet under-discovered research topic. It benefits various clinical decision support applications, e.g., medication outcome prediction or patient similarity search. Current approaches focus on task-specific label supervision on vectorized sequential EHR, which is not applicable to large-scale unsupervised scenarios. Recently, contrastive learning shows great success on self-supervised representation learning problems. However, complex temporality often degrades the performance. We propose Graph Kernel Infomax, a self-supervised graph kernel learning …

abstract applications arxiv benefits clinical cs.cy cs.lg current decision decision support ehr electronic electronic health records focus graph health kernel patient prediction records representation representation learning research search supervision support type