RAM-EHR: Retrieval Augmentation Meets Clinical Predictions on Electronic Health Records

TL;DR

RAM-EHR introduces a retrieval-augmented approach that enhances clinical predictions on EHRs by integrating multiple knowledge sources and co-training with local data, leading to significant performance improvements.

Contribution

It proposes a novel retrieval augmentation pipeline for EHR predictions that combines knowledge retrieval with co-trained models, addressing complex medical concept names.

Findings

3.4% gain in AUROC over baselines

7.2% gain in AUPR over baselines

Effective use of summarized knowledge improves clinical prediction accuracy

Abstract

We present RAM-EHR, a Retrieval AugMentation pipeline to improve clinical predictions on Electronic Health Records (EHRs). RAM-EHR first collects multiple knowledge sources, converts them into text format, and uses dense retrieval to obtain information related to medical concepts. This strategy addresses the difficulties associated with complex names for the concepts. RAM-EHR then augments the local EHR predictive model co-trained with consistency regularization to capture complementary information from patient visits and summarized knowledge. Experiments on two EHR datasets show the efficacy of RAM-EHR over previous knowledge-enhanced baselines (3.4% gain in AUROC and 7.2% gain in AUPR), emphasizing the effectiveness of the summarized knowledge from RAM-EHR for clinical prediction tasks. The code will be published at \url{https://github.com/ritaranx/RAM-EHR}.