RAM-EHR: Retrieval Augmentation Meets Clinical Predictions on Electronic Health Records
This work addresses the challenge of enhancing predictive accuracy for clinical tasks using EHR data, representing an incremental improvement over existing knowledge-enhanced methods.
The paper tackles the problem of improving clinical predictions on Electronic Health Records by introducing RAM-EHR, a retrieval augmentation pipeline that integrates multiple knowledge sources, resulting in a 3.4% gain in AUROC and 7.2% gain in AUPR over previous baselines.
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}.