RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mechanism

TLDR

In predictive modeling, accuracy and interpretability are key, yet choosing between complex black‑box models like RNNs and interpretable models such as logistic regression creates a trade‑off that is especially problematic in medicine. The authors developed RETAIN, a reverse‑time attention model for EHR data, to reconcile accuracy and interpretability. RETAIN uses a two‑level neural attention mechanism that processes EHR records in reverse chronological order, highlighting recent visits and key clinical variables. On a large EHR dataset, RETAIN matched RNNs in predictive accuracy and scalability while offering interpretability comparable to traditional models.

Abstract

Accuracy and interpretability are two dominant features of successful predictive models. Typically, a choice must be made in favor of complex black box models such as recurrent neural networks (RNN) for accuracy versus less accurate but more interpretable traditional models such as logistic regression. This tradeoff poses challenges in medicine where both accuracy and interpretability are important. We addressed this challenge by developing the REverse Time AttentIoN model (RETAIN) for application to Electronic Health Records (EHR) data. RETAIN achieves high accuracy while remaining clinically interpretable and is based on a two-level neural attention model that detects influential past visits and significant clinical variables within those visits (e.g. key diagnoses). RETAIN mimics physician practice by attending the EHR data in a reverse time order so that recent clinical visits are likely to receive higher attention. RETAIN was tested on a large health system EHR dataset with 14 million visits completed by 263K patients over an 8 year period and demonstrated predictive accuracy and computational scalability comparable to state-of-the-art methods such as RNN, and ease of interpretability comparable to traditional models.