Resset: A Recurrent Model for Sequence of Sets with Applications to Electronic Medical Records

TL;DR

Resset is a novel recurrent neural network model designed to analyze sequences of medical record sets, capturing complex interactions between diseases and treatments to improve healthcare risk prediction and management.

Contribution

It introduces a new modeling framework for sequences of sets in electronic medical records, capturing interactions between diseases and treatments with an end-to-end recurrent model.

Findings

Resset outperforms existing methods in predicting hospital readmissions.

The model effectively recommends treatments and tracks disease progression.

Demonstrates promise on large-scale patient data for chronic diseases.

Abstract

Modern healthcare is ripe for disruption by AI. A game changer would be automatic understanding the latent processes from electronic medical records, which are being collected for billions of people worldwide. However, these healthcare processes are complicated by the interaction between at least three dynamic components: the illness which involves multiple diseases, the care which involves multiple treatments, and the recording practice which is biased and erroneous. Existing methods are inadequate in capturing the dynamic structure of care. We propose Resset, an end-to-end recurrent model that reads medical record and predicts future risk. The model adopts the algebraic view in that discrete medical objects are embedded into continuous vectors lying in the same space. We formulate the problem as modeling sequences of sets, a novel setting that have rarely, if not, been addressed. Within Resset, the bag of diseases recorded at each clinic visit is modeled as function of sets. The same hold for the bag of treatments. The interaction between the disease bag and the treatment bag at a visit is modeled in several, one of which as residual of diseases minus the treatments. Finally, the health trajectory, which is a sequence of visits, is modeled using a recurrent neural network. We report results on over a hundred thousand hospital visits by patients suffered from two costly chronic diseases -- diabetes and mental health. Resset shows promises in multiple predictive tasks such as readmission prediction, treatments recommendation and diseases progression.