DeepCare: A Deep Dynamic Memory Model for Predictive Medicine

TL;DR

DeepCare is a novel deep neural network model that captures long-term patient health trajectories from irregular medical records to improve disease prediction and intervention planning.

Contribution

It introduces a deep dynamic memory network based on LSTM with time parameterizations and multiscale pooling for modeling patient health over time.

Findings

Improved accuracy in disease progression modeling.

Effective in predicting future medical risks.

Demonstrated on diabetes and mental health cohorts.

Abstract

Personalized predictive medicine necessitates the modeling of patient illness and care processes, which inherently have long-term temporal dependencies. Healthcare observations, recorded in electronic medical records, are episodic and irregular in time. We introduce DeepCare, an end-to-end deep dynamic neural network that reads medical records, stores previous illness history, infers current illness states and predicts future medical outcomes. At the data level, DeepCare represents care episodes as vectors in space, models patient health state trajectories through explicit memory of historical records. Built on Long Short-Term Memory (LSTM), DeepCare introduces time parameterizations to handle irregular timed events by moderating the forgetting and consolidation of memory cells. DeepCare also incorporates medical interventions that change the course of illness and shape future medical risk. Moving up to the health state level, historical and present health states are then aggregated through multiscale temporal pooling, before passing through a neural network that estimates future outcomes. We demonstrate the efficacy of DeepCare for disease progression modeling, intervention recommendation, and future risk prediction. On two important cohorts with heavy social and economic burden -- diabetes and mental health -- the results show improved modeling and risk prediction accuracy.