Predicting healthcare trajectories from medical records: A deep learning approach

• Published in: Journal of biomedical informatics Vol. 69; pp. 218 - 229
• Main Authors: Pham, Trang, Tran, Truyen, Phung, Dinh, Venkatesh, Svetha
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.05.2017
• Subjects: Delivery of Health Care; Disease Progression; Electronic Health Records; Electronic medical records; Health Status; Healthcare processes; Humans; Irregular timing; Long-Short Term Memory; Neural Networks (Computer); Predictive medicine; Irregular timing; Healthcare processes; Electronic medical records; Predictive medicine; Long-Short Term Memory
• ISSN: 1532-0464; 1532-0480; 1532-0480
• DOI: 10.1016/j.jbi.2017.04.001

[Display omitted] •DeepCare – a model that reads EMRs and predicts future medical outcomes.•DeepCare is an end-to-end deep neural network built on Long Short-Term Memory.•Variable size admissions are represented as fixed-size vectors.•Irregularly timed events and interventions are modeled explicitly.•The results of 4 prediction tasks on diabetes and mental health cohorts are improved. Personalized predictive medicine necessitates the modeling of patient illness and care processes, which inherently have long-term temporal dependencies. Healthcare observations, stored 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 and models patient health state trajectories by the memory of historical records. Built on Long Short-Term Memory (LSTM), DeepCare introduces methods to handle irregularly timed events by moderating the forgetting and consolidation of memory. DeepCare also explicitly models 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 prediction accuracy.