Doctor AI: Predicting Clinical Events via Recurrent Neural Networks

Abstract

Leveraging large historical data in electronic health record (EHR), we developed Doctor AI, a generic predictive model that covers observed medical conditions and medication uses. Doctor AI is a temporal model using recurrent neural networks (RNN) and was developed and applied to longitudinal time stamped EHR data from 260K patients over 8 years. Encounter records (e.g. diagnosis codes, medication codes or procedure codes) were input to RNN to predict (all) the diagnosis and medication categories for a subsequent visit. Doctor AI assesses the history of patients to make multilabel predictions (one label for each diagnosis or medication category). Based on separate blind test set evaluation, Doctor AI can perform differential diagnosis with up to 79% recall@30, significantly higher than several baselines. Moreover, we demonstrate great generalizability of Doctor AI by adapting the resulting models from one institution to another without losing substantial accuracy.

Presented at 2016 Machine Learning and Healthcare Conference (MLHC 2016), Los Angeles, CA

References in corpus (8)

• Deep Learning in Neural Networks: An Overview
• Sequence to Sequence Learning with Neural Networks
• Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
• Unifying Visual-Semantic Embeddings with Multimodal Neural Language Models
• Theano: new features and speed improvements
• Visualizing and Understanding Recurrent Networks
• A Simple Way to Initialize Recurrent Networks of Rectified Linear Units
• Distilling Knowledge from Deep Networks with Applications to Healthcare Domain

Cited by in corpus (22)

• Deep Representation Learning of Patient Data from Electronic Health Records (EHR): A Systematic Review
• Recurrent Neural Networks for Multivariate Time Series with Missing Values
• RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mechanism
• Deep Learning in Pharmacogenomics: From Gene Regulation to Patient Stratification
• Multi-Label Classification of Patient Notes a Case Study on ICD Code Assignment
• Deep Learning Convolutional Networks for Multiphoton Microscopy Vasculature Segmentation
• Multi-task Prediction of Disease Onsets from Longitudinal Lab Tests
• Learning for Biomedical Information Extraction: Methodological Review of Recent Advances
• Deepr: A Convolutional Net for Medical Records
• MeDAL: Medical Abbreviation Disambiguation Dataset for Natural Language Understanding Pretraining
• Bayesian LSTMs in medicine
• Multi-layer Representation Learning for Medical Concepts
• Predictive Clinical Decision Support System with RNN Encoding and Tensor Decoding
• Mining Electronic Health Records: A Survey
• Personalized Event Prediction for Electronic Health Records
• CovidCare: Transferring Knowledge from Existing EMR to Emerging Epidemic for Interpretable Prognosis
• Multivariate Arrival Times with Recurrent Neural Networks for Personalized Demand Forecasting
• Long-term Blood Pressure Prediction with Deep Recurrent Neural Networks
• Memory and attention in deep learning
• A Simple and Interpretable Predictive Model for Healthcare
• Improving the Decision-Making Process of Self-Adaptive Systems by Accounting for Tactic Volatility
• Pastprop-RNN: improved predictions of the future by correcting the past