Doctor AI: Predicting Clinical Events via Recurrent Neural Networks

2015/11/18 by Edward Choi, Mohammad Taha Bahadori, Choi, Edward +7 · 16 citations
Biochemistry, Genetics and Molecular Biology · Computer Science · #Biomedical Text Mining and Ontologies #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning in Healthcare #Topic Modeling

paper · pdf · doi:10.48550/arxiv.1511.05942

openalex publication_date 2015/11/18 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

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.

Citations

• Unifying Visual-Semantic Embeddings with Multimodal Neural Language Models
• Learning to Execute
• Deep learning in neural networks: An overview
• A Novel Connectionist System for Unconstrained Handwriting Recognition
• Exact solutions to the nonlinear dynamics of learning in deep linear neural networks
• Sequence to Sequence Learning with Neural Networks

Cited by

• Generative Modeling of Clinical Time Series via Latent Stochastic Differential Equations
• FlowPath: Learning Data-Driven Manifolds with Invertible Flows for Robust Irregularly-sampled Time Series Classification
• CURENet: Combining Unified Representations for Efficient Chronic Disease Prediction
• Adversarial Spatio-Temporal Attention Networks for Epileptic Seizure Forecasting
• KAT-GNN: A Knowledge-Augmented Temporal Graph Neural Network for Risk Prediction in Electronic Health Records
• Modeling Irregular Astronomical Time Series with Neural Stochastic Delay Differential Equations
• TANDEM: Temporal Attention-guided Neural Differential Equations for Missingness in Time Series Classification
• Explainable AI for Infection Prevention and Control: Modeling CPE Acquisition and Patient Outcomes in an Irish Hospital with Transformers
• ASCENDgpt: A Phenotype-Aware Transformer Model for Cardiovascular Risk Prediction from Electronic Health Records
• Multi-Ontology Integration with Dual-Axis Propagation for Medical Concept Representation
• Structure-aware Hypergraph Transformer for Diagnosis Prediction in Electronic Health Records
• Memory and attention in deep learning
• The Physiological Deep Learner: First application of multitask deep learning to predict hypotension in critically ill patients
• Learning for Biomedical Information Extraction: Methodological Review of Recent Advances
• Prediction of Survival Outcomes under Clinical Presence Shift: A Joint Neural Network Architecture
• Quantifying surprise in clinical care: Detecting highly informative events in electronic health records with foundation models

Related

• A Comprehensive Survey of Electronic Health Record Modeling: From Deep Learning Approaches to Large Language Models
• Learning to Adapt Clinical Sequences with Residual Mixture of Experts
• MedGPT: Medical Concept Prediction from Clinical Narratives
• RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mechanism
• Scalable and accurate deep learning with electronic health records