An Improved Multi-Output Gaussian Process RNN with Real-Time Validation for Early Sepsis Detection

Abstract

Sepsis is a poorly understood and potentially life-threatening complication that can occur as a result of infection. Early detection and treatment improves patient outcomes, and as such it poses an important challenge in medicine. In this work, we develop a flexible classifier that leverages streaming lab results, vitals, and medications to predict sepsis before it occurs. We model patient clinical time series with multi-output Gaussian processes, maintaining uncertainty about the physiological state of a patient while also imputing missing values. The mean function takes into account the effects of medications administered on the trajectories of the physiological variables. Latent function values from the Gaussian process are then fed into a deep recurrent neural network to classify patient encounters as septic or not, and the overall model is trained end-to-end using back-propagation. We train and validate our model on a large dataset of 18 months of heterogeneous inpatient stays from the Duke University Health System, and develop a new "real-time" validation scheme for simulating the performance of our model as it will actually be used. Our proposed method substantially outperforms clinical baselines, and improves on a previous related model for detecting sepsis. Our model's predictions will be displayed in a real-time analytics dashboard to be used by a sepsis rapid response team to help detect and improve treatment of sepsis.

Presented at Machine Learning for Healthcare 2017, Boston, MA

References in corpus (4)

• Auto-Encoding Variational Bayes
• A Recurrent Latent Variable Model for Sequential Data
• Learning to Diagnose with LSTM Recurrent Neural Networks
• Sparse Multi-Output Gaussian Processes for Medical Time Series Prediction

Cited by in corpus (19)

• MGP-AttTCN: An Interpretable Machine Learning Model for the Prediction of Sepsis
• Dropout Feature Ranking for Deep Learning Models
• Explainable artificial intelligence model to predict acute critical illness from electronic health records
• "The Human Body is a Black Box": Supporting Clinical Decision-Making with Deep Learning
• Early detection of sepsis utilizing deep learning on electronic health record event sequences
• Temporal-Clustering Invariance in Irregular Healthcare Time Series
• Counterfactual Normalization: Proactively Addressing Dataset Shift and Improving Reliability Using Causal Mechanisms
• Sequential Gaussian Processes for Online Learning of Nonstationary Functions
• EVA: Generating Longitudinal Electronic Health Records Using Conditional Variational Autoencoders
• Medical Dead-ends and Learning to Identify High-risk States and Treatments
• Modeling Irregularly Sampled Clinical Time Series
• Integrating Physiological Time Series and Clinical Notes with Deep Learning for Improved ICU Mortality Prediction
• Dynamic Measurement Scheduling for Event Forecasting using Deep RL
• Deep Mixed Effect Model using Gaussian Processes: A Personalized and Reliable Prediction for Healthcare
• The Consequences of the Framing of Machine Learning Risk Prediction Models: Evaluation of Sepsis in General Wards
• Dynamic Measurement Scheduling for Adverse Event Forecasting using Deep RL
• UNITE: Uncertainty-based Health Risk Prediction Leveraging Multi-sourced Data
• Nonstationary Multivariate Gaussian Processes for Electronic Health Records
• Patient-Specific Effects of Medication Using Latent Force Models with Gaussian Processes