• Corpus ID: 7539338

Learning to Detect Sepsis with a Multitask Gaussian Process RNN Classifier

  title={Learning to Detect Sepsis with a Multitask Gaussian Process RNN Classifier},
  author={Joseph D. Futoma and Sanjay Hariharan and Katherine A. Heller},
  booktitle={International Conference on Machine Learning},
We present a scalable end-to-end classifier that uses streaming physiological and medication data to accurately predict the onset of sepsis, a life-threatening complication from infections that has high mortality and morbidity. [] Key Method The Gaussian process is directly connected to a black-box classifier that predicts whether a patient will become septic, chosen in our case to be a recurrent neural network to account for the extreme variability in the length of patient encounters.

Figures from this paper

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

This work develops a flexible classifier that leverages streaming lab results, vitals, and medications to predict sepsis before it occurs, and substantially outperforms clinical baselines, and improves on a previous related model for detectingsepsis.

Early Recognition of Sepsis with Gaussian Process Temporal Convolutional Networks and Dynamic Time Warping

This empirical study proposes two novel approaches for the early detection of sepsis: a deep learning model and a lazy learner based on time series distances based on a temporal convolutional network that is embedded in a Multi-task Gaussian Process Adapter framework.

MGP-AttTCN: An interpretable machine learning model for the prediction of sepsis

This work proposes MGP-AttTCN: a joint multitask Gaussian Process and attention-based deep learning model to early predict the occurrence of sepsis in an interpretable manner and shows that the model outperforms the current state-of-the-art and presents evidence that different labelling heuristics lead to discrepancies in task difficulty.


An anonymised and appropriately processed data is taken in order to detect infections which imply future probability for sepsis and highlight the relation of specific clinical input features to the prediction outcome, possibly contributing to an improved, data-driven understanding of this multi-factorial dysfunction.

Machine learning for early detection of sepsis: an internal and temporal validation study

A novel deep learning model to detect sepsis was developed and validated and using the data elements and feature set, the modeling approach outperformed other machine learning methods and clinical scores.

Temporal Convolutional Networks and Dynamic Time Warping can Drastically Improve the Early Prediction of Sepsis

Two approaches for the early detection of sepsis are proposed: a new deep learning model (MGP-TCN) and a data mining model (DTW-KNN), which employs a temporal convolutional network as embedded in a Multitask Gaussian Process Adapter framework, making it directly applicable to irregularly spaced time series data.

Early Prediction of Sepsis: Using State-of-the-art Machine Learning Techniques on Vital Sign Inputs

This work presents a random forest-based ensemble machine learning technique to work on patient data, also called vital sign input, from ICU, and predicts the onset of sepsis based on the patient's information it received from previous recordings of vital sign inputs.

DBNet: a novel deep learning framework for mechanical ventilation prediction using electronic health records

A prediction model to estimate the probability of requiring mechanical ventilation for in-hospital patients at least 24 hours after their admission that outperforms the state-of-the-art baseline deep learning models in predicting the future requirement of mechanical ventilation.



A Multivariate Timeseries Modeling Approach to Severity of Illness Assessment and Forecasting in ICU with Sparse, Heterogeneous Clinical Data

This work evaluates the use of multivariate timeseries modeling with the multi-task Gaussian process (GP) models using noisy, incomplete, sparse, heterogeneous and unevenly-sampled clinical data, including both physiological signals and clinical notes to assess and forecast patient acuity.

Learning to Diagnose with LSTM Recurrent Neural Networks

This first study to empirically evaluate the ability of LSTMs to recognize patterns in multivariate time series of clinical measurements considers multilabel classification of diagnoses, and establishes the effectiveness of a simple LSTM network for modeling clinical data.

Using recurrent neural network models for early detection of heart failure onset

Deep learning models adapted to leverage temporal relations appear to improve performance of models for detection of incident heart failure with a short observation window of 12–18 months.

Multitask Gaussian Processes for Multivariate Physiological Time-Series Analysis

This work investigates MTGPs for physiological monitoring with synthetic data sets and two real-world problems from the field of patient monitoring and radiotherapy, and shows that the framework learned the correlation between physiological time series efficiently, outperforming the existing state of the art.

ForecastICU: A Prognostic Decision Support System for Timely Prediction of Intensive Care Unit Admission

Using a real-world dataset obtained from UCLA Ronald Reagan Medical Center, it is shown that ForecastICU can predict ICU admissions 9 hours before a physician's decision and performs consistently better than other state-of-the-art machine learning algorithms in terms of sensitivity, precision, and timeliness.

A Framework for Individualizing Predictions of Disease Trajectories by Exploiting Multi-Resolution Structure

A hierarchical latent variable model that individualizes predictions of disease trajectories is proposed and validated on the task of predicting the course of interstitial lung disease, a leading cause of death among patients with the autoimmune disease scleroderma.

A targeted real-time early warning score (TREWScore) for septic shock

Continuous sampling of data from the electronic health records and calculation of TREWScore may allow clinicians to identify patients at risk for septic shock and provide earlier interventions that would prevent or mitigate the associated morbidity and mortality.

Recurrent Neural Networks for Multivariate Time Series with Missing Values

Novel deep learning models are developed based on Gated Recurrent Unit, a state-of-the-art recurrent neural network that takes two representations of missing patterns, i.e., masking and time interval, and effectively incorporates them into a deep model architecture so that it not only captures the long-term temporal dependencies in time series, but also utilizes the missing patterns to achieve better prediction results.

Scalable Joint Modeling of Longitudinal and Point Process Data for Disease Trajectory Prediction and Improving Management of Chronic Kidney Disease

A novel probabilistic generative model is proposed that can provide individualized predictions of future disease progression while jointly modeling the pattern of related recurrent adverse events and gives superior performance in terms of both prediction offuture disease trajectories and of future serious events when compared to non-joint models.

A Non-parametric Learning Method for Confidently Estimating Patient's Clinical State and Dynamics

A non-parametric learning algorithm is constructed to estimate the clinical state of a patient by eliminating bias introduced by therapeutic intervention censoring, increasing the timeliness of state estimation while ensuring a sufficient accuracy, and the ability to detect anomalous clinical states.