Improving Clinical Predictions through Unsupervised Time Series Representation Learning

Abstract

In this work, we investigate unsupervised representation learning on medical time series, which bears the promise of leveraging copious amounts of existing unlabeled data in order to eventually assist clinical decision making. By evaluating on the prediction of clinically relevant outcomes, we show that in a practical setting, unsupervised representation learning can offer clear performance benefits over end-to-end supervised architectures. We experiment with using sequence-to-sequence (Seq2Seq) models in two different ways, as an autoencoder and as a forecaster, and show that the best performance is achieved by a forecasting Seq2Seq model with an integrated attention mechanism, proposed here for the first time in the setting of unsupervised learning for medical time series.

arXiv commentMachine Learning for Health (ML4H) Workshop at NeurIPS 2018 arXiv:1811.07216

References in corpus (6)

• Sequence to Sequence Learning with Neural Networks
• DRAW: A Recurrent Neural Network For Image Generation
• Sequence-to-Sequence Models Can Directly Translate Foreign Speech
• The Use of Autoencoders for Discovering Patient Phenotypes
• Recurrent Neural Network-Based Semantic Variational Autoencoder for Sequence-to-Sequence Learning
• Canonical Correlation Analysis for Analyzing Sequences of Medical Billing Codes

Cited by in corpus (9)

• Unsupervised Representation Learning for Time Series with Temporal Neighborhood Coding
• CLOCS: Contrastive Learning of Cardiac Signals Across Space, Time, and Patients
• Medical Time Series Classification with Hierarchical Attention-based Temporal Convolutional Networks: A Case Study of Myotonic Dystrophy Diagnosis
• Convolution-enhanced Evolving Attention Networks
• Deep Bayesian Gaussian Processes for Uncertainty Estimation in Electronic Health Records
• Unsupervised Representation for EHR Signals and Codes as Patient Status Vector
• Target-Embedding Autoencoders for Supervised Representation Learning
• A Self-Supervised Framework for Function Learning and Extrapolation
• Piece-wise Matching Layer in Representation Learning for ECG Classification