Phenotyping of Clinical Time Series with LSTM Recurrent Neural Networks

TL;DR

This paper introduces a new approach using LSTM recurrent neural networks for multilabel classification of clinical diagnoses from variable-length time series data, demonstrating superior performance over existing baselines.

Contribution

It applies LSTM networks to clinical time series classification, showcasing improved accuracy and effectiveness in multilabel diagnosis prediction.

Findings

LSTM models outperform baseline methods on multiple metrics

Effective handling of variable-length clinical time series

Enhanced multilabel diagnosis classification accuracy

Abstract

We present a novel application of LSTM recurrent neural networks to multilabel classification of diagnoses given variable-length time series of clinical measurements. Our method outperforms a strong baseline on a variety of metrics.