Bayesian Recurrent Framework for Missing Data Imputation and Prediction with Clinical Time Series

Page view(s)
15
Checked on Jun 07, 2024
Bayesian Recurrent Framework for Missing Data Imputation and Prediction with Clinical Time Series
Title:
Bayesian Recurrent Framework for Missing Data Imputation and Prediction with Clinical Time Series
Journal Title:
Machine Learning for Health (ML4H) at NeurIPS 2019
DOI:
Keywords:
Publication Date:
18 November 2019
Citation:
Abstract:
Real-world clinical time series data sets exhibit a high prevalence of missing values. Hence, there is an increasing interest in missing data imputation. Traditional statistical approaches impose constraints on the data-generating process and decouple imputation from prediction. Recent works propose recurrent neural network based approaches for missing data imputation and prediction with time series data. However, they generate deterministic outputs and neglect the inherent uncertainty. In this work, we introduce a unified Bayesian recurrent framework for simultaneous imputation and prediction on time series data sets. We evaluate our approach on two real-world mortality prediction tasks using the MIMIC-III and PhysioNet benchmark datasets. We demonstrate significant performance gains over state-of-the-art methods, and provide strategies to use the resulting probability distributions to better assess reliability of the imputations and predictions.
License type:
PublisherCopyrights
Funding Info:
The authors would like to acknowledge grant funding for Digital Health from the Science and Engineering Research Council, A*STAR, Singapore (Project No. A1818g0044).
Description:
ISBN:

Files uploaded:
File Size Format Action
There are no attached files.