Missing data imputation for noisy time-series data and applications in healthcare

TL;DR

This paper compares various imputation methods for noisy, missing healthcare time-series data, demonstrating that traditional methods like MICE-RF outperform deep learning approaches in accuracy and can also provide denoising benefits.

Contribution

It provides a comprehensive comparison of imputation techniques for healthcare time series, highlighting the effectiveness of MICE-RF over deep learning methods in noisy, missing data scenarios.

Findings

MICE-RF outperforms deep learning methods in imputation accuracy.

Imputation methods can also serve as denoising tools.

Effective imputation improves downstream classification performance.

Abstract

Healthcare time series data is vital for monitoring patient activity but often contains noise and missing values due to various reasons such as sensor errors or data interruptions. Imputation, i.e., filling in the missing values, is a common way to deal with this issue. In this study, we compare imputation methods, including Multiple Imputation with Random Forest (MICE-RF) and advanced deep learning approaches (SAITS, BRITS, Transformer) for noisy, missing time series data in terms of MAE, F1-score, AUC, and MCC, across missing data rates (10 % - 80 %). Our results show that MICE-RF can effectively impute missing data compared to deep learning methods and the improvement in classification of data imputed indicates that imputation can have denoising effects. Therefore, using an imputation algorithm on time series with missing data can, at the same time, offer denoising effects.