Generalised mixed effects models for changepoint analysis of biomedical time series data

TL;DR

This paper introduces a new method for detecting multiple changepoints in biomedical time series data using a generalized linear mixed model framework, applicable to digital health and neuroimaging data.

Contribution

It develops a novel changepoint detection approach that incorporates data structure via mixed models within a dynamic programming algorithm.

Findings

Effective in simulated scenarios

Successfully applied to fMRI data

Demonstrates utility in digital health monitoring

Abstract

Motivated by two distinct types of biomedical time series data, digital health monitoring and neuroimaging, we develop a novel approach for changepoint analysis that uses a generalised linear mixed model framework. The generalised linear mixed model framework lets us incorporate structure that is usually present in biomedical time series data. We embed the mixed model in a dynamic programming algorithm for detecting multiple changepoints in the fMRI data. We evaluate the performance of our proposed method across several scenarios using simulations. Finally, we show the utility of our proposed method on our two distinct motivating applications.