Two-Stage Representation Learning for Analyzing Movement Behavior Dynamics in People Living with Dementia

TL;DR

This paper introduces a two-stage self-supervised learning framework that transforms high-frequency movement data from dementia patients into interpretable low-rank representations, aiding in behavioral analysis and clinical prediction.

Contribution

It proposes a novel two-stage approach combining text encoding and PageRank-based state transition analysis for dementia patient activity data.

Findings

Enhanced interpretability of behavioral data

Improved clustering and transition analysis

Potential for cognitive status prediction

Abstract

In remote healthcare monitoring, time series representation learning reveals critical patient behavior patterns from high-frequency data. This study analyzes home activity data from individuals living with dementia by proposing a two-stage, self-supervised learning approach tailored to uncover low-rank structures. The first stage converts time-series activities into text sequences encoded by a pre-trained language model, providing a rich, high-dimensional latent state space using a PageRank-based method. This PageRank vector captures latent state transitions, effectively compressing complex behaviour data into a succinct form that enhances interpretability. This low-rank representation not only enhances model interpretability but also facilitates clustering and transition analysis, revealing key behavioral patterns correlated with clinicalmetrics such as MMSE and ADAS-COG scores. Our findings demonstrate the framework's potential in supporting cognitive status prediction, personalized care interventions, and large-scale health monitoring.