Two-Stage Representation Learning for Analyzing Movement Behavior Dynamics in People Living with Dementia
This study addresses the problem of analyzing movement behavior dynamics in people living with dementia, which is significant for healthcare professionals and caregivers who need to provide personalized care interventions.
The study tackled analyzing movement behavior dynamics in people living with dementia and resulted in a framework that can reveal key behavioral patterns correlated with clinical metrics. The framework has the potential to support cognitive status prediction and personalized care interventions.
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.