Using wearable sensors and machine learning to assess upper limb function in Huntington's disease.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Digital Health

Background:

• Huntington's disease (HD) is a neurodegenerative disorder affecting limb function.
• Clinical assessments of HD symptoms are limited to specific settings.
• Wearable sensors can capture real-world data to complement clinical evaluations.

Purpose of the Study:

• To investigate the utility of wearable sensors for monitoring upper limb function in individuals with Huntington's disease.
• To apply deep learning and machine learning models to analyze sensor data and predict disease status and clinical scores.

Main Methods:

• A wrist-worn wearable sensor was used to collect data over 7 days from individuals with Huntington's disease (HD), prodromal HD (pHD), and controls (CTR).
• A deep learning model identified goal-directed hand movements, and kinematic features were analyzed.
• Statistical and machine learning models were employed to predict disease groups and clinical scores.

Main Results:

• Significant differences in goal-directed movement features were observed between HD, pHD, and CTR groups.
• Movement features strongly correlated with clinical scores.
• Classification models accurately distinguished between groups (67% balanced accuracy, 0.72 recall for HD), and regression models predicted clinical scores.

Conclusions:

• Wearable sensors combined with machine learning offer a promising approach for monitoring upper limb function in Huntington's disease.
• This technology can support early detection, remote patient monitoring, and assessment of treatment efficacy in clinical trials.