Category: Huntington's Disease
Objective: To develop and evaluate machine learning models to identify patients with Huntington’s disease (HD) and to predict their symptom severity related to upper limb function using outcomes derived from a wearable sensor worn by the participants during activities of daily living.
Background: HD is a neurodegenerative disorder characterized by progressive motor, cognitive, and behavioral symptoms. Current clinical assessment methods for HD are limited in sensitivity and provide only a snapshot of patient disability during clinic visits. Remote monitoring through wearable sensors offers a promising approach to capture continuous data on upper limb function, potentially enabling early intervention and more comprehensive disease management.
Method: We recruited 18 individuals with manifest HD(mean age 49.9 ± 11 SD), 7 individuals with prodromal HD (mean age 34.6 ± 12.9), and 11 controls (mean age 55 ± 14 SD). Participants wore a PAMSys ULM wrist sensor for seven consecutive days on their dominant hand to monitor hand goal-directed movements during activities of daily living. Machine learning techniques were employed to classify HD, prodromal HD, and controls and regression analysis was used to predict clinical assessment scores such as such as UHDRS functional, UHDRS motor, UHDRS total functional capacity and UHDRS upper limb function subdomain scores.
Results: Classification achieved a balanced accuracy of 0.62 (chance level 0.33). Prediction of clinical scores explained variance ranged from 73% to 43%.
Conclusion: Our study demonstrates the feasibility of using wearable sensors and machine learning to track upper limb function in individuals with Huntington’s disease, including those in prodromal stages. The classification model achieved a balanced accuracy significantly above chance level, and regression analysis revealed good predictability of clinical assessments. These findings suggest that remote monitoring with wearable technology holds promise for improving the early detection and management of Huntington’s disease and can be used as sensitive tools in HD clinical trials and care.
To cite this abstract in AMA style:
J. Adams, A. Nunes, R. Mishra, E. Dorsey, A. Vaziri. Remote monitoring of upper limb function in Huntington’s disease using a wearable sensor [abstract]. Mov Disord. 2024; 39 (suppl 1). https://www.mdsabstracts.org/abstract/remote-monitoring-of-upper-limb-function-in-huntingtons-disease-using-a-wearable-sensor/. Accessed October 7, 2024.« Back to 2024 International Congress
MDS Abstracts - https://www.mdsabstracts.org/abstract/remote-monitoring-of-upper-limb-function-in-huntingtons-disease-using-a-wearable-sensor/