Objective: To determine whether changes in EDA and HRV are associated with motor fluctuations in patients with PD

Background: EDA and HRV are affected in individuals with PD. New wrist-worn sensors allow EDA and HRV to be monitored longitudinally, which may allow inference of sympathetic flow in PD patients, which in turn, may allow prediction of pending motor fluctuations. Ideally, detectable subtle changes may occur before motor manifestations, and may serve as a reminder for taking medications.

Method: A total of 10 patients with Parkinson’s disease were recruited from the Movement Disorder Clinic at UBC. A data-driven method was employed to predict the ON/OFF status of patients wearing the Empatica E4 wrist-worn sensor using their EDA and HRV data as well as their activity logs. Data was divided into 10 min epochs (5 min before and 5 min after the time tagged by the patient) and, based on the activity logs were classified as “ON” or “OFF”. The R package of CARET was used to predict motor status based on EDA and HRV values.

Results: Penalized Discriminant Analysis (PDA) validated by ten-fold cross validation could classify the EDA data with 63% accuracy (SD=0.04) and HRV data with 61% accuracy (SD=0.004). Combining features of HRV and EDA using PDA could classify the data with 70% accuracy (SD=0.06).

Conclusion: There is an association between motor fluctuations in patients with PD and EDA/HRV data, however more study is required to determine if this can be used as a practical predictive tool.

« Back to 2019 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/ambulatory-monitoring-of-electrodermal-activity-eda-and-heart-rate-variability-hrv-for-prediction-of-motor-fluctuations-in-parkinsons-disease-pd/