Session Time: 1:45pm-3:15pm
Location: Hall 3FG
Objective: To develop a self-administered instrumented motor exam enabling remote measurement of Parkinson’s Disease (PD) signs.
Background: Current methods for PD assessment in clinical trials rely on rating scales including the Movement Disorder Society Unified Parkinson’s Disease Rating Scale motor exam (MDS-UPDRS-III). The MDS-UPDRS-III requires in-clinic administration and assessment by neurologists creating significant burden for patients in upcoming disease-modification drug trials, which are likely to benefit from more frequent longitudinal measures. Wearable inertial sensors may be deployed to obtain a lower burden, more precise measurement of standardized exam movements in the home.
Methods: Twenty-nine mild to moderate PD patients (Hoehn & Yahr no more than 3; Age: 68.83±7.04 years; females/males: 9/20), taking regular dopaminergic medication (Average Levodopa-equivalent dose 166.8±85.6 mg) had in-clinic MDS-UPDRS-III assessments performed by a neurologist during their ON and OFF states, while wearing sensors located at each limb (wrist and foot), sternum and lumbar regions. For each MDS-UPDRS-III item, features characterizing motor performance were derived from accelerometer and gyroscope data at relevant sensor locations. Variation of these features with the neurologist’s item score was quantified by the Kruskal-Wallis test. These features were used to build a step-wise linear mixed effects regression model to predict the item score. The model was tested by leave-one-subject-out cross validation.
Results: Considering the MDS-UPDRS-III toe tapping task as an example, kinematic features (amplitude and temporal variability), and signal features (spectral entropy), varied significantly with the item-score (p<0.009). The regression model achieved a root-mean-square-error of 0.64 and an R-square of 0.59. This approach is currently being applied to the other MDS-UPDRS-III items.
Conclusions: We have developed an analytical framework that leverages data from wearable sensors to predict MDS-UPDRS-III item scores. This approach may enable frequent home-based assessments of PD motor function and more precise quantification of medication effects. # This abstract will be presented at American Academy of Neurology (4/21-27, 2018)
To cite this abstract in AMA style:S. Patel, C. Demanuele, B. Ho, P. Wacnik, H. Zhang, T. Kangarloo, V. Ramos, S. Amato, D. Volfson, P. Bergethon, M. Erb. Developing a Self-Administered Instrumented Motor Exam for Home-based Parkinson’s Disease Assessment Using Wearable Sensors [abstract]. Mov Disord. 2018; 33 (suppl 2). https://www.mdsabstracts.org/abstract/developing-a-self-administered-instrumented-motor-exam-for-home-based-parkinsons-disease-assessment-using-wearable-sensors/. Accessed December 10, 2023.
« Back to 2018 International Congress
MDS Abstracts - https://www.mdsabstracts.org/abstract/developing-a-self-administered-instrumented-motor-exam-for-home-based-parkinsons-disease-assessment-using-wearable-sensors/