Objective: To evaluate the classification accuracy of a computer vision-based feature extraction and classifier method to detect changes in finger tap bradykinesia associated with dopaminergic medication in Parkinson’s disease (PD).

Background: Clinical evaluation is the gold standard for assessing the effects of dopaminergic replacement therapy (DRT) in PD. Subtle changes in bradykinesia after DRT should be assessed by a trained eye, thus limiting inter-rater reliability and applicability.

Method: Standard 10-s video recordings of 419 hands performing finger tapping (103 PD patients), collected as part of the routine UPDRS-III, were analyzed.  Of these videos, 108 corresponded to evaluations in the “on” and “off” dopaminergic medication (from 66 patients), recorded in the same session. Videos had an associated severity rating from a trained clinician, following the rating criteria of UPDRS item 23 (range 0-4). The MediaPipe Deep Learning Library [1] was used to extract key-point coordinates of the fingers and arms for each frame. Three feature extraction methods were tested, (i) conventional kinematic feature design to capture key characteristics of item 23 as described in the UPDRS manual, (ii) massive higher-order feature extraction based on the HCTSA library [2], and (iii) a deep learning neural network based on a Multi-Layer Perceptron (MLP) classifier for time series and applied to perform supervised statistical learning against reference clinical diagnosis. An ordinal classifier, based on an extra-tree classifier, was trained and evaluated using 100 random splits with a stratification strategy. Furthermore, an extra-tree classifier was trained in the subset of ON-OFF measures in the same patient, to detect changes in the UPDRS item 23 score after DRT. Classification performance is reported using a balanced accuracy score.

Results: For estimating UPDRS ratings, our classifier achieved a balanced accuracy of 76.4% using conventional kinematic features, 72.3% using massive feature extraction, and 73.0% using the MLP classifier. The accuracy to detect changes in the UPDRS score after DRT was 77.8%.

Conclusion: Although it requires further optimization, our computer vision-based method can accurately quantify the ability of PD patients to perform the finger tapping item of UPDRS-III. Interestingly, our classifier also produces good accuracy to detect changes in this item after DRT.

References: [1] Lugaresi C et al. 2019 Mediapipe: A framework for building perception pipelines. arXiv preprint arXiv:1906.08172.
[2] Fulcher BD & Jones NS 2017. hctsa: A computational framework for automated time-series phenotyping using massive feature extraction. Cell systems, 5, 527-31.

« Back to 2022 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/computer-vision-can-detect-changes-in-bradykinesia-associated-with-dopaminergic-state-in-parkinsons-disease/