Category: Technology
Objective: To objectively analyze motor symptoms with 2D RGB videos, which are easily accessible, attachment free and cost-effective.
Background: Quantitative measurement of parkinsonian symptoms is crucial in clinical practice and research. However, the current Unified PD Rating Scale (UPDRS) is based on semi-quantitative evaluation with high inter- and intra-rater variability. Sensor-based measurement has been widely studied but is limited for its accessibility.
Method: We analyzed 2D-RGB videos recording finger tapping and leg agility tests in 29 PD patients. The position of the finger tips and heels were tracked with deep-learning based tracking algorithm. We verified the tracking performance with the simultaneous application of accelerometer. Four parameters (mean amplitude, mean interpeak interval, amplitude variability and interpeak interval variability) were calculated from the position tracking.
Results: The performance of video-tracking was comparable with the result from the accelerometer (Squared R > 0.93). The video-tracking successfully captured variable aspects of limb bradykinesia which have a distinct correlation to the general parkinsonian motor symptoms and gait. Amplitude of the leg agility test was significantly correlated with UPDRS part II gait score (coefficient = -3.5, p = 0.0055) and gait speed (coefficient = -0.038, p = 0.029) and velocity of the leg agility test was correlated with UPDRS part III total score (coefficient = 1.7, p = 0.0024). Limb rigidity affected velocity (coefficient = 1.7, p = 0.0034) and rhythm (coefficient = 20, p = 0.0089) of leg agility test.
Conclusion: We showed that video-based tracking could objectively measure limb bradykinesia in PD patients. Video-based tracking is easily accessible, attachment free and cost-effective. Future studies with 3-dimensional analysis would further enhance the applicability of the technique.
To cite this abstract in AMA style:
J.H Shin, S.H Jeon, R. Kim, S.M Park, J.H Choi, J.N Ong, H.J Kim, B. Jeon. Objective measurement of limb bradykinesia using marker-less tracking algorithm with 2D-video of PD patients [abstract]. Mov Disord. 2020; 35 (suppl 1). https://www.mdsabstracts.org/abstract/objective-measurement-of-limb-bradykinesia-using-marker-less-tracking-algorithm-with-2d-video-of-pd-patients/. Accessed November 22, 2024.« Back to MDS Virtual Congress 2020
MDS Abstracts - https://www.mdsabstracts.org/abstract/objective-measurement-of-limb-bradykinesia-using-marker-less-tracking-algorithm-with-2d-video-of-pd-patients/