Objective: To assess the feasibility of extracting video-derived motor biomarkers from retrospective clinical recordings in Parkinson’s disease (PD) and quantify agreement with clinician-rated MDS-UPDRS Part III item scores, including inter-rater variability in remote scoring.

Background: The MDS-UPDRS Part III is the gold standard for PD motor assessment but requires in-person administration and is subject to inter-rater variability [1]. Automated video-based biomarker extraction offers a scalable alternative; however, feasibility and validity from retrospective clinical recordings remain understudied [2]. Existing trial recordings provide an efficient pathway for generating early alignment evidence and supporting prospective validation.

Method: Video recordings were collected from a previously conducted RCT examining effectiveness of photobiomodulation therapy (PBMt) combined with exercise in individuals with PD and were paired with MDS-UPDRS Part III scores [3]. Two clinicians independently rated recordings in a blinded manner. Video-derived measures include speech volume and blink rate. Speech volume and blink rate will be compared to corresponding MDS-UPDRS Part III items. Feasibility was evaluated by the proportion of videos yielding analyzable output. Agreement was assessed using exact and near agreement (within one point).

Results: Of 60 enrolled participants, 54 (90%) had usable video data, yielding 270 task-specific recordings across 5 motor domains, demonstrating strong feasibility for retrospective biomarker extraction. Agreement analyses will be presented at the conference.

Conclusion: High usability of retrospective clinical recordings supports the feasibility of automated motor biomarker extraction in PD. Agreement analyses are expected to clarify which biomarkers most closely align with clinician-rated MDS-UPDRS scores, informing prospective validation and standardization priorities.

References: [1] Goetz, C.G., Tilley, B.C., Shaftman, S.R., et al. Movement Disorder Society-sponsored revision of the Unified Parkinson’s Disease Rating Scale (MDS-UPDRS): Scale presentation and clinimetric testing results. Movement Disorders. 2008;23(15),2129–2170.

[2] Sibley, K.M., Girges, C., & Lam, T. Reliability and validity of video-based scoring of MDS-UPDRS Part III in Parkinson’s disease. Movement Disorders Clinical Practice. 2022;9(3), 343–350.

[3] Saltmarche, A. E., Hares, O., Bicknell, B., Liebert, A., Naeser, M., Ramachandran, S., Sykes, J., Togeretz, K., Namini, A., Heller, G. Z., & Herkes, G. (2025). Effectiveness of Photobiomodulation to Treat Motor and Non-Motor Symptoms of Parkinson’s Disease: A Randomised Clinical Trial with Extended Treatment. Journal of Clinical Medicine, 14(21), 7463. https://doi.org/10.3390/jcm14217463

« Back to 2026 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/retrospective-video-based-motor-assessment-feasibility-and-inter-rater-variability-in-remote-scoring/