Category: Parkinson’s Disease: Clinical Trials
Objective: To assess hypomimia longitudinally in individuals with Parkinson’s disease (PD) using a video analytics tool.
Background: Recent advancements in video analytics – the use of algorithms to determine a video’s spatial and temporal content  – may facilitate objective evaluation of PD symptoms and progression.
Method: Previously, Ali et al. used the Facial Action Coding System (FACS) to assess features of hypomimia from videos of participants with and without PD independently completing facial mimicry tasks on a publicly available website (www.parktest.net) . FACS translates movement of specific facial muscles into discrete action units (AUs) . However, Ali et al.’s participant cohort only included a small proportion (10.1%) of individuals with PD . In this 24-month observational study with in-person and remote components, we applied similar computational methods to a formally screened cohort that used the video analytics tool at baseline, month 6, month 12, and month 24 visits. Notably, this cohort included a much larger proportion of participants with PD (70.0%) [table1].
Results: Refer to [table1] for participant characteristics. While completing the tool’s smile mimicry task at in-person baseline visits, participants with PD experienced less intense movement of the orbicularis oculi and pars orbitalis when raising the cheeks, and of the zygomatic major when pulling the corner of the lip [table2]  compared to controls.
Conclusion: Our findings are consistent with those from Ali et al.’s work , and may demonstrate the feasibility of using the Facial Action Coding System to evaluate hypomimia in individuals with PD. Additional efforts to deploy this tool for use by diverse cohorts (for example, genetically at-risk populations) are warranted. Collecting data longitudinally with this tool may assist with PD detection and provide insights into PD progression.
References:  A Guide to Video Analytics: Applications and Opportunities. Tryolabs. Published October 29, 2019. https://tryolabs.com/guides/video-analytics-guide
 Ali, M.R., Myers, T., Wagner, E. et al. Facial expressions can detect Parkinson’s disease: preliminary evidence from videos collected online. npj Digit. Med. 4, 129 (2021). https://doi.org/10.1038/s41746-021-00502-8
 Facial Action Coding System. Paul Ekman Group. https://www.paulekman.com/facial-action-coding-system
 Farnsworth B. Facial Action Coding System (FACS) – A Visual Guidebook. iMotions. Published February 18, 2019. https://imotions.com/blog/facial-action-coding-system
 Baltrusaitis T. Action Units. GitHub. Published June 30, 2019. https://github.com/TadasBaltrusaitis/OpenFace/wiki/Action-Units
To cite this abstract in AMA style:P. Yang, M. Islam, A. Abdelkader, W. Rahman, M. Pawlik, S. Jensen-Roberts, E. Waddell, T. Myers, J. Soto, E. Hartman, E. Nnadika, R. Wilson, K. Lizarraga, C. Tarolli, R. Schneider, E. Dorsey, J. Adams, E. Hoque. Longitudinal evaluation of hypomimia in individuals with Parkinson’s disease using a video analytics tool [abstract]. Mov Disord. 2022; 37 (suppl 2). https://www.mdsabstracts.org/abstract/longitudinal-evaluation-of-hypomimia-in-individuals-with-parkinsons-disease-using-a-video-analytics-tool/. Accessed September 22, 2023.
« Back to 2022 International Congress
MDS Abstracts - https://www.mdsabstracts.org/abstract/longitudinal-evaluation-of-hypomimia-in-individuals-with-parkinsons-disease-using-a-video-analytics-tool/