Human vs. Machine Learning Based Detection of Facial Weakness Using Video Analysis.
Publication/Presentation Date
1-1-2022
Abstract
BACKGROUND: Current EMS stroke screening tools facilitate early detection and triage, but the tools' accuracy and reliability are limited and highly variable. An automated stroke screening tool could improve stroke outcomes by facilitating more accurate prehospital diagnosis and delivery. We hypothesize that a machine learning algorithm using video analysis can detect common signs of stroke. As a proof-of-concept study, we trained a computer algorithm to detect presence and laterality of facial weakness in publically available videos with comparable accuracy, sensitivity, and specificity to paramedics.
METHODS AND RESULTS: We curated videos of people with unilateral facial weakness (
CONCLUSIONS: These preliminary results suggest that a machine learning algorithm using computer vision analysis can detect unilateral facial weakness in pre-recorded videos with an accuracy and sensitivity comparable to trained paramedics. Further research is warranted to pursue the concept of augmented facial weakness detection and external validation of this algorithm in independent data sets and prospective patient encounters.
Volume
13
First Page
878282
Last Page
878282
ISSN
1664-2295
Published In/Presented At
Aldridge, C. M., McDonald, M. M., Wruble, M., Zhuang, Y., Uribe, O., McMurry, T. L., Lin, I., Pitchford, H., Schneider, B. J., Dalrymple, W. A., Carrera, J. F., Chapman, S., Worrall, B. B., Rohde, G. K., & Southerland, A. M. (2022). Human vs. Machine Learning Based Detection of Facial Weakness Using Video Analysis. Frontiers in neurology, 13, 878282. https://doi.org/10.3389/fneur.2022.878282
Disciplines
Medicine and Health Sciences
PubMedID
35847210
Department(s)
Department of Medicine
Document Type
Article