Markerless Video-Based Gait Analysis in People With Multiple Sclerosis

Gait analysis plays a crucial role in assessing mobility impairments and monitoring disease progression in individuals with Multiple Sclerosis (MS). Markerless, video-based methods offer a non-invasive, practical alternative to traditional marker-based systems, making them particularly suitable for clinical applications. This study employs a markerless video-based approach to extract spatio-temporal and kinematic parameters from 25 individuals with MS and 25 age- and sex-matched unimpaired controls. The MS cohort was divided into two subgroups based on the Expanded Disability Status Scale (EDSS): "high" disability ( $\textit {EDSS} \geq {3}$ ) and "low" disability ( $\textit {EDSS} \lt {3}$ ). Both normal and tandem gait patterns were evaluated. In normal gait, significant spatio-temporal and joint kinematic differences were observed between the high EDSS group and unimpaired controls, while the low EDSS group exhibited no notable deviations. In contrast, tandem gait analysis revealed significant differences in heel-to-toe distance between the low EDSS group and unimpaired controls, highlighting subtle changes that were undetectable in normal gait. These findings underscore the potential of video-based methods to enhance disease monitoring and guide targeted rehabilitation strategies in MS.