Video-Based Pairwise Comparison: Enabling the Development of Automated Rating of Motor Dysfunction in Multiple...

Area of Science:

• Neurology
• Biomedical Engineering
• Clinical Assessment

Background:

• Multiple sclerosis (MS) significantly impacts motor function, necessitating reliable assessment tools.
• Current methods for evaluating motor dysfunction in MS may lack granularity and consistency.
• Video analysis offers potential for objective and detailed motor function evaluation.

Purpose of the Study:

• To assess the feasibility, reliability, granularity, and convergent validity of a video-based pairwise comparison technique for rating motor dysfunction in MS patients.
• To explore the utility of algorithmic support for automated rating of motor deficits.
• To establish a more precise method for quantifying motor ability in individuals with MS.

Main Methods:

• A cross-sectional cohort study involving 42 patients with MS from two specialist university medical centers.
• Neurologists rated standardized patient movement videos (e.g., finger-to-nose test) using a pairwise comparison approach.
• The TrueSkill algorithm calculated comparative assessment scores, analyzed for intrarater and interrater reliability and convergent validity with the Expanded Disability Status Scale (EDSS).

Main Results:

• Excellent intrarater reliability (median ratio of agreement ≥0.87) and good-to-excellent interrater reliability (ICC: 0.89; 0.71) were observed.
• Comparative assessment scores showed high correlation with Neurostatus-EDSS equivalents (ρ=0.78, P<.001; ρ=0.91, P<.05).
• The method demonstrated high granularity, capturing subtle differences in patient performance.

Conclusions:

• Video-based pairwise comparison is a feasible and reliable method for assessing motor dysfunction in MS.
• This technique provides a fine-grained capture of clinical judgment on neurological performance.
• It holds potential for developing a consistent, quantified metric of motor ability in multiple sclerosis.