AUTHOR=Riazati Sherveen , McGuirk Theresa E. , Perry Elliott S. , Sihanath Wandasun B. , Patten Carolynn 

TITLE=Absolute Reliability of Gait Parameters Acquired With Markerless Motion Capture in Living Domains

JOURNAL=Frontiers in Human Neuroscience

VOLUME=Volume 16 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/human-neuroscience/articles/10.3389/fnhum.2022.867474

DOI=10.3389/fnhum.2022.867474

ISSN=1662-5161

ABSTRACT=<p><bold>Purpose</bold>: To examine the between-day absolute reliability of gait parameters acquired with Theia3D markerless motion capture for use in biomechanical and clinical settings.</p><p><bold>Methods</bold>: Twenty-one (7 M,14 F) participants aged between 18 and 73 years were recruited in community locations to perform two walking tasks: self-selected and fastest-comfortable walking speed. Participants walked along a designated walkway on two separate days.Joint angle kinematics for the hip, knee, and ankle, for all planes of motion, and spatiotemporal parameters were extracted to determine absolute reliability between-days. For kinematics, absolute reliability was examined using: full curve analysis [root mean square difference (RMSD)] and discrete point analysis at defined gait events using standard error of measurement (SEM). The absolute reliability of spatiotemporal parameters was also examined using SEM and SEM%.</p><p><bold>Results</bold>: Markerless motion capture produced low measurement error for kinematic full curve analysis with RMSDs ranging between 0.96° and 3.71° across all joints and planes for both walking tasks. Similarly, discrete point analysis within the gait cycle produced SEM values ranging between 0.91° and 3.25° for both sagittal and frontal plane angles of the hip, knee, and ankle. The highest measurement errors were observed in the transverse plane, with SEM &gt;5° for ankle and knee range of motion. For the majority of spatiotemporal parameters, markerless motion capture produced low SEM values and SEM% below 10%.</p><p><bold>Conclusion</bold>: Markerless motion capture using Theia3D offers reliable gait analysis suitable for biomechanical and clinical use.</p>