Exploring the validity of smartphone based single camera markerless motion capture technology to quantify knee range of motion in patients with knee osteoarthritis

BACKGROUND: Single camera markerless motion capture technology offers a potential means of assessing joint function in patients with musculoskeletal disorders/diseases. The aim of this study was to determine the validity and within-session reliability of sagittal plane knee joint kinematics quantified using the Deep Vision platform in patients with knee osteoarthritis. METHODS: Sagittal plane knee joint kinematics were quantified using a 3D motion capture system, the msk.ai Deep Vision platform and Kinovea in fifteen patients with knee osteoarthritis. Bland Altman plots with 95 % limits of agreement were used to assess validity and within-session reliability, with mean differences and limits of agreement explored relative to ± 5° and 10° clinically meaningful thresholds, respectively. This is based on the assumption that changes in joint angles greater than 5° are clinically meaningful. RESULTS: Mean differences and 95 % limits of agreement were within the clinically meaningful thresholds when using the Deep Vision platform to quantify peak knee flexion and extension, and range of motion. Mean differences were within the clinically meaningful threshold for Kinovea based assessments, but 95 % limits of agreement exceeded the ± 10° clinically meaningful threshold for peak flexion and range of motion. All assessment methods displayed mean differences and 95 % confidence intervals within the clinically meaningful thresholds on average when comparing across repetitions to quantify within-session reliability. CONCLUSION: The findings of the study demonstrate that the msk.ai Deep Vision platform provides a valid and reliable means of quantifying peak knee flexion, extension and range of motion. CONTRIBUTION OF THE PAPER.