Inter-session reliability of markerless lower extremity joint kinematics and kinetics for return-to-sport screening tasks

Evaluating movement quality in dynamic tasks is vital in return-to-sport (RTS) decision-making, yet marker-based motion capture (MoCap) can be complex. This study assessed the inter-session reliability of lower extremity joint kinematics and kinetics obtained using Theia3D, a markerless MoCap solution, during dynamic RTS screening tasks. Eighteen healthy participants performed six tasks over two sessions. Joint kinematics and kinetics were evaluated for absolute reliability between sessions, for discrete values and time-varying profiles. Test-retest reliability was evaluated using (integrated) intraclass correlation coefficient (ICC), standard error of measurement (SEM), and minimal detectable change (MDC). With few exceptions, integrated ICC values of joint angular displacements and moments showed moderate-to-excellent reliability (0.61-0.86 and 0.62-0.90 for angles in the sagittal and frontal plane, respectively; 0.54-0.88 and 0.51-0.85 for moments). SEM values for joint angles were low, indicating minimal session-to-session variation for kinematics. However, most MDC values were relatively high compared to expected intervention effects, particularly showing that kinetic measurements may not yet accurately detect clinically meaningful changes. Overall, these findings thus support the markerless MoCap's potential for tracking lower extremity dynamics during RTS screening, particularly for kinematics and sagittal plane kinetics, but highlight limited inter-session reliability for frontal plane kinetic parameters at the hip and ankle.