Clinical reliability and validity of a video-based markerless gait evaluation method

Area of Science:

• Biomedical Engineering
• Clinical Biomechanics
• Pediatric Gait Analysis

Background:

• Gait analysis is crucial for assessing pediatric conditions like developmental dysplasia of the hip (DDH).
• Traditional 3-dimensional gait analysis (3D GA) is accurate but resource-intensive.
• Markerless gait assessment offers a potentially more accessible alternative.

Purpose of the Study:

• To evaluate the reliability and validity of a markerless gait assessment system in children.
• To compare gait parameters derived from markerless video analysis with 3D GA.
• To determine the clinical feasibility of markerless gait analysis in pediatric populations.

Main Methods:

• Recruited 18 typically developing children and 10 with DDH.
• Utilized both 3D GA and a 2D markerless (2D ML) system to capture gait kinematics.
• Measurements were repeated within a 2-hour interval to assess reliability.

Main Results:

• Markerless gait parameters showed no significant differences between repeated measurements (P > 0.05).
• High intra-class correlation coefficients (ICC > 0.7) indicated good reliability.
• Excellent validity was observed when compared to 3D GA (ICC 0.835–0.957), with low standard error of measurement.

Conclusions:

• The markerless gait assessment method using human posture estimation is reliable and valid for children.
• This technology shows feasibility for practical clinical gait analysis.
• Markerless systems offer a promising, accessible approach for pediatric gait evaluation.