Related Experiment Video
Updated: Aug 25, 2025
Clinical-oriented Three-dimensional Gait Analysis Method for Evaluating Gait Disorder
Published on: March 4, 2018
Usefulness of the Kinect-V2 System for Determining the Global Gait Index to Assess Functional Recovery after Total
Hong Man Cho1, Jangwon Seon1, JiYeon Park1
1Department of Orthopedic Surgery, Gwangju Veterans Hospital, Gwangju, South Korea.
Objective:
The Korean Knee Society (KKS) score is used for functional evaluation during follow-up after total knee arthroplasty (TKA), but it is time-consuming to measure and is limited by its subjective nature. We investigated whether the global gait asymmetry index (GGA) that can be obtained using the Kinect-V2 system could overcome the KKS limitations.
Methods:
Forty-three patients who underwent TKA from January 2019 to December 2019 were included. Postoperatively, regular follow-up was performed at 2, 4, 6, 8, and 12 weeks, and at 4, 6, and 12 months. At each follow-up visit, the KKS was measured, and the walking path was followed with six Kinect-V2 systems. After allowing the participants to walk naturally, the range of motion of each joint of the lower extremity and GGA were obtained. Changes in the KKS and GGA scores and measurement times were investigated until the final follow-up. A statistical model was made to predict the KKS from the GGA score using data at all observed time points, and analysis of variance (ANOVA) with Turkey's post-hoc tests and Pearson correlation tests were used for evaluation.
Results:
Both the KKS and GGA scores improved significantly from 4 weeks postoperatively until the final follow-up. The measurement time was significantly shorter for the GGA (9.3 ± 1.4 min) than for the KKS (32.4 ± 9.2 min; P < 0.001) score. The predicted and actual KKS values clustered close to a straight line on the scatter plot, but the prediction was less accurate in the initial stage (2 weeks post-surgery) than at later time points. The mean absolute error (MAE) and root mean square of the error (RMSE) were considered to be poorly predicted in the initial stage (8 weeks post-surgery) compared to the later time-points (MAE ≥ 5 and RMSE ≥ 6 for 8 weeks post-surgery).
Conclusion:
In the early phase after knee joint surgery (up to 12 weeks post-surgery), the GGA index does not predict the KKS well. However, after this time point, the GGA index can be simply measured in the outpatient department and may be able to replace the KKS. Thus, evaluation of the GGA index using the Kinect-V2 may be a useful method to evaluate functional recovery in the outpatient clinic after knee joint surgery.
