Alterations in gait patterns often indicate health status, and their analysis enables the diagnosis and assessment of various health conditions. This study aimed to develop a noncontact gait analysis system using red, green, and blue-depth (RGB-D) cameras and to evaluate its potential clinical applicability. A single case study was conducted to assess changes in the gait patterns of a patient with stroke before and after the application of an ankle-foot orthosis. Twenty walking trials were recorded to evaluate the key gait parameters. The custom RGB-D camera-based gait analysis system demonstrated the potential to rapidly quantify key gait parameters in the patient. Compared with normative data, it effectively identified characteristic stroke-related gait impairments such as shorter step lengths and slower gait speeds. However, the intraclass correlation coefficient analysis indicated low measurement reliability. Although the stance time and minimum knee angle on the left and right sides exceeded the standard error of measurement (SEM), no changes exceeded the minimal detectable change (MDC) criteria. Moreover, other gait parameters did not show significant changes beyond SEM or MDC, limiting the interpretability of the results. Therefore, further technological developments and data collection are required to improve test-retest reliability and sensitivity to change.
Chang et al. (Sat,) studied this question.