Testing the Performance of an Innovative Markerless Technique for Quantitative and Qualitative Gait Analysis

Abstract

Gait abnormalities such as high stride and step frequency/cadence (SF—stride/second, CAD—step/second), stride variability (SV) and low harmony may increase the risk of injuries and be a sentinel of medical conditions. This research aims to present a new markerless video-based technology for quantitative and qualitative gait analysis. 86 healthy individuals (mead age 32 years) performed a 90 s test on treadmill at self-selected walking speed. We measured SF and CAD by a photoelectric sensors system; then, we calculated average ± standard deviation (SD) and within-subject coefficient of variation (CV) of SF as an index of SV. We also recorded a 60 fps video of the patient. With a custom-designed web-based video analysis software, we performed a spectral analysis of the brightness over time for each pixel of the image, that reinstituted the frequency contents of the videos. The two main frequency contents (F1 and F2) from this analysis should reflect the forcing/dominant variables, i.e., SF and CAD. Then, a harmony index (HI) was calculated, that should reflect the proportion of the pixels of the image that move consistently with F1 or its supraharmonics. The higher the HI value, the less variable the gait. The correspondence SF-F1 and CAD-F2 was evaluated with both paired t-Test and correlation and the relationship between SV and HI with correlation. SF and CAD were not significantly different from and highly correlated with F1 (0.893 ± 0.080 Hz vs. 0.895 ± 0.084 Hz, p < 0.001, r2 = 0.99) and F2 (1.787 ± 0.163 Hz vs. 1.791 ± 0.165 Hz, p < 0.001, r2 = 0.97). The SV was 1.84% ± 0.66% and it was significantly and moderately correlated with HI (0.082 ± 0.028, p < 0.001, r2 = 0.13). The innovative video-based technique of global, markerless gait analysis proposed in our study accurately identifies the main frequency contents and the variability of gait in healthy individuals, thus providing a time-efficient, low-cost means to quantitatively and qualitatively study human locomotion.

Quantitative and Qualitative Running Gait Analysis through an Innovative Video-Based Approach

“…In the present study an additional index (i.e., Synchron Index) was developed and proposed to describe the global gait quality. Both the Harmony and Synchrony Indexes are in arbitrary units [35]. The analysis starts with the subdivision of the video file into frames, allowing us to obtain a number (n) of individual JPEG images For each pixel of each image, the software computes the brightness via a brightness edi algorithm [35,43], followed by the application of a specific correction method called th Hanning window and a high-pass Butterworth filter with order 2 at 0.5 Hz to attenuat the component due to direct current [44].…”

“…The analysis starts with the subdivision of the video file into frames, allowing us to obtain a number (n) of individual JPEG images For each pixel of each image, the software computes the brightness via a brightness edi algorithm [35,43], followed by the application of a specific correction method called th Hanning window and a high-pass Butterworth filter with order 2 at 0.5 Hz to attenuat the component due to direct current [44]. After this preprocessing, the fast Fourier trans form (FFT) algorithm used in [35] is applied to the vector of brightnes values over time for each pixel to obtain its magnitude in the frequency domain. The resul is a power spectrum for each pixel that displays the magnitude as a function of frequency and permits us to identify the peaks of the dominant frequencies [35].…”

“…To ameliorate th process of peak frequency detection presented in Simoni et al (2020), power spectra fo each pixel are then averaged in order to obtain the Averaged Power Spectrum of the video and the signal is then further processed using the detrend MATLAB function in order to attenuate the effect of flickering or pink noise [44] (Figure 1). A simple peak detection is thus performed to extract the peaks with the highes power between 0 and 15 Hz, as indicated in [35]. The frequencies o the first two largest peaks are named F1 and F2 and have been demonstrated to corre spond, respectively, to stride frequency and cadence, expressed in Hz units, for walking gait [35].…”

“…Assuming that F2 corresponds to step frequency or cadence, the Graal system indi rectly estimates the spatial parameter step length as follows [35,45,46]: A simple peak detection is thus performed to extract the peaks with the highest power between 0 and 15 Hz, as indicated in [35]. The frequencies of the first two largest peaks are named F1 and F2 and have been demonstrated to correspond, respectively, to stride frequency and cadence, expressed in Hz units, for walking gait [35].…”

“…Assuming that F2 corresponds to step frequency or cadence, the Graal system indirectly estimates the spatial parameter step length as follows [35,45,46]:…”