Agreement between spatiotemporal parameters from a photoelectric system with different filter settings and high-speed...

The aim of this study was to determine the level of agreement between spatiotemporal gait characteristics from a photoelectric system with different filter settings and high-speed video analysis during running on a treadmill at comfortable velocity. Forty-nine runners performed a running protocol on a treadmill at comfortable velocity. Two systems were used to determine spatiotemporal parameters (i.e. contact time [CT], flight time [FT], step frequency [SF] and step length [SL]) during running: OptoGait system and high-speed video analysis at 1000 Hz. The collected data was re-filtered in the OptoGait software by using nine different settings (i.e. 0_0, 1_1, 2_2, 3_3, 3_4, 4_4, 4_5, 5_4 and 5_5), and compared to those obtained through video analysis. The Pearson correlation analysis revealed very large correlations (r > 0.9, p < 0.001) in CT, FT, SF and SL between both systems, regardless of the OptoGait's filter settings. The ICC reported an almost perfect association (ICC > 0.9) for both SL and SF regardless of the filter setting. However, large variations between filter settings according to the data from video analysis were reported in CT and FT (0_0, 1_1 and 2_2 filter settings obtained an association ICC > 0.9, whereas other filters obtained lower ICCs). Bland-Altman plots revealed small bias and error and no presence of heteroscedasticity of error for 1_1 setting. In conclusion, the filter setting for the OptoGait system should be considered to minimize the bias and error of spatiotemporal parameters measurement. For running on a treadmill, the 1_1 filter setting is recommended if gait parameters are to be compared to a high-speed video analysis (1000 Hz).