Validity and reliability of smartphone high-speed camera and Kinovea for velocity-based training measurement

Abstract

The aim of this study was to validate the combination of smartphone high-speed camera and motion analysis software Kinovea methodology (SHSC-Kinovea) to measure kinematic variables of velocity-based training during back squat exercises. Fifteen athletes were voluntarily recruited for the study (age 22.8 ± 2.9 years, height 182.9 ± 8.9 cm, body mass 79.5 ± 9.6 kg). High-speed video recordings with a smartphone at 240 fps were used against a criterion linear force transducer (LPT) for measuring displacement of the barbell (RB), mean velocity (MV), maximum velocity (Vmax) and concentric phase time (CPT). The intra-class correlations coefficient between LPT and SHSC-Kinovea showed almost perfect agreement for consistency (.992, .995, .997, .993) and absolute agreement (.975, .978, .980, .964) for RB, MV, Vmax and CPT, respectively. The mean differences between instruments were 1.11 mm for RB, 0.03 m/s for MV, 0.05 m/s for Vmax and 65.91 ms for CPT, all p < .001. Bland-Altman plots showed low systematic bias ± random error for RB: 1.11 ± 1.50 cm (r2: .006), MV: 0.03 ± 0.33 m/s (r2: .001) and Vmax: 65.91 ± 63.82 m/s (r2: .11), whereas Vmax showed overestimation for the high range of measures: 0.55 ± 0.42 m/s (r2: .31). Pearson’s product moment correlation coefficient showed almost perfect association between all variables: (r = .985 - .990) (p < .001). The SHSC-Kinovea methodology resulted in similar kinematic values than criterion so it can be considered as a trustworthy instrument for measuring velocity-based training.