Validity of a Smartphone App Using Artificial Intelligence for the Real-Time Measurement of Barbell Velocity in the Bench Press Exercise
Carlos Balsalobre‐Fernández, Jiaqing Xu, Paul Jarvis, Steve Thompson, Kyran Tannion, Chris Bishop
Abstract
ABSTRACT: Balsalobre-Fernández, C, Xu, J, Jarvis, P, Thompson, S, Tannion, K, and Bishop, C. Validity of a smartphone app using artificial intelligence for the real-time measurement of barbell velocity in the bench press exercise. J Strength Cond Res 37(12): e640-e645, 2023-The purpose of this study was to explore the validity and within-session reliability of the newly developed My Jump Lab application (app), which uses artificial intelligence techniques to monitor barbell velocity in real time. Twenty-seven sport science students performed 5 repetitions at 50 and 75% of their self-reported bench press 1 repetition maximum (1RM) during a single testing session, whereas barbell velocity was concurrently measured using the app (installed on an iPhone 12 Pro) and the GymAware linear position transducer (LPT). A very high correlation was observed between devices at each loading condition (50% 1RM: r = 0.90 [0.82-0.97]; 75% 1RM: r = 0.92 [0.86-0.98]). Results showed trivial differences between the app and LPT at both 50% 1RM (g = -0.06) and 75% 1RM (g = -0.12). Bland-Altman analysis showed a bias estimate of -0.010 m·s-1 and -0.026 m·s-1 for the 50 and 75% 1RM, respectively. Finally, similar levels of reliability, as revealed by the coefficient of variation, were observed for both devices (50% 1RM: LPT = 6.52%, app = 8.17%; 75% 1RM: LPT = 12.10%, app = 13.55%). Collectively, the findings of this study support the use of My Jump Lab for the measurement of real-time barbell velocity in the bench press exercise.