Document Type



This study compared body position on the bicycle using manual and automatically determined body segments during stationary cycling. High speed video (120 fps) was obtained from 14 cyclists using their own bikes on a stationary cycle trainer in a single session. Torso, hip, knee and ankle angles were measured in two positions of the crank (3 o’clock and 6 o’clock-dynamically) to determine body position on the bike. Significant differences (3-12°, d=0.38-1.92) were observed for all joints between the manual and automated method for both crank positions (except for the ankle at the 6 o’clock). Overall, the automated method overestimated torso and knee flexions and underestimated hip flexion and ankle dorsiflexion. The implications of these changes in terms of bike setup are to be determined in future research but it is unlikely that errors from the automated method will result in large changes in joint forces.