PT - JOURNAL ARTICLE AU - Kim, Heung-Youl TI - 3 Influence of input data errors on the inverse dynamics analysis of human locomotion: time-series changes in lower extremity joint moment and power during running AID - 10.1136/bjsports-2018-ISSSMC2018.3 DP - 2018 Dec 01 TA - British Journal of Sports Medicine PG - A1--A2 VI - 52 IP - Suppl 3 4099 - http://bjsm.bmj.com/content/52/Suppl_3/A1.3.short 4100 - http://bjsm.bmj.com/content/52/Suppl_3/A1.3.full SO - Br J Sports Med2018 Dec 01; 52 AB - The purpose of this study was to investigate the influence of inaccuracies between the center of pressure (CoP) recorded from a force platform and the body segment parameters (BSP) calculated from the different anthropometric models on resultant joint moment and power. Biomechanical data while running were obtained for one healthy male subject. Using inverse dynamics procedures, the lower extremity joint moment and power were computed for five successful trials. Thereafter, recalculations of the joint kinetic values were carried out under the following conditions: (1) five different BSP models (i.e., Ae (Japanese model), Vaughan, Chandler, Zatsiorsky, and De Leva), and (2) The CoP shifted in the anterior-posterior direction by ±5% and±10% from the actual location. Changes of the joint kinetics resulting from difference in BSP models showed similar changes in all joints. There was similar changes between the curves, and statistically significant difference in the results was not observed. On the other hand, the anterior shifts in the CoP caused an elevation in the ankle plantal flextion and hip extension moment curves and a decrease in the knee extension moment curves. The posterior shifts in the CoP had the opposite effects. But the times of occurrence of the peak values were not affected. When CoP was shifted in the anterior direction, the ankle and hip joint power generations were increased, and decreased in the posterior direction. In the knee joint power generation had the opposite results. The adjustments in the CoP produced changes in the observed joint moment variables of 14% and 34%, on average for the ±5% and±10% shifts, respectively. On the other hand, in the joint powers, greater power generation was observed with hip joint than other joint powers. Using a one-way repeated measures ANOVA, when CoP location was shifted in the anterior-posterior direction (±5% and±10%) were significantly associated with joint moment and power values (p<0.01). The joint moment and power are more sensitive to errors in the CoP calculated from a force plate measurements than in BSP parameters. Consequently, those results suggest that it is necessary to pay close attention to the level of experimental errors to ensure meaningful results when using inverse dynamics procedures in various fields.