Patients with anterior cruciate ligament deficiencies (ACLD) often exhibit altered knee kinematics and dynamics. Simulations that do not rely on tracking experimental or kinematic data can help elucidate the causal relationship between injuries and gait abnormalities. Unlike methods relying on experimental data tracking, predictive simulations, especially those based on the single shooting method, enable the exploration of movement patterns emerging from gait control policies, and thus well suited for studying the causal relationship of gait changes in patients. However, there are few studies on the direct generation of the ACLD gait based on single shooting predictive forward simulations. This study developed a cost function to obtain the motions of unilateral ACLD patients accurately. Using an asymmetric reflex-based controller to activate the muscles, the ACLD gait was simulated based on single shooting predictive forward dynamics. The gait kinematic and dynamic changes of ACLD patients were assessed based on comparisons with healthy individuals at different walking speeds, and the simulation results of their kinematics, dynamics, and muscle forces were examined. The simulation approach evaluated the differences in the knee flexion angles and moments observed in the ACLD leg. Additionally, the velocity distribution properties, and the role and limitation of the quadriceps avoidance strategy, were evaluated. Ultimately, the ability of the proposed method to simulate the ACLD gait was confirmed.
Keywords:
Cost function; Predictive simulation; Anterior cruciate ligament deficiency; Gait; Forward dynamics; Knee flexion moment; Optimization; Single shooting