EMG-driven modeling: Forward simulation and knee-ligament loading simulation

The human neuromusculoskeletal system is complicated and different muscles are finely coordinated to accomplish various tasks. Electromyography (EMG) includes real-time information about the electrical activity of a specific muscle. Different EMG-driven biomechanical models have been developed to estimate muscle forces. They can implicitly account for a subject's individual muscle activation patterns and help reveal underlying neuromuscular control strategies. However, these EMG-driven models have not been applied to study patients with neurological disorder, or finish forward simulation and knee-ligament loading simulation.The first study of this dissertation reviewed how EMG signal is generated, measured and processed, presented an EMG-driven model that could be used as a tool to estimate muscle forces. It provided a comprehensive description of the model, and applied the model in the knee and ankle joint of healthy subjects during gait.The second study of this dissertation used an EMG-driven model to estimate muscle forces and joint moments of patients following stroke during walking. The EMG-driven model did predict the ankle joint moment for patients following stroke, despite the variability in muscle activation patterns and joint moments between trials. The predictable ability of the EMG-driven model demonstrated that it could be used to estimate muscle forces and joint kinetics in patients with neurological disorder.The third study of this dissertation developed a forward dynamics model that incorporated an EMG-driven model. The model used EMGs, kinematics and ground reaction force data as inputs, and calculated muscle forces, as well as joint torques to drive a forward simulation during the stance phase of normal gait for five healthy subjects. The muscle forces calculated from the EMG-driven model were used to drive the knee and ankle joint. Therefore this EMG-driven approach has the advantage of identifying different muscle activation patterns, and it has great potential in applications to the rehabilitation of patients with neurological disorders.The fourth study of this dissertation developed a biomechanical model using EMG, joint position and force plate data as inputs to estimate anterior tibial translation (ATT), anterior shear forces and ligament loading in the healthy and anterior cruciate ligament (ACL)-deficient knee joint during gait. The model predicted that ATT increased throughout stance phase for the ACL-deficient knee compared with the healthy knee. The medial collateral ligament functioned as the main passive restraint to anterior shear force in the ACL-deficient knee. The calculated results were consistent with previous in vitro and in vivo studies, and this gave us confidence that our model could be used to study how ACL-deficient knees compensate for the loss of the ACL using abnormal muscle activation strategies. Posterior inclination angle of the tibial plateau was found to be a crucial parameter in determining knee mechanics, and increasing the tibial slope in our model would increase the resulting ATT and ligament forces in both healthy and ACL-deficient knees.The findings of this dissertation provide insight on neuromuscular control strategies of healthy subjects, post-stroke patients and patients without an ACL. The models have great potential in studying the outcome of different rehabilitation protocols on patients with neurological disorder.