No abstract

Neurological disorders such as stroke impair locomotor control and result in abnormal 3-D gait kinematics. Establishment of effective rehabilitation strategies requires an understanding of how individual muscles contribute to pathological movement. Forward dynamic simulations account for complexities of interjoint coupling and can be used to predict dynamic muscle function. However to date, limited experimental validations of dynamic models have been performed. Our objective was to measure 3-D movement induced by the biceps femoris (BF), rectus femoris (RF), and vastus lateralis (VL) in limb configurations corresponding to the swing phase of gait, and to assess the biomechanical factors that affect dynamic function. Subjects were positioned in a robotic gait orthosis that included a compliant interface. Electrical stimulation was introduced into individual muscles while induced hip and knee joint movements were recorded. Measured hip to knee sagittal plane acceleration ratios were consistent with dynamic musculoskeletal model simulations. However RF and VL induced substantially larger frontal plane hip movements than model-based predictions. Sensitivity analyses on musculoskeletal model parameters revealed that muscle function depends primarily on moment arm assumptions. Though generic musculoskeletal models are suitable for predicting sagittal plane muscle function, improvements in moment arm accuracy are essential for investigation of 3-D pathological gait.

Increased frontal-plane hip movement of the affected leg during the swing phase is a commonly observed gait adaptation in stroke patients. Recent evidence suggests that pathologically-induced torque coupling may contribute to asymmetric gait behaviors observed following stroke. This study proposes to use a CPG-controlled three-dimensional (3D) bipedal model to quantify the effects of abnormal torque coupling on frontal plane gait kinematics. Model dynamics have been evaluated using overground data collection observed under comparable in vivo experimental conditions. The CPG controller has demonstrated ability to provide sustained stable gait over an inclined surface in a simplified model. Preliminary results indicate that the proposed framework is feasible to control a 3D model for investigating the effects of torque coupling on the abnormal frontal plane kinematics of pathological gait.

No abstract