There is compelling evidence that human movement may be produced through the flexible combination of a limited set of muscle activation patterns, or muscle synergies, which are differentially weighted according to task. As defined here, a muscle synergy is composed of a fixed balance of activation across muscles. Effectively, muscle synergies may simplify motor control by reducing the dimensionality (degrees of freedom) of the control problem. The central thesis of the proposed work is that cerebrovascular accidents alter the number and composition of muscle synergies available for control of lower limb motor tasks, resulting in altered gait and postural patterns of muscle activation. The recovery of lower limb function following stroke is characterized by the emergence of stereotypical movement patterns involving a relatively tight coupling of motions at the hip, knee and ankle. While these kinematic patterns suggest the existence of abnormal muscle synergies, the surprisingly few quantitative studies of these synergies (in terms of joint torque patterns) have yielded conflicting results. Our preliminary studies strongly suggest that these kinematic disturbances are related, at least in part, to abnormal torque coupling between the impaired hip and knee. The proposed work will expand on the quantitative evaluation of lower limb torque patterns, and for the first time, use sophisticated matrix factorization techniques to identify the muscle synergies underlying post-stroke muscle activation in the lower limb. These underlying muscle synergies will be identified under isometric conditions and related to muscle activation during gait. In particular, we seek to demonstrate that muscle activation patterns observed during the swing phase of gait are constrained by the set of muscle synergies identified under isometric conditions. In this study, we will take an important first step toward the identification of the neural substrates that mediate abnormal lower limb muscle synergies following stroke. Specifically, we will investigate the roles of the contralateral and ipsilateral motor cortices in mediating abnormal torque coupling and muscle synergies in the paretic lower limb observed under static conditions. The knowledge generated by our study will form a cornerstone for future clinical research and provide a rational scientific basis for the design of interventions that seek to overcome abnormal across-joint coupling. Furthermore, longitudinal static and dynamic assessments can be designed to study the development of abnormal muscle synergies and torque coupling during recovery from stroke. It may also be possible to develop clinical assessment tools based on matrix factorization algorithms that will allow a succinct characterization of coordination disturbances following stroke.
Following stroke, stereotypical abnormal gait patterns emerge during recovery. It is possible that these abnormal patterns are the result of inappropriate combinations of muscle activations. Since the few existing studies have shown conflicting results, we propose to quantify these lower limb patterns, identify the possible muscle activation combinations producing these patterns, and take the important first stop toward identifying the neurological basis for these gait impairments. The results of these studies will assist in developing rehabilitation techniques and strategies and identify novel neurological targets to improve functional movement in stroke survivors.
