Graph-Theoretical Analysis using Data-Driven Features: Application to Rehabilitation After Stroke

The assessment of neuroplasticity after stroke through functional magnetic resonance imaging (fMRI) analysis is a developing field where the objective is to understand the neural process of recovery and to better target rehabilitation. In this study, the connectivity structure of the stroke-affected brain is analyzed before and after rehabilitation. The main challenge associated with fMRI data from stroke patients is the significant individual variability that is present. We provide a graph-theoretical framework to effectively compare algorithm performance for capturing subject variability for real data. We demonstrate that independent vector analysis provides superior performance when compared with group independent component analysis. In this study, increased small worldness across components and greater centrality in key motor networks are demonstrated as a result of the intervention, suggesting improved efficiency in neural communication. Clinically, these results bring forth new possibilities as a means to observe the neural processes underlying improvements in motor function.