NEUROIMAGING INVESTIGATION OF COMPENSATORY BRAIN MECHANISMS AFTER CORPUS CALLOSUM SURGICAL RESECTION

Interhemispheric communication in the human brain is primarily mediated by the corpus callosum (CC), a major white matter commissure connecting the two hemispheres. Other alternative extracallosal pathways may support limited interhemispheric coordination, but their contribution is difficult to isolate in healthy brains because the CC provides the dominant route. ‘Split-brain’ conditions, in which the CC is surgically resected (callosotomy), most often as a treatment for drug-resistant epilepsy, offer a unique opportunity to examine interhemispheric communication in the absence of the main commissural pathway. Findings from prior studies indicate that some degree of coordinated activity between hemispheres may persist even when direct callosal fibers are absent, raising questions about the structural substrates that could support residual functional coupling. This thesis investigates the relationship between resting-state functional networks and structural connectivity in a split-brain patient and in a healthy control subject. The control subject provides a physiological reference, whereas the split-brain patient analysis focuses on whether canonical networks retain a bilateral spatial organization and on potential indirect pathways that may support residual interhemispheric coupling. A multimodal Magnetic Resonance Imaging (MRI) analysis framework was implemented by combining structural MRI, resting-state functional MRI and diffusion MRI data, using FSL and MRtrix3 as primary software tools. Functional connectivity was characterized by identifying canonical resting-state networks (RSNs) through independent component analysis followed by manual component classification into signal and noise. Structural connectivity was investigated from diffusion data using constrained spherical deconvolution for fiber orientation estimation and anatomically constrained tractography, a technique that reconstructs the most plausible white matter pathways by tracking the directionality of water diffusion in brain tissue while being guided by anatomical tissue constraints. RSN-guided tractography was then employed, integrating RSN maps with diffusion tractography, to explore structural pathways that may link bilateral RSN activations and to assess whether plausible indirect routes could support interhemispheric coupling despite the absence of callosal fibers. In the healthy control subject, tractography as expected indicated that the CC represents the primary structural route linking homologous regions across hemispheres, supporting the validity of the analysis workflow and strengthening confidence in the split-brain findings. In contrast, the analysis in the split-brain patient revealed a marked reduction of direct interhemispheric structural connections consistent with callosal disruption, while several bilateral RSNs were still detectable, including the medial visual, occipital visual, sensorimotor, auditory and executive control networks. RSN-guided tractography in the split-brain patient highlighted recurrent trajectories intersecting subcortical structures and the brainstem, suggesting that indirect anatomical routes may become relatively more relevant when commissural fibers are unavailable. Overall, this work provides an integrated functional-structural perspective on brain network organization after callosal disconnection and supports the use of RSN-guided tractography for investigating candidate compensatory pathways supporting residual interhemispheric communication when callosal fibers are disrupted.