Predicting Cognitive Outcomes by Mapping White Matter Tracts to Surface

White matter tracts are bundles of myelinated nerve fibers that connect different regions of the brain, facilitating communication between them. These tracts play an important role in the cognitive and behavioral functioning of the brain. Understanding the structure and connectivity of white matter tracts is crucial for studying brain function and diagnosing neurological disorders. In this study, we propose a new method to map fiber tract information onto the cortical regions via fiber to cortex minimal distances. Diffusion properties of the fiber tracts weighted by these distances can then be incorporated as adjacency weights in a graph convolutional neural network. Our approach provides a multi-modality framework that integrates structural and diffusion MRI, providing a comprehensive view of the brain’s architecture. We evaluate this framework in two longitudinal studies, predicting later cognitive outcomes.