Visualization of disconnection syndromes in humans

Abstract

Knowledge of the relationship between structure and function is essential to the exploration of the architecture of cognition. Cognitive processes require the coordinated activity of large-scale brain networks consisting of distant cortical regions, connected by long-range white matter tracts. Despite decades of connectional tracing studies in monkeys, the backwardness of human anatomy makes it difficult to draw conclusions from lesion studies and functional neuroimaging when brain connectivity is at issue. We propose an approach to clinico-anatomical correlation, based on a standardized atlas of white matter tracts derived from diffusion tensor imaging tractography. Using OVER-TRACK, a method based on tracking and overlapping white matter tracts, we mapped the course of three rostro-caudal association pathways in the Montreal Neurological Institute space. For each voxel we defined the probability of finding fibers belonging to individual tracts. This method is defined to localize in the white matter the overlapping lesion derived from a group of patients with brain damage. Our study provides a general approach for establishing anatomo-functional correlations by estimating the cortical areas connected in normal subjects, or disconnected by white matter lesions. This method will help researchers and clinicians to identify the neural bases of cognitive abilities and the behavioral consequences of brain lesions.