Anatomically Consistent Segmentation of the Human Cortex for Functional MRI Visualization

We describe a system that is being used to segment gray matter from volumetric representations of the human cortex obtained using magnetic resonance imaging. The segmentation algorithm identi es gray matter voxels and computes their connectivity. The method di ers from existing schemes in that it exploits knowledge of the anatomy of human cortex and produces anatomically consistent segmentations. The method is based on a novel and computationally e cient technique of incorporating structural constraints into the segmentation algorithm. Because the gray matter segmentation is anatomically consistent, it can be used together with functional magnetic resonance imaging measurements to visualize the spatial pattern of cortical activity within the gray matter.