Brain segmentation and the generation of cortical surfaces

This paper describes methods for white matter segmentation in brain images and the generation of cortical surfaces from the segmentations. We have dev eloped a system that allows a user to start with a brain volume, obtained b y modalities such as MRI or cryosection, and constructs a complete digital representation of the cortical surface. The methodology consists of three b asic components: local parametric modeling and Bayesian segmentation; surfa ce generation and local quadratic coordinate fitting; and surface editing. Segmentations are computed by parametrically fitting known density function s to the histogram of the image using the expectation maximization algorith m [DLR77]. The parametric fits are obtained locally rather than globally ov er the whole volume to overcome local variations in gray levels. To represe nt the boundary of the gray and white matter we use triangulated meshes gen erated using isosurface generation algorithms [GH95]. A complete system of local parametric quadratic charts [JWM(+)95] is superimposed on the triangu lated graph to facilitate smoothing and geodesic curve tracking. Algorithms for surface editing include extraction of the largest closed surface. Resu lts for several macaque brains are presented comparing automated and hand s urface generation. (C) 1999 Academic Press.