Cortical Surface Models from MRI
David Shattuck
UCLA
Surface models of the cerebral cortex are useful in a variety of applications, including visualization, intersubject registration and comparison, and EEG/MEG source localization. We have developed an automated processing sequence that produces cerebral cortex models with spherical topology from T1-weighted MRI of the human head.
We first skull-strip the brain using our Brain Surface Extractor (BSE). BSE combines anisotropic diffusion filtering, Marr–Hildreth edge detection, and a sequence of mathematical morphology operators to segment and extract the brain from surrounding tissues. We then compensate for tissue intensity non-uniformities using our Bias Field Corrector (BFC). BFC computes local estimates of gain variation using an adaptive partial volume tissue model and fits a tricubic B-spline to estimate gain variation throughout the brain volume. The image is then corrected using the spline values.
Next, each voxel in the brain is labeled according to tissue content. Our tissue classifier combines the tissue model with a Gibbs spatial prior, encouraging contiguous regions of similar tissue types. The cerebrum is then extracted from the labeled volume by registering a labeled atlas to the skull-stripped MRI data. Isocontour algorithms are subsequently used to generate a surface model from the identified cerebrum.
However, such models do not typically exhibit spherical topology. To address this, we apply a Topological Correction Algorithm (TCA), which uses a graph-based approach to ensure that the resulting tessellations have spherical topology. Finally, the surfaces are parameterized using a method based on p-harmonic energy minimization.
In this presentation, we describe each stage of the method and demonstrate related software tools that we have developed.
An example includes a cortical surface extracted using our BrainSuite tools.
Audio (MP3 File, Podcast Ready)