A hybrid approach to the Skull Stripping problem in MRI

Abstract

We present a novel skull-stripping algorithm based on a hybrid approach that combines watershed algorithms and deformable surface models. Our method takes advantage of the robustness of the former as well as the surface information available to the latter. The algorithm first localizes a single white matter voxel in a T1-weighted MRI image, and uses it to create a global minimum in the white matter before applying a watershed algorithm with a preflooding height. The watershed algorithm builds an initial estimate of the brain volume based on the three-dimensional connectivity of the white matter. This first step is robust, and performs well in the presence of intensity nonuniformities and noise, but may erode parts of the cortex that abut bright nonbrain structures such as the eye sockets, or may remove parts of the cerebellum. To correct these inaccuracies, a surface deformation process fits a smooth surface to the masked volume, allowing the incorporation of geometric constraints into the skullstripping procedure. A statistical atlas, generated from a set of accurately segmented brains, is used to validate and potentially correct the segmentation, and the MRI intensity values are locally re-estimated at the boundary of the brain. Finally, a high-resolution surface deformation is performed that accurately matches the outer boundary of the brain, resulting in a robust and automated procedure. Studies by our group and others outperform other publicly available skullstripping tools.

Shape Modeling and Analysis with Entropy-Based Particle Systems

“…Surface representations were constructed from the MRI data using the FreeSurfer system, which provides an entire framework for the segmentation, surface reconstruction, topology correction, inflation and spherical parameterization of the cortex [94,25,36]. Correspondences were optimized on the cortical surfaces using the generalized PBM algorithm, with sulcal depth as the functional data.…”

Differential tinnitus-related neuroplastic alterations of cortical thickness and surface area

“…In particular, bilateral 55 peri-auditory regions in the superior temporal lobe [26,27,28] and subcortical areas that are considered as part of the (anatomically ill-defined) limbic system [29,34] often show anatomical changes. However, to date, neither the direction of tinnitus-related changes (increase or decrease of neuroanatomical gray matter (GM)) is clear, nor is it clarified whether the changes in the neural architecture 60 of distinct regions occur independently or are related.…”

“…This software is documented online and freely available for download (http://freesurfer.net). The technical details of these procedures are described in prior publications [49,50,51,52,53,54,55]. In short, the FreeSurfer pipeline generates models of the individual cortical surface with sub-voxel/-millimeter precision, yielding measures of CT, CSA and CV at each vertex of the surface.…”

Verbal paired-associate learning by APOE genotype in non-demented older adults: fMRI evidence of a right hemispheric compensatory response

“…Tissue segmentation: Whole brain segmentation of same session structural MRI scans was undertaken. Following N3 bias correction of field inhomogeneities [40], each scan was processed with one or both of the following automated skull-stripping methods: FreeSurfer's Hybrid Watershed Algorithm [38] and Brain Surface Extractor (Version 3.3) [36,39]. Both methods have been shown to be particularly effective when working with older adult images [19].…”