Multi-phase three-dimensional level set segmentation of brain MRI

Abstract

This paper presents the implementation and quantitative evaluation of a four-phase three-dimensional active contour implemented with a level set framework for automated segmentation of cortical structures on brain T1 MRI images. The segmentation algorithm performed an optimal partitioning of three-dimensional data based on homogeneity measures that naturally evolves to the extraction of different tissue types in the brain. Random seed initialization was used to speed up numerical computation and avoid the need for a priori information. A simple post-processing, based on morphological operators, was applied to correct for segmentation artifacts. The segmentation method was tested on ten MRI brain data sets and quantitative evaluation was performed by comparison to manually labeled data, Computation of false positive and false negative assignments of voxels for white matter, gray matter and cerebrospinal fluid were performed. Results reported high accuracy of the segmentation methods, demonstrating the efficiency and flexibility of the multi-phase level set segmentation framework to perform the challenging task of automatically extracting cortical brain tissue volume contours. © Springer-Verlag Berlin Heidelberg 2004.