An automatic segmentation approach for boundary delineation of corpus callosum based on cell competition
Journal
30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society
Pages
5514-5517
Date Issued
2008
Author(s)
Abstract
The size and shape of corpus callosum are important indicators for assisting diagnosis of many neurological diseases involving morphological changes of corpus callosum. A new automatic segmentation approach was proposed in this paper for boundary delineation of corpus callosum. The basic idea of the proposed approach was to perform segmentation on the red component of color-coded map of diffusion tensor magnetic resonance image (MR-DTI). The boundary of corpus callosum was delineated in two phases. Firstly, a rough boundary surrounding corpus callosum was derived by using a built-in contour function in Matlab. Then, this cell competition algorithm was applied to the area inside the rough boundary derived in the first phase. The proposed segmentation approach has been evaluated and compared to the Chan and Vese level set method by using the MR-DTI images of a healthy volunteer and a systemic lupus erythematorsus (SLE) patient. The implementation results showed that the proposed approach could delineate the boundaries of corpus callosum reasonably well for both cases, whereas the Chan and Vese level set method failed to catch the weak edge for the SLE patient. ? 2008 IEEE.
SDGs
Other Subjects
Competition; Drop breakup; Edge detection; Level measurement; Magnetic resonance imaging; MATLAB; Automatic segmentations; Basic ideas; Boundary delineations; Cell competitions; Contour functions; Corpus callosum; Diffusion tensor magnetic resonance images; Level set methods; Morphological changes; Neurological disease; Rough boundaries; Size and shapes; Tensors; algorithm; article; artificial intelligence; automated pattern recognition; brain vasculitis; computer assisted diagnosis; corpus callosum; diffusion weighted imaging; human; image enhancement; methodology; pathology; reproducibility; sensitivity and specificity; Algorithms; Artificial Intelligence; Corpus Callosum; Diffusion Magnetic Resonance Imaging; Humans; Image Enhancement; Image Interpretation, Computer-Assisted; Lupus Vasculitis, Central Nervous System; Pattern Recognition, Automated; Reproducibility of Results; Sensitivity and Specificity
Type
conference paper