Segmentation of MRI images for brain cancer detection
Wassim El Hajj Chehade, Riham Abdel Kader, Ali El‐Zaart · 2018 International Conference on Information and Communications Technology (ICOIACT) · 2018
Brain cancer detection without human interference is a major challenge in the field of medical image processing. Segmentation of MRI brain images is a technique used as a first step towards extracting different features from these images for analyzing, interpreting and understanding. The objective of MRI brain segmentation is to detect the type of brain abnormality. Many segmentation techniques have been proposed in the literature. These techniques use Gaussian distribution to estimate the image threshold. Gaussian distribution assumes that the histogram of the image has symmetric distribution. However, if the histogram is non-symmetric, a more generic distribution, i.e. a Gamma distribution, must be used. The aim of this paper is to enhance Li's method that has been proved good for image segmentation, by using Between-Class Variance with Gamma distributions. The proposed method will be tested on MRI brain images. Experiments show good results for our enhanced segmentation formula.