Automatic brain tumor detection and segmentation from multi-modal MRI images based on region growing and level set evolution

Abstract

Glioma is a type of brain tumor, originates from glial cells. Approximately 80% of them are malignant. Based on pathological evolution of tumor, they can be classified into two types of tumor — high grade & low grade glioma. In this paper, the segmented area obtained from the conventional region-growing approach is automatically selected as the the initial contour to the iterative distance regularized level set evolution method thus removing the need of selecting the initial region of interest by the user. Therefore, a computer aided fully automated technique is developed to detect glioma from multimodal MRI images & segment the tumor region from whole image. The proposed method is capable of improving the overall detection and segmentation performance of tumor for different glioma cases of BRATS 2012 publicly available database.