Autors: Georgieva, V. M., Katsarova V. V., Laskov, L. B. Title: A Hybrid Approach for Brain Tumor Detection in MRI Images Keywords: MRI of the brain, texture analysis and active contour segmen Abstract: In this paper we present a hybrid approach based on texture analysis and active contour segmentation for the 3 grades of meningiomas. Since the shape of the tumors plays a major role in its classification, we propose to segment the area of the meningioma on the base of the active contour without edges. We use the second-order texture analysis through Gray-Level Co-occurrence Matrices to extract and analyse some texture features, such as contrast, correlation, homogeneity, energy and entropy. The experiments, performed in MATLAB environment with real MRI images of a human brain have shown good results for detection of meningioma in the diagnostic phase, which can help for successfully treatment of this disease. References Issue
|
Вид: публикация в международен форум, публикация в реферирано издание, индексирана в Scopus