Edge detection in MRI brain tumor images based on fuzzy C-means clustering
URI (for links/citations):https://www.sciencedirect.com/science/article/pii/S1877050918313474
Zotin, Alexander Gennadievich
Simonov, Konstantin Vasilyevich
Kurako, Mikhail Aleksandrovich
Hamad, Yousif Ahmed
Кириллова, С. В.
Институт космических и информационных технологий
Кафедра прикладной математики и компьютерной безопасности
Journal Name:Procedia Computer Science
Journal Quartile in Scopus:без квартиля
Journal Quartile in Web of Science:без квартиля
Bibliographic Citation:Zotin, Alexander Gennadievich. Edge detection in MRI brain tumor images based on fuzzy C-means clustering [Текст] / Alexander Gennadievich Zotin, Konstantin Vasilyevich Simonov, Mikhail Aleksandrovich Kurako, Yousif Ahmed Hamad, С. В. Кириллова // Procedia Computer Science. — 2018. — Т. 126. — С. 1261-1270
Nowadays, medical image processing is the most challenging and emerging field. Edge detection of MRI images is one of the most important stage in this field. The paper describes the proposed strategy to detect the edges of brain tumor from patient’s MRI scan images of the brain. At the first stage, this method includes some noise removal functions improving features that provides better characteristics of medical images for reliable diagnosis using Balance Contrast Enhancement Technique (BCET). The result of second stage is subjected to image segmentation using Fuzzy c-Means (FCM) clustering method. Finally, Canny edge detection method is applied to detect the fine edges. During the experimental study, we used images containing brain tumors that were characterized by different location, type of pathology, shape, size and density, as well as the size of the area of the affected tissue near the tumor space. Detection and extraction of tumor from MRI scan images of the brain is done using MATLAB software. The obtained results demonstrate some resistivity to a noise. Also, the accuracy of segmentation, in some cases of tumor pathology, was increased up to 10-15% regarding the expert estimates.