Quantification of Brain Tumor region in MR Images using FCM technique

Medical image processing confines the use and exploration of image datasets of human body obtained from CT or MRI scanner to guide medical interventions. The variations in tumor location, shape, and size present an important challenge to the identification of brain tumors. Although many clinical specialists or radiologists are skilled in spotting tumors and other abnormalities through imaging modalities using the MRI technology in the brain but it is noted that it takes more time and effort to identify, locate, and segment the affected part of the brain. In order to identify brain tumors in MR images, this effort first removes noise using noise removal technique Adaptive Median Filter(AMF) before enhancing the pictures with the contrast limited adaptive histogram equalization(CLAHE). Additionally, image segmentation is carried out using both fuzzy c-means, thresholding and segmented image is created as input to canny edge detection that yields the tumor image and area of the region is calculated by finding area of largest contour region. The computational complexity that plagues earlier systems that mix fuzzy and k-means clustering for this purpose has been solved by the system created in this research.