Segmentation and Detection of Brain Tumor Using 2D and 3D Image Processing

Abstract

Brain tumor is a growth of abnormal cells in the brain that can be either benign or malignant. It tends to affect people irrespective of age group. Thus earlier detection and treatment of brain tumor is needed for quick recovery. Doctors use Magnetic Resonance Imaging (MRI) to identify the intensity of tumor. The MRI images are pre-processed and then Deep learning neural network and spearman algorithms are used to provide segmentation of images. The segmented images are grouped using Fuzzy clustering technique. The CNN and SVM Classifiers are used to classify and detect the size of tumor cell. On comparison, it was found that CNN classifiers are more accurate, sensitive and yields high performance as compared to SVM classifiers for both 2D and 3D images.