Detection Of Brain Tumor İn MRI Images Using Anisotropic Filtering And Kernel SVM Classification

Abstract

A Brain Tumor is a cancerous or non-cancerous mass or irregular cell growth in the Brain. It is a fatal disease which cannot be detected without MRI Scans. It became very hard for doctors to identify the exact location of a tumor. For medical analysis, an accurate description of MRI images is very important. The technique used in this paper assists in classifying a given MRI of the brain as normal or abnormal. The approach propound uses the discrete wavelet transform to extract the details or characteristics from images. Principal component analysis is used to normalize the data and reduce the dimensions of features. Image compression and recognition is done by PCA as well. The reduced features are sent to the KSVM to the classification of brain tumor automatically. Preprocessing and morphological operations are performed by using Anisotropic Filtering technique. This technique helps to detect the exact tumor spread region. The method used provides the accuracy of 99% in classifying tumors. This propound method can help the medical diagnosis to detect whether the patient is healthy or unhealthy under certain conditions