An Automated Framework to Segment and Classify Gliomas Using Efficient Segmentation and Classification

Abstract

A brain disease that has high levels of disability, mortality, and malignancy is known as Glioma. The analysis of image classification is divided into different image features and this will organize all data into various categories. In this work, an automated method for segmenting and classifying Gliomas is presented. The effectiveness of various classifiers are evaluated. Extraction of features will provide higher levels of imaging features for the image with regard to colour, texture, contrast, and shape. Gabor Filtered images can capture all relevant spectrums of frequency for extracting the features that are aligned at certain specific orientations. The Naïve Bayes (NB) is a simple algorithm that implements this giving good results. The Support Vector Machine (SVM) is yet another statistical learning method that may be applied for solving problems in regression or classification. The design of the Artificial Neural Network (ANN) is formulated in a way in which it is quite similar to the brain of humans. It can detect and also discover relationships and certain common patterns of raw data.