Convolutional-Neural-Network Assisted Segmentation and SVM Classification of Brain Tumor in Clinical MRI Slices

Due to the increased disease occurrence rates in humans, need for Automated Disease Diagnosis (ADD) systems is also raised. Most of ADD are proposed support doctor during screening and decision making process. This research aims at developing a Computer Aided (CADD) scheme categorize brain tumour 2D MRI slices into Glioblastoma/Glioma class with better accuracy. The main contribution this work develop CADD system Convolutional-Neural-Network (CNN) supported segmentation classification. framework consist following phases; (i) Image collection resizing, (ii) using VGG-UNet, (iv) Deep-feature extraction VGG16 network, (v) Handcrafted feature extraction, (vi) Finest choice by firefly-algorithm, (vii) Serial concatenation binary merit executed confirmed an investigation realized benchmark as well clinically collected slices. In work, classification 10-fold cross validation implemented known classifiers results attained SVM-Cubic (accuracy >98%) superior. result confirms that combination CNN assisted helps achieve enhanced detection