Open AccessA Unique Six Sigma Based Segmentation Technique for Brain Tumor Detection and Classification using Hybrid CNN-SVM Model
Author(s) -
Arati Kothari,
B. Indira
Publication year - 2019
Publication title -
international journal of recent technology and engineering
Language(s) - English
Resource type - Journals
ISSN - 2277-3878
DOI - 10.35940/ijrte.a1239.058119
Subject(s) - support vector machine , artificial intelligence , computer science , pattern recognition (psychology) , convolutional neural network , segmentation , feature (linguistics) , brain tumor , medicine , pathology , philosophy , linguistics
An intelligent organizing scheme to detect andclassify normal, abnormal MRI brain sequences has beenillustrated here. At present, handling of brain tumors diseaseand decision is based on radiological appearance and itssymptoms. Magnetic-Resonance-Imaging (MRI) is a powerfulsubstantial precise instrument for functional conclusion of braintumorous. In existing study, broad range of methods is used forbrain cancer detection and classification. Under this methodsviz., image pre-processing, enhancement, segmentation, featuremining and resulting classification is efficiently conducted.Furthermore, when various machine learning algorithms like:Six Sigma, Convolutional Neural Network (CNN), SupportVector Machine (SVM), are employed to detect and extract thetumor region and classify numerous sequence of imageries, it iswitnessed from our results that this Hybrid CNN-SVM modelgives maximum classification accuracy rate of 99.33%compared to previous models. The foremost aim of this researchis to get an effective result for detecting type of brain tumorusing six sigma based segmentation technique, and to achieveefficient classification rate, using hybrid CNN-SVM model.