Open AccessBenign and Malignant Tumor Classification using Machine Learning Technique
Author(s) -
Ch. Usha Kumari*,
N.Madhusudhana Rao,
M.N.V.S.S Kumar,
Pinninti Kishore,
M Pala Prasad Reddy
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.c6297.098319
Subject(s) - artificial intelligence , thresholding , support vector machine , pattern recognition (psychology) , feature extraction , computer science , brain tumor , contextual image classification , segmentation , medicine , pathology , image (mathematics)
Brain tumor is the growth of large mass of abnormal cells in human brain. Brain tumors are of different types benign and malignant. Benign tumor is noncancerous and malignant tumor is cancerous. This research paper classifies benign and malignant using SVM classifier. The image processing techniques are used for image enhancement and restoration. Total 60 images are taken out of which 30 images are benign i.e., noncancerous images and 30 images are malignant i.e., tumor images. Image Filtration is performed on the input MRI image by using median filter. Segmentation is done using thresholding technique and Gray Level Concurrence matrix is used for the feature extraction. Features such as entropy, energy, homogeneity, correlation, contrast, IDM, RMS, standard deviation and mean are extracted for tumor region. After feature extraction all these features are given to SVM classifier to classify benign and malignant tumors. The SVM classifier has given 97.7% accuracy.