(Publisher of Peer Reviewed Open Access Journals)
ICETT-2012
Full-Text PDF
Paper Title : Automatic Medical Image Classification and Abnormality Detection Using K-Nearest Neighbour
Author Name : R. J. Ramteke, Khachane Monali Y.
Abstract : This research work presents a method for automatic classification of medical images in two classes Normal and Abnormal based on image features and automatic abnormality detection. Our proposed system consists of four phases Preprocessing, Feature extraction, Classification, and Post processing. Statistical texture feature set is derived from normal and abnormal images. We used the KNN classifier for classifying image. The KNN classifier performance compared with kernel based SVM classifier (Linear and RBF). The confusion matrix computed and result shows that KNN obtain 80% classification rate which is more than SVM classification rate. So we choose KNN algorithm for classification of images. If image classified as abnormal then post processing step applied on the image and abnormal region is highlighted on the image. The system has been tested on the number of real CT scan brain images.
Keywords : K –Nearest Neighbour (KNN), Computed Tomography (CT), Support Vector Machine (SVM), Radial Basis Function (RBF).
Cite this article : R. J. Ramteke, Khachane Monali Y. " Automatic Medical Image Classification and Abnormality Detection Using K-Nearest Neighbour " ,ICETT-2012 ,Page No : 186-192.