Bone Cancer Identification and Separation Using K-Means and KNN Classifiers

Abstract

The unregulated cell growth will lead to the dangerous and deadliest cancer disease. In the body of humans various kinds of cancer has been detected, after going on many researches. Among all these, bone cancer is the one which will spread more widely, because of this it will leads to death. There is no anticipation for the bone cancer, therefore the bone cancer detection is more critical. Nowadays, methods like data mining and the image processing methods are utilized for the most of the studies in the process of medical image analysis. Many scientific researchers have been predicted the data, related websites and the collection of knowledge from the large databases. There are many methods used in the approaches for bone cancer detection and classification like supports vector machines, Association rule mining and fuzzy theory. In this approach of segmentation k-means will be utilized for segmenting the bone regions. In further processes for bone cancer detection the segmented image will be used by the mean intensity evaluation of the area identified. To check whether there is a presence or absence of bone cancer in the medical images that is for the classification process threshold values are used. This approach can be used for jpeg images, CT scan images. The proposed work will use the K-Nearest Neighbor (KNN) classifier as a classification technique and achieved to produce better accuracy.