MR Images Classification Using Hybrid KNNSVM ...

In the field of medical, image data play a vital role to assist the physicians in all kinds. Especially, Magnetic Resonance Image data will be very useful to diagnose brain tumors in human brain. Unfortunately, there are certain difficulties to classify those images to take sudden appropriate decisions to recover the identified disease. Hence, the concept of image mining is used to extract potential hidden information from the image data and those can be classified to take right decision for the early recover of the patient. Support Vector Machine (SVM) and K-Nearest Neighbor (KNN) are very powerful and popular classification algorithms to classify image data. However, these two have their own drawbacks in certain situations. In this paper, both SVM and KNN have been merged to derive a hybrid KNNSVM algorithm to diagnose the MR Images in an effective manner with high accuracy rate and low error rate.