A Segmentation Method and Comparison of Classification Methods for Thyroid Ultrasound Images

In the conventional and relatively simple image processing techniques are most important task in the field of medical imaging. In this work to provide information about segmentation and classification methods that are very important for medical image processing. Ultrasound is unique in its ability to image patient anatomy and physiology in real time, providing an important, rapid and non-invasive means of evaluation. In this paper uses the groups of Benign and Malignant thyroid nodules images. These images used to analysis the classification accurately. GLCM extracts the total 13 features and these features are used to analysis in classifiers such as SVM, KNN and Bayesian. Experimental results illustrated that the classifiers like SVM/k-NN/Bayesian comparing to each other and enhanced classification accuracy. Result shows the SVM is best classification method and the performance measure such as accuracy. It is observed that the SVM gives much better accuracy than KNN and Bayesian.