Classification of X-rays using statistical moments and SVM

Due to collection of thousands of medical digital images every day in medical institutions, the usage of medical digital images has been increasing rapidly. Because of the increase in medical digital images, managing this data properly and accessing it accurately is a raising need. To overcome the difficulties of the manual classification automated method is proposed. In this work an effort has been made to automatically classify X-rays at the macro level (global level) using statistical moments and SVM classifier, six classes of X-ray images are taken namely chest, foot, spine, neck, head, and palm. Each class consists of 60 images and it is collected from IRMA database. Initially pre-processing is performed by using the M3 filter and its region of interest is found by applying Connected Component Labeling (CCL), feature is extracted by applying statistical moments. The extracted features are used for classification using Support Vector Machines (SVM) which gave an accuracy rate of 92.58%.