<![CDATA[Early detection of kidney tumors is essential to increase the chances of a patient's recovery. This study aims to develop a classification system for kidney CT scan images to distinguish between normal kidneys and kidneys containing tumors. The classification method used is Support Vector Machine (SVM) with three types of kernels, namely linear, polynomial, and radial basis function (RBF). Previously, feature extraction was performed using two approaches, namely Histogram of Oriented Gradients (HOG) to obtain shape values, and Gray Level Co-occurrence Matrix (GLCM) to obtain texture characteristics of the image. The test results show that SVM with a linear kernel gives the highest accuracy of 90%, followed by polynomial at 85%, while the RBF kernel only reaches 50%. Based on these results, it can be concluded that the combination of HOG and GLCM feature extraction followed by classification using linear kernel SVM is effective for distinguishing normal kidney images and kidney tumors. This research makes a positive contribution to the development of a medical image-based kidney disease diagnosis support system.]]>
Copyrights © 2025
