SVM (Support Vector Machine) was used for classification based onthe work by Khandokar et al. [22]. In this method, a part of a dataset istypically used for training (healthy and patients are denoted by labels of 1 and þ1 respectively), while the remaining data is used to validate thealgorithm. The algorithm compares the test data with true labels, wherethe accuracy of the classifier increases with the number of true labels.The leave-one-out method is used in the cross validation technique forthe low number of observations [22]. Furthermore, two indicators aredefined for the specificity and sensitivity in the process of finding theoptimum classifier. These indicators along with the classification accuracyare considered as the overall quality measures of the classifier.