“…Handcrafted features like texture [ 30 , 32 ] were extracted from the left–right lung zones [ 40 , 41 , 42 , 43 ]. Following feature selection, they were fed into various machine learning classifiers, including support vector machines (SVM) [ 40 , 41 , 43 , 44 , 45 , 46 , 47 , 48 , 49 , 50 , 51 ], decision trees (DT) [ 47 , 48 ], random trees (RT) [ 44 , 49 , 50 , 51 ], artificial neural networks (ANNs) [ 52 , 53 , 54 ], K-nearest neighbours (KNN) [ 55 ], self-organizing map (SOM) [ 55 ], backpropagation (BP), the radial basis function (RBF) neural networks (NN) [ 44 , 49 , 50 , 51 , 55 , 56 ], and Ensemble classifier [ 41 , 43 , 48 ]. Among the classifiers, SVM had the best overall detection accuracy, with a 73.17 percent success rate when using the same dataset as this study.…”