Classification of normal and abnormal lung sounds using neural network and support vector machines

This work proposes feature extraction of lung sounds using wavelet coefficients and their classification by neural network and support vector machines. The lung sounds were classified into 6 classes. The results stated the advantages of a support vector machines for the classification of normal and abnormal lung sounds, and indicated that SVMs are a highly successful classifier with accuracy about 93.51-100 for classification of lung sounds.