\r\nabnormalities in heart sounds. Since accurate auscultation is

\r\na crucial first step in screening patients with heart diseases,

\r\nthere is a need to develop computer-aided detection\/diagnosis

\r\n(CAD) systems to assist cardiologists in interpreting heart sounds

\r\nand provide second opinions. In this paper different algorithms

\r\nare implemented for automated heart sound classification using

\r\nunsegmented phonocardiogram (PCG) signals. Support vector

\r\nmachine (SVM), artificial neural network (ANN) and cartesian

\r\ngenetic programming evolved artificial neural network (CGPANN)

\r\nwithout the application of any segmentation algorithm has been

\r\nexplored in this study. The signals are first pre-processed to remove

\r\nany unwanted frequencies. Both time and frequency domain features

\r\nare then extracted for training the different models. The different

\r\nalgorithms are tested in multiple scenarios and their strengths and

\r\nweaknesses are discussed. Results indicate that SVM outperforms

\r\nthe rest with an accuracy of 73.64%.","references":"[1] B. M. Whitaker, P. B. Suresha, C. Liu, G. D. Clifford, and D. V.\r\nAnderson, \u201cCombining sparse coding and time-domain features for heart\r\nsound classification,\u201d Physiological measurement, vol. 38, no. 8, p. 1701,\r\n2017.\r\n[2] D. B. Springer, L. Tarassenko, and G. D. Clifford, \u201cLogistic\r\nregression-hsmm-based heart sound segmentation,\u201d IEEE Transactions\r\non Biomedical Engineering, vol. 63, no. 4, pp. 822\u2013832, 2016.\r\n[3] I. Turkoglu, A. Arslan, and E. Ilkay, \u201cAn expert system for diagnosis\r\nof the heart valve diseases,\u201d Expert systems with applications, vol. 23,\r\nno. 3, pp. 229\u2013236, 2002.\r\n[4] T. J. Hirschauer, H. Adeli, and J. A. Buford, \u201cComputer-aided diagnosis\r\nof parkinsons disease using enhanced probabilistic neural network,\u201d\r\nJournal of medical systems, vol. 39, no. 11, p. 179, 2015.\r\n[5] J.-S. Chou and A.-D. Pham, \u201cSmart artificial firefly colony\r\nalgorithm-based support vector regression for enhanced forecasting\r\nin civil engineering,\u201d Computer-Aided Civil and Infrastructure\r\nEngineering, vol. 30, no. 9, pp. 715\u2013732, 2015.\r\n[6] Z. Sankari and H. Adeli, \u201cProbabilistic neural networks for diagnosis of\r\nalzheimer\u2019s disease using conventional and wavelet coherence,\u201d Journal\r\nof neuroscience methods, vol. 197, no. 1, pp. 165\u2013170, 2011.\r\n[7] J. J. G. Ortiz, C. P. Phoo, and J. Wiens, \u201cHeart sound classification\r\nbased on temporal alignment techniques,\u201d in Computing in Cardiology\r\nConference (CinC), 2016. IEEE, 2016, pp. 589\u2013592.\r\n[8] A. Ganguly and M. Sharma, \u201cDetection of pathological heart murmurs\r\nby feature extraction of phonocardiogram signals,\u201d Journal of Applied\r\nand Advanced Research, vol. 2, no. 4, pp. 200\u2013205, 2017.\r\n[9] C. Potes, S. Parvaneh, A. Rahman, and B. Conroy, \u201cEnsemble of\r\nfeature-based and deep learning-based classifiers for detection of\r\nabnormal heart sounds,\u201d in Computing in Cardiology Conference (CinC),\r\n2016. IEEE, 2016, pp. 621\u2013624.\r\n[10] M. Zabihi, A. B. Rad, S. Kiranyaz, M. Gabbouj, and A. K. Katsaggelos,\r\n\u201cHeart sound anomaly and quality detection using ensemble of\r\nneural networks without segmentation,\u201d in Computing in Cardiology\r\nConference (CinC), 2016. IEEE, 2016, pp. 613\u2013616.\r\n[11] E. Kay and A. Agarwal, \u201cDropconnected neural network trained\r\nwith diverse features for classifying heart sounds,\u201d in Computing in\r\nCardiology Conference (CinC), 2016. IEEE, 2016, pp. 617\u2013620.\r\n[12] G. Redlarski, D. Gradolewski, and A. Palkowski, \u201cA system for heart\r\nsounds classification,\u201d PloS one, vol. 9, no. 11, p. e112673, 2014.\r\n[13] J.-b. Wu, S. Zhou, Z. Wu, and X.-m. Wu, \u201cResearch on the method\r\nof characteristic extraction and classification of phonocardiogram,\u201d in\r\nSystems and Informatics (ICSAI), 2012 International Conference on.\r\nIEEE, 2012, pp. 1732\u20131735.\r\n[14] M. Abdollahpur, A. Ghaffari, S. Ghiasi, and M. J. Mollakazemi,\r\n\u201cDetection of pathological heart sounds,\u201d Physiological measurement,\r\nvol. 38, no. 8, p. 1616, 2017.\r\n[15] K. Ek\u02c7stein and T. Pavelka, \u201cEntropy and entropy-based features in signal\r\nprocessing.\u201d\r\n[16] M. M. Azmy, \u201cClassification of normal and abnormal heart sounds\r\nusing new mother wavelet and support vector machines,\u201d in Electrical\r\nEngineering (ICEE), 2015 4th International Conference on. IEEE,\r\n2015, pp. 1\u20133.\r\n[17] J. F. Miller and P. Thomson, \u201cCartesian genetic programming,\u201d in\r\nEuropean Conference on Genetic Programming. Springer, 2000, pp.\r\n121\u2013132.\r\n[18] M. M. Khan, A. M. Ahmad, G. M. Khan, and J. F. Miller, \u201cFast learning\r\nneural networks using cartesian genetic programming,\u201d Neurocomputing,\r\nvol. 121, pp. 274\u2013289, 2013.\r\n[19] G. Khattak, M. Khan, G. Khan, F. Huenupan, and M. Curilem,\r\n\u201cAutomatic classification of seismic signals of the chilean llaima volcano\r\nusing cartesian genetic programming based artificial neural network,\u201d\r\n2017.\r\n[20] Z. Jiang and H. Wang, \u201cA new approach on heart murmurs classification\r\nwith svm technique,\u201d in Information Technology Convergence, 2007.\r\nISITC 2007. International Symposium on. IEEE, 2007, pp. 240\u2013244.\r\n[21] A. H. Salman, N. Ahmadi, R. Mengko, A. Z. Langi, and T. L. Mengko,\r\n\u201cEmpirical mode decomposition (emd) based denoising method for heart\r\nsound signal and its performance analysis,\u201d International Journal of\r\nElectrical and Computer Engineering, vol. 6, no. 5, p. 2197, 2016.","publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 140, 2018"}