PHONOCARDIOGRAPHY-BASED MITRAL VALVE PROLAPSE DETECTION USING ARTIFICIAL NEURAL NETWORK
Abstract
Mitral valve prolapse (MVP) is the most common valve anomaly and most often cause of isolated mitral insufficiency. MVP has mostly benign course and prognosis in childhood, while the complications like severe mitral regurgitation, infectious endocarditis, pulmonary embolism, arrhythmia and sudden death more often occur with elderly people, which demands prompt diagnostics and prevention. Due to its frequent occurrence, diagnostic failure and clinical importance of early MVP detection, the aim of this study was to develop an original, non-invasive and easily applicable diagnostic method for MVP detection in children and adolescents by using artificial neural network (ANN). Cardiac sounds in 48 children with MVP, 49 healthy children and 38 children with pathological heart murmur of atrial septal defect (ASD), ventricular septal defect (VSD), ductus arteriousus persistence (DAP), aortic stenosis (AS), pulmonic stenosis (PS), aortic coarctation (ACo), mitral regurgitation (MR), mitral insufficiency (MI) and tricuspid insufficiency (TI) were recorded by auscultation using electronic stethoscope. In electronic stethoscope the sound is archived in internal memory of the stethoscope and then transmitted to a computer memory by a transmitter. Basic software for checkup and analysis of sound is delivered together with electronic stethoscope and provides phonocardiograph and spectral presentation of auscultative finding. In further qualitative analysis, phonocardiogram in digital form (format *. е4k) is transformed into standard*.wav format, which is the first step in the processing of digital signal, study and testing of ANN. The obtained precision of MVP classification category was 71.2%, while total precision of category identification of 73.62% was the result of averaging all four cross-validation steps. These results can be interesting in phonocardiograph MVP diagnostics of children and adolescents.
Key words: phonocardiography – mitral valve prolapse –neural networkReferences
Ostojić M, Vladimir K, Beleslin B. Kardiologija. Beograd: Zavod za udžbenike, 2011.
Završnik J, Malčić, I. Mitral valve prolapse and mitral valve prolapse sindrome in children / Pedijatrijska kardiologija, odabrana poglavlja - 2. dio / Ivan Malčić, (ed). Zagreb: Medicinska naklada, 2003.
Rajakumar K, Weisse M, Rosas A, et al. Comparative study of clinical evaluation of heart murmurs by general pediatricians and pediatric cardiologists. Clin Pediatr 1999; 38: 511-18.
Becket MC, Nowalk A, Hofkosh D, Zuberbuhler JR, Law YM. Comparison of Two Educational Interventions on Pediatric Resident Auscultation Skills. Pediatrics 2004; 113: 1331-35
Sztajzelag JM, Kossovsky MP, Lerchab R, Vuilleac C, Sarasind FP. Accuracy of cardiac auscultation in the era of Doppler-echocardiography: A comparison between cardiologists and internists. Cardiology 2008; 138: 308-10.
Tavel ME. Cardiac auscultation: a glorious past: and it does have a future. Circulation 2006; 113:1255-59.
Pelech N. The physiology of cardiac auscultation. Pediatr Clin North Am 2004; 51:1515–35.
Haykin S. Neural Network: A Comprehensive Foundation. Pearson, 1999.
Hu YH, Hwang J. Handbook of Neural Network Signal Processing. CRC Press, 2001.
Bhatikar SR, DeGroff C and Mahajan RL. A classifier based on the artificial neural network approach for cardiologic auscultation in pediatrics. Artificial Intelligence in Medicine. 2005; 33: 251–60.
Reed TR, Reed NE, and Fritzson P. Heart sound analysis for symptom detection and computer-aided diagnosis. Simulation Modelling Practice and Theory. 2004; 12: 129-46.
Ari S, Hembram K, Saha G. Detection of cardiac abnormality from PCG signal using LMS based least square SVM classifier. Expert Systems with Applications: An International Journal 2010; 37: 8019-26.