Classification of Heartbeat Sounds Using Deep Learning

Yıl 2024, Cilt: 13 Sayı: 4, 1927 - 1933, 25.12.2024

Gökhan Tutar , Serdar Aydın

https://doi.org/10.37989/gumussagbil.1580777

Öz

Death rates due to heart disease have a large place in the daily death rates in the world. For many years, doctors have first listened to the patient's heartbeat sound to diagnose heart diseases and have tried to make an initial diagnosis based on this data.
Understanding whether there is a disease from the heart sound is a difficult task because it requires experience. Classifying the heartbeat sound with deep learning is also a difficult task. Because both lung sounds and other external environmental sounds in the sounds taken with a stethoscope can cause misdiagnosis.
In this study, the publicly available library "Pascal Heart Sound Challenge" was used as the dataset. There are three categories in the dataset. These are; "Normal", "Murmur" and "Extra-systole".
In this study, it is aimed to predict the class of the heartbeat sound signal with a high degree of accuracy.

Anahtar Kelimeler

Deep learning, Heartbeat sound, CNN, SVM, Classification

Kalp Atış Seslerinin Derin Öğrenme Kullanarak Sınıflandırılması

Öz

Kalp hastalığına bağlı ölüm oranları dünyadaki günlük ölüm oranlarında büyük bir yer edinmektedir. Uzun yıllardır doktorlar kalp hastalıklarının teşhisi için ilk olarak hastanın kalp atış sesini dinlemekte ve bu veriye dayalı olarak hastaya ilk tanıyı koymaya çalışmaktadır.
Kalp sesinden hastalık olup olmadığını anlamak tecrübe gerektirdiği için zor bir iştir. Derin öğrenme ile kalp atış sesinin sınıflandırması da zor bir iştir. Çünkü stetoskop ile alınan seslerde hem akciğer sesi hem de diğer dış ortam sesleri yanlış tanıya neden olabilmektedir.
Bu çalışmada veri kümesi olarak halka açık bir kütüphane olan “Pascal Heart Sound Challenge” kullanılmıştır. Veri kümesinde üç kategori bulunmaktadır. Bunlar; “Normal”, “Murmur” ve “Extra-systole” dur.
Bu çalışmada kalp atışı ses sinyalinin hangi sınıfa ait olduğunu yüksek oranda doğru tahmin etmek amaçlanmaktadır.

Anahtar Kelimeler

Derin öğrenme, Kalp atış sesi, CNN, SVM, Sınıflandırma

Kaynakça

• 1. Somers VK, Dyken ME, Clary MP, et al. "Sympathetic neural mechanisms in obstructive sleep apnea". J Clin Invest 1995; 96: 1897–1904.
• 2. Cibo M, Brigic L, Tukulija S, et al. "Management and Invasive Cardiological Review by Comparison of Percutaneous Coronary Intervention in Left Anterior Descending Artery with Drug Eluting and Bare Metal Stents". Acta Informatica Medica 2018; 26: 130.
• 3. Hanna IR, Silverman ME. "A history of cardiac auscultation and some of its contributors". Am J Cardiol 2002; 90: 259–267.
• 4. Gomes EF, Bentley PJ, Pereira E, et al. "Classifying Heart Sounds-Approaches to the PASCAL Challenge". Healthinf 2013; 2013: 337–340.
• 5. Busono P, Karim S, Kamaruddin A, et al. "Heart Sound Signal Analysis for Digital Auscultation". J Phys Conf Ser 2022; 2377: 12024.
• 6. Kumar D, Carvalho P, Antunes M, et al. "Heart Murmur Classification With Feature Selection". Epub ahead of print 2010. DOI: 10.1109/iembs.2010.5625940.
• 7. Jiang Z, Choi S. "A cardiac sound characteristic waveform method for in-home heart disorder monitoring with electric stethoscope". Expert Syst Appl 2006; 31: 286–298.
• 8. Gomes EF, Pereira E. "Classifying heart sounds using peak location for segmentation and feature construction". In: Workshop Classifying Heart Sounds. 2012, pp. 480–492.
• 9. Pandey SK, Shukla A, Bhatia S, et al. "Detection of Arrhythmia Heartbeats From ECG Signal Using Wavelet Transform-Based CNN Model". International Journal of Computational Intelligence Systems. Epub ahead of print 2023. DOI: 10.1007/s44196-023-00256-z.
• 10. Pandey SK, Kumar G, Shukla S, et al. "Automatic Detection of Atrial Fibrillation From ECG Signal Using Hybrid Deep Learning Techniques". Journal of Sensors. Epub ahead of print 2022. DOI: 10.1155/2022/6732150.
• 11. Zheng Y, Guo X, Ding X. "A novel hybrid energy fraction and entropy-based approach for systolic heart murmurs identification". Expert Syst Appl 2015; 42: 2710–2721.
• 12. Deng S-W, Han J-Q. "Towards heart sound classification without segmentation via autocorrelation feature and diffusion maps". Future Generation Computer Systems 2016; 60: 13–21.
• 13. Zhang W, Han J, Deng S. "Heart sound classification based on scaled spectrogram and tensor decomposition". Expert Syst Appl 2017; 84: 220–231.
• 14. Yaseen, Son G-Y, Kwon S. "Classification of Heart Sound Signal Using Multiple Features". Applied Sciences 2018; 8: 2344.
• 15. Madani A, Ong JR, Tibrewal A, et al. "Deep Echocardiography: Data-Efficient Supervised and Semi-Supervised Deep Learning Towards Automated Diagnosis of Cardiac Disease". NPJ Digital Medicine. Epub ahead of print 2018. DOI: 10.1038/s41746-018-0065-x.
• 16. Raza A, Mehmood A, Ullah S, et al. "Heartbeat Sound Signal Classification Using Deep Learning". Sensors 2019, Vol 19, Page 4819 2019; 19: 4819.
• 17. Cortes C. "Support-Vector Networks". Mach Learn.
• 18. Schölkopf B. "SVMs - A practical consequence of learning theory". IEEE Intelligent Systems and Their Applications 1998; 13: 18–21.
• 19. Li F, Li X, Wang F, et al. "A Novel P300 Classification Algorithm Based on a Principal Component Analysis-Convolutional Neural Network". Applied Sciences. Epub ahead of print 2020. DOI: 10.3390/app10041546.
• 20. Liu Q, Liu B, Du Y. "An Algorithm to Improve the Performance of Convolutional Neural Networks for <scp>TSC</Scp> Tasks". Engineering Reports. Epub ahead of print 2022. DOI: 10.1002/eng2.12589.