نشریه علمی مهندسی پزشکی زیستی
Iranian Journal of Biomedical Engineering (IJBME)

طبقه‌بندی صداهای طبیعی از صداهای غیرطبیعی قلب با استفاده از روش‌های مبتنی بر یادگیری ماشین

نوع مقاله : مقاله کامل پژوهشی

نویسندگان

پرستو صادقی‌نیا 1
حامد داننده حصار 2

1 دانشجوی کارشناسی ارشد، گروه بیوالکتریک، دانشکده‌ی مهندسی پزشکی، دانشگاه صنعتی سهند، تبریز، ایران

2 استادیار، گروه بیوالکتریک، دانشکده‌ی مهندسی پزشکی، دانشگاه صنعتی سهند، تبریز، ایران

https://doi.org/10.22041/ijbme.2023.1987058.1821
چکیده
سیگنال‌های فونوکاردیوگرافی (PCG) اطلاعات ارزشمندی در مورد عمل‌کرد دریچه‌های قلبی دارند. از این رو این سیگنال‌ها می‌توانند در تشخیص زودهنگام بیماری‌های قلبی مفید واقع شوند. طبقه‌بندی خودکار صدای قلب دارای پتانسیل امیدوار کننده‌ای در آسیب‌شناسی قلبی است. در این پژوهش روشی خودکار برای تشخیص صداهای طبیعی از غیرطبیعی قلب پیشنهاد شده است. در روش پیشنهادی ابتدا صداهای قلبی به چهار بخش صدای S1، S2، سیستول و دیاستول قطعه‌بندی شده و سپس ویژگی‌های زمانی آماری و زمانی فرکانسی از هر کدام از این بخش‌ها استخراج شده است. پیش از عملیات طبقه‌بندی داده‌ها، از دو ره‌یافت برای انتخاب ویژگی‌های موثر استفاده شده است. انتخاب ویژگی در ره‌یافت اول با استفاده از الگوریتم بهینه‌سازی ازدحام ذرات (PSO) و در ره‌یافت دوم با استفاده از جست‌وجوی سلسله مراتبی (SFFS) انجام شده است. روش پیشنهادی روی پایگاه داده‌ی چالش 2016 فیزیونت ارزیابی شده و در نهایت عمل‌کرد آن با استفاده از روش اعتبارسنجی متقابل 10-لایه‌ای مورد ارزیابی قرار گرفته است. هم‌چنین به دلیل نامتوازن بودن تعداد صداهای طبیعی نسبت به صداهای غیرطبیعی، از تکنیک بیش‌نمونه‌برداری اقلیت مصنوعی (SMOTE) برای تولید مجموعه‌ی داده‌های متعادل استفاده شده است. نتایج ارزیابی روی پایگاه داده نشان داده که روش پیشنهادی دارای صحت 03/98%، حساسیت 64/97% و اختصاصیت 43/98% در تشخیص صداهای طبیعی از غیرطبیعی است.

کلیدواژه‌ها

فونوکاردیوگرام
الگوریتم بهینه‌سازی ازدحام ذرات
انتخاب ویژگی
طبقه‌بندی

موضوعات

عنوان مقاله English

Classification of Normal and Abnormal Heart Sounds using Machine Learning Techniques

نویسندگان English

Parastoo Sadeghi Nia 1
Hamed Danandeh Hesar 2
1 M.Sc. Student, Faculty of Biomedical Engineering, Sahand University of Technology, Tabriz, Iran
2 Assistant Professor, Faculty of Biomedical Engineering, Sahand University of Technology, Tabriz, Iran
چکیده English

Phonocardiography (PCG) signals provide valuable information about the heart valves. These auditory signals can be useful in the early diagnosis of heart diseases. Automatic heart sound classification has a promising potential in the field of heart pathology. In this research, a new method based on machine learning techniques is proposed for discriminating normal and abnormal heart sounds. In this method, first, the heart sounds are segmented into 4 main parts: S1, S2, systole and diastole segments. From these segments, statistical and time-frequency features are extracted for classification. Before classification, the distinctive features are selected using two approaches. In the first approach, the feature selection is accomplished using particle swarm optimization algorithm (PSO).  In the second approach, we use Sequential Forward Feature Selection (SFFS) method. The proposed method was evaluated on the Physionet 2016 Challenge database using 10-fold cross-validation method. In this database, the number of normal and abnormal PCG signals are not balanced. Therefore, in this paper, the synthetic minority over-sampling technique (SMOTE) is applied to produce balanced data. The evaluation results showed that the proposed method can distinguish the normal heart sounds from abnormal ones with accuracy of 98.03% and sensitivity and specificity of 97.64% and 98.43% respectively.

کلیدواژه‌ها English

Phonocardiogram
Particle Swarm Optimization Algorithm
Feature Selection
Classification
  1. Rezaee and J. Haddadnia, "Design and performance evaluation of intelligent system to segregate and classify the phonocardiograph abnormalities using matched filter and multilayer perceptron-back propagation neural networks," Pajoohandeh Journal, vol. 18, no. 5, pp. 277-286, 2013.
  2. K. Abbas and R. Bassam, "Phonocardiography signal processing," Synthesis Lectures on Biomedical Engineering, vol. 4, no. 1, pp. 1-194, 2009.
  3. Nilanon, J. Yao, J. Hao, S. Purushotham, and Y. Liu, "Normal/abnormal heart sound recordings classification using convolutional neural network," in 2016 computing in cardiology conference (CinC), 2016: IEEE, pp. 585-588.
  4. Rubin, R. Abreu, A. Ganguli, S. Nelaturi, I. Matei, and K. Sricharan, "Classifying heart sound recordings using deep convolutional neural networks and mel-frequency cepstral coefficients," in 2016 Computing in cardiology conference (CinC), 2016: IEEE, pp. 813-816.
  5. Mei, H. Wang, Y. Zhang, F. Liu, X. Jiang, and S. Wei, "Classification of heart sounds based on quality assessment and wavelet scattering transform," Computers in Biology and Medicine, vol. 137, p. 104814, 2021.
  6. A. Goda and P. Hajas, "Morphological determination of pathological PCG signals by time and frequency domain analysis," in 2016 computing in cardiology conference (CinC), 2016: IEEE, pp. 1133-1136.
  7. Milani, P. E. Abas, L. C. De Silva, and N. D. Nanayakkara, "Abnormal heart sound classification using phonocardiography signals," Smart Health, vol. 21, p. 100194, 2021.
  8. Chen, S. Wei, and Y. Zhang, "Classification of heart sounds based on the combination of the modified frequency wavelet transform and convolutional neural network," Medical & Biological Engineering & Computing, vol. 58, no. 9, pp. 2039-2047, 2020.
  9. J. G. Ortiz, C. P. Phoo, and J. Wiens, "Heart sound classification based on temporal alignment techniques," in 2016 computing in cardiology conference (CinC), 2016: IEEE, pp. 589-592.
  10. Tschannen, T. Kramer, G. Marti, M. Heinzmann, and T. Wiatowski, "Heart sound classification using deep structured features," in 2016 Computing in Cardiology Conference (CinC), 2016: IEEE, pp. 565-568.
  11. Li, H. Tang, S. Shang, K. Mathiak, and F. Cong, "Classification of heart sounds using convolutional neural network," Applied Sciences, vol. 10, no. 11, p. 3956, 2020.
  12. Nabih-Ali, E.-S. A. El-Dahshan, and A. S. Yahia, "Heart diseases diagnosis using intelligent algorithm based on PCG signal analysis," International Journal of Biology and Biomedicine, vol. 2, 2017.
  13. D. Clifford et al., "Classification of normal/abnormal heart sound recordings: The PhysioNet/Computing in Cardiology Challenge 2016," in 2016 Computing in cardiology conference (CinC), 2016: IEEE, pp. 609-612.
  14. B. Springer, L. Tarassenko, and G. D. Clifford, "Logistic regression-HSMM-based heart sound segmentation," IEEE transactions on biomedical engineering, vol. 63, no. 4, pp. 822-832, 2015.
  15. Liu et al., "An open access database for the evaluation of heart sound algorithms," Physiological Measurement, vol. 37, no. 12, p. 2181, 2016.
  16. Leatham, Auscultation of the Heart and Phonocardiography. Churchill London, 1970.
  17. A. Innes, A. R. Dover, and K. Fairhurst, Macleod's clinical examination. Elsevier Health Sciences, 2018.
  18. N. Homsi et al., "Automatic heart sound recording classification using a nested set of ensemble algorithms," in 2016 Computing in Cardiology Conference (CinC), 2016: IEEE, pp. 817-820.
  19. Wang, D. Li, Y. Wei, and H. Li, "A feature selection method based on fisher’s discriminant ratio for text sentiment classification," in International Conference on Web Information Systems and Mining, 2009: Springer, pp. 88-97.
  20. Sarabi, M. Asadnejad, and S. Rajabi, "Using neural network for drowsiness detection based on EEG signals and optimization in the selection of its features using genetic algorithm," Innovaciencia Facultad de Ciencias Exactas Físicas y Naturales, vol. 8, no. 1, pp. 1-9, 2020.
  21. Shi, "Particle swarm optimization," IEEE connections, vol. 2, no. 1, pp. 8-13, 2004.
  22. Umapathy, C. Venkataseshaiah, and M. S. Arumugam, "Particle swarm optimization with various inertia weight variants for optimal power flow solution," Discrete Dynamics in Nature and Society, vol. 2010, 2010.
  23. He, W. J. Ma, and J. P. Zhang, "The parameters selection of PSO algorithm influencing on performance of fault diagnosis," in MATEC Web of conferences, 2016, vol. 63: EDP Sciences, p. 02019.
  24. Shirbani and H. Soltanian Zadeh, "Fast SFFS-based algorithm for feature selection in biomedical datasets," AUT Journal of Electrical Engineering, vol. 45, no. 2, pp. 43-56, 2013.
  25. Ashok and P. Aruna, "Comparison of Feature selection methods for diagnosis of cervical cancer using SVM classifier," Int. J. Eng. Res. Appl, vol. 6, pp. 94-99, 2016.
  26. Wang, Support vector machines: theory and applications. Springer Science & Business Media, 2005.
  27. M. Bishop and N. M. Nasrabadi, Pattern recognition and machine learning (no. 4). Springer, 2006.
  28. Dey, A. S. Ashour, F. Shi, and V. E. Balas, Soft Computing Based Medical Image Analysis. Academic Press, 2018.
  29. V. Chawla, K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer, "SMOTE: synthetic minority over-sampling technique," Journal of artificial intelligence research, vol. 16, pp. 321-357, 2002.
  30. Potes, S. Parvaneh, A. Rahman, and B. Conroy, "Ensemble of feature-based and deep learning-based classifiers for detection of abnormal heart sounds," in 2016 computing in cardiology conference (CinC), 2016: IEEE, pp. 621-624.
  31. M. Nogueira, C. A. Ferreira, and A. M. Jorge, "Classifying heart sounds using images of MFCC and temporal features," in EPIA Conference on Artificial Intelligence, 2017: Springer, pp. 186-203.
  32. Hazeri, P. Zarjam, and G. Azemi, "Classification of normal/abnormal PCG recordings using a time–frequency approach," Analog Integrated Circuits and Signal Processing, vol. 109, no. 2, pp. 459-465, 2021.
نشریه علمی مهندسی پزشکی زیستی
دوره 16، شماره 3
پاییز 1401
صفحه 257-270

  • تاریخ دریافت 19 دی 1401
  • تاریخ بازنگری 08 اسفند 1401
  • تاریخ پذیرش 15 اسفند 1401

صفحه اصلی

تماس با ما