Analysis of Coronary Heart Diseases by Kinetic Features: Applying Variational Mode Decomposition to ECG Signals and Classification Using Machine Learning Algorithms

Öz

This study presents an approach for the diagnosis of myocardial infarction (MI) and other coronary heart diseases using 12-lead electrocardiogram (ECG) signals. In the presented approach, 12-lead ECG signals recordings of MI types (STEMI-NSTEMI), other heart diseases (OHD) and healthy control (HC) participants, who presented to the Emergency Department of Erciyes University Hospital for heart disease, were used. In the first stage, the noise-cleaned ECG signals were decomposed into subbands by applying the Variational Mode Decomposition (VMD) method and kinetic features were obtained, and the ones that would positively affect the performance of the classifiers were determined by Chi-square test. In the classification stage, these features were evaluated by Support Vector Machine (SVM), Random Forest (RF), and Artificial Neural Network (ANN) algorithms, and AUC, Accuracy, and Negative Predictive Value ratios were obtained. Classification procedures were performed for HC-OHD, HC-MI (NSTEMI+STEMI), and STEMI-NSTEMI-OHD groups. When evaluated in terms of AUC, rates that can be considered successful (80% and above) were obtained. The findings of this research may contribute to the systems that can be developed for the rapid and accurate diagnosis of coronary heart diseases from ECG signals, which can be difficult to interpret manually.

Anahtar Kelimeler

• Coronary heart disease
• 12-lead electrocardiogram (ECG) signal
• Kinetic features
• Variational mode decomposition
• Machine learning algorithms

Koroner Kalp Hastalıklarının Kinetik Özelliklere Göre Analizi: EKG Sinyallerine Varyasyonel Mod Ayrıştırmanın Uygulanması ve Makine Öğrenme Algoritmaları Kullanılarak Sınıflandırılması

Öz

Bu çalışmada 12 kanallı elektrokardiyogram (EKG) sinyalleri kullanılarak miyokard enfarktüsü (MI) ve diğer koroner kalp hastalıklarının tanısı için bir yaklaşım sunulmaktadır. Sunulan yaklaşımda Erciyes Üniversitesi Hastanesi Acil Servisine kalp rahatsızlığı nedeniyle başvuran MI tipleri (STEMI-NSTEMI), diğer kalp hastalıkları (OHD) ve sağlıklı kontrol (SK) katılımcılarının 12 kanallı EKG sinyalleri kayıtları kullanılmıştır. İlk aşamada, gürültüden arındırılmış EKG sinyalleri Varyasyonel Mod Ayrıştırma (VMD) yöntemi uygulanarak alt bantlara ayrıştırılmış ve kinetik özellikler elde edilmiş, sınıflandırıcıların performansını olumlu yönde etkileyecek olanlar Ki-kare testi ile belirlenmiştir. Sınıflandırma aşamasında bu özellikler Destek Vektör Makinesi (SVM), Rastgele Orman (RF) ve Yapay Sinir Ağı (YSA) algoritmaları ile değerlendirilmiş ve AUC, Doğruluk ve Negatif Tahmini Değer oranları elde edilmiştir. HC-OHD, HC-MI (NSTEMI+STEMI) ve STEMI-NSTEMI-OHD grupları için sınıflandırma işlemleri gerçekleştirildi. AUC açısından değerlendirildiğinde başarılı sayılabilecek oranlar (%80 ve üzeri) elde edildi. Bu araştırmanın bulguları, manuel olarak yorumlanması zor olabilen EKG sinyallerinden koroner kalp hastalıklarının hızlı ve doğru tanısı için geliştirilebilecek sistemlere katkıda bulunabilir.

Anahtar Kelimeler

• Koroner kalp hastalığı
• 12 derivasyonlu elektrokardiyogram (EKG) sinyali
• Kinetik özellikler
• Varyasyonel mod ayrıştırma
• Makine öğrenimi algoritmaları

Kaynakça

1. [1] World Health Organization. (2023). World Health Statistics 2023 Monitoring health for the SDGs Sustainable Development Goals HEALTH FOR ALL.
2. [2] Ansari, S., Farzaneh, N., Duda, M., Horan, K., Andersson, H. B., Goldberger, Z. D., ... & Najarian, K. (2017). A review of automated methods for detection of myocardial ischemia and infarction using electrocardiogram and electronic health records. IEEE reviews in biomedical engineering, 10, 264-298.
3. [3] Ribeiro, A. H., Ribeiro, M. H., Paixão, G. M., Oliveira, D. M., Gomes, P. R., Canazart, J. A., ... & Ribeiro, A. L. P. (2020). Automatic diagnosis of the 12-lead ECG using a deep neural network. Nature communications, 11(1), 1760.
4. [4] Chauhan, C., Tripathy, R. K., & Agrawal, M. (2024). Third-order tensor-based cardiac disease detection from 12-lead ECG signals using deep convolutional neural network. In Signal Processing Driven Machine Learning Techniques for Cardiovascular Data Processing (pp. 19-34). Academic Press.
5. [5] Schläpfer, J., & Wellens, H. J. (2017). Computer-interpreted electrocardiograms: benefits and limitations. Journal of the American College of Cardiology, 70(9), 1183-1192.
6. [6] Sun, Q., Wang, L., Li, J., Liang, C., Yang, J., Chen, Y., & Wang, C. (2024). Multi-phase ECG dynamic features for detecting myocardial ischemia and identifying its etiology using deterministic learning. Biomedical Signal Processing and Control, 88, 105498.
7. [7] Sadhukhan, D., Pal, S., & Mitra, M. (2018). Automated identification of myocardial infarction using harmonic phase distribution pattern of ECG data. IEEE Transactions on Instrumentation and Measurement, 67(10), 2303-2313.
8. [8] Zhang, J., Liu, M., Xiong, P., Du, H., Zhang, H., Lin, F., ... & Liu, X. (2021). A multi-dimensional association information analysis approach to automated detection and localization of myocardial infarction. Engineering Applications of Artificial Intelligence, 97, 104092.

9. [9] Dohare, A. K., Kumar, V., & Kumar, R. (2018). Detection of myocardial infarction in 12 lead ECG using support vector machine. Applied Soft Computing, 64, 138-147.
10. [10] Arif, M., Malagore, I. A., & Afsar, F. A. (2012). Detection and localization of myocardial infarction using k-nearest neighbor classifier. Journal of medical systems, 36, 279-289.
11. [11] Muminov, B., Nasimov, R., Mirzahalilov, S., Sayfullaeva, N., & Gadoyboyeva, N. (2020, May). Localization and classification of myocardial infarction based on artificial neural network. In 2020 Information Communication Technologies Conference (ICTC) (pp. 245-249). IEEE.
12. [12] Sahu, G., & Ray, K. C. (2021). An efficient method for detection and localization of myocardial infarction. IEEE Transactions on Instrumentation and Measurement, 71, 1-12.
13. [13] Anwar, S. M. S., Pal, D., Mukhopadhyay, S., & Gupta, R. (2024). A Lightweight Method of Myocardial Infarction Detection and Localization from Single Lead ECG Features Using Machine Learning Approach. IEEE Sensors Letters.
14. [14] Acharya, U. R., Fujita, H., Adam, M., Lih, O. S., Sudarshan, V. K., Hong, T. J., ... & San, T. R. (2017). Automated characterization and classification of coronary artery disease and myocardial infarction by decomposition of ECG signals: A comparative study. Information Sciences, 377, 17-29.
15. [15] Zeng, W., Yuan, J., Yuan, C., Wang, Q., Liu, F., & Wang, Y. (2020). Classification of myocardial infarction based on hybrid feature extraction and artificial intelligence tools by adopting tunable-Q wavelet transform (TQWT), variational mode decomposition (VMD) and neural networks. Artificial Intelligence in Medicine, 106, 101848.
16. [16] Sharma, L. D., & Sunkaria, R. K. (2018). Inferior myocardial infarction detection using stationary wavelet transform and machine learning approach. Signal, Image and Video Processing, 12(2), 199-206.
17. [17] Chakraborty, A., Chatterjee, S., Majumder, K., Shaw, R. N., & Ghosh, A. (2022). A comparative study of myocardial infarction detection from ECG data using machine learning. In Advanced Computing and Intelligent Technologies: Proceedings of ICACIT 2021 (pp. 257-267). Springer Singapore.
18. [18] Satty, A., Salih, M. M., Hassaballa, A. A., Gumma, E. A., Abdallah, A., & Khamis, G. S. M. (2024). Comparative Analysis of Machine Learning Algorithms for Investigating Myocardial Infarction Complications. Engineering, Technology & Applied Science Research, 14(1), 12775-12779.
19. [19] Maindarkar, P., & Reka, S. S. (2022, April). Machine Learning-Based Approach for Myocardial Infarction. In International Conference on Artificial Intelligence and Sustainable Engineering: Select Proceedings of AISE 2020, Volume 1 (pp. 17-27). Singapore: Springer Nature Singapore.
20. [20] Dragomiretskiy, K., & Zosso, D. (2013). Variational mode decomposition. IEEE transactions on signal processing, 62(3), 531-544.
21. [21] Maji, U., & Pal, S. (2016, September). Empirical mode decomposition vs. variational mode decomposition on ECG signal processing: A comparative study. In 2016 international conference on advances in computing, communications and informatics (ICACCI) (pp. 1129-1134). IEEE.
22. [22] Xie, L., Li, Z., Zhou, Y., He, Y., & Zhu, J. (2020). Computational diagnostic techniques for electrocardiogram signal analysis. Sensors, 20(21), 6318.
23. [23] Memiş, G., & Sert, M. (2019, April). Classification of Obstructive Sleep Apnea using Multimodal and Sigma-based Feature Representation. In 2019 27th Signal Processing and Communications Applications Conference (SIU) (pp. 1-4). IEEE.
24. [24] Azar, A. T., & El-Said, S. A. (2014). Performance analysis of support vector machines classifiers in breast cancer mammography recognition. Neural Computing and Applications, 24, 1163-1177.
25. [25] Xiong, P., Lee, S. M. Y., & Chan, G. (2022). Deep learning for detecting and locating myocardial infarction by electrocardiogram: A literature review. Frontiers in cardiovascular medicine, 9, 860032.
26. [26] Sraitih, M., Jabrane, Y., & Hajjam El Hassani, A. (2022). A robustness evaluation of machine learning algorithms for ECG myocardial infarction detection. Journal of Clinical Medicine, 11(17), 4935.
27. [27] Latifoğlu, F., Zhusupova, A., İnce, M., Ertürk, N. A., Özdet, B., İçer, S., ... & Kalay, N. (2024). Preliminary Study Based on Myocardial Infarction Classification of 12-Lead Electrocardiography Images with Deep Learning Methods. The European Journal of Research and Development, 4(1), 42-54.
28. [28] Jahmunah, V., Ng, E. Y. K., San, T. R., & Acharya, U. R. (2021). Automated detection of coronary artery disease, myocardial infarction and congestive heart failure using GaborCNN model with ECG signals. Computers in biology and medicine, 134, 104457.
29. [29] Acharya, U. R., Fujita, H., Sudarshan, V. K., Oh, S. L., Adam, M., Tan, J. H., ... & Chua, K. C. (2017). Automated characterization of coronary artery disease, myocardial infarction, and congestive heart failure using contourlet and shearlet transforms of electrocardiogram signal. Knowledge-Based Systems, 132, 156-166.
30. [30] Baloglu, U. B., Talo, M., Yildirim, O., San Tan, R., & Acharya, U. R. (2019). Classification of myocardial infarction with multi-lead ECG signals and deep CNN. Pattern recognition letters, 122, 23-30.
31. [31] Han, C., & Shi, L. (2020). ML–ResNet: A novel network to detect and locate myocardial infarction using 12 leads ECG. Computer methods and programs in biomedicine, 185, 105138.
32. [32] Barandas, M., Famiglini, L., Campagner, A., Folgado, D., Simão, R., Cabitza, F., & Gamboa, H. (2024). Evaluation of uncertainty quantification methods in multi-label classification: A case study with automatic diagnosis of electrocardiogram. Information Fusion, 101, 101978.