Performance Analysis of Nonlinear Features in Detection of Myocardial Infarction Patients

Year 2024, Volume: 24 Issue: 06, 1499 - 1505

Ali Narin , Merve Keser

Abstract

Myocardial infarction (MI), one of the heart diseases, is a condition in which the heart muscle is damaged as a result of partial or complete interruption of blood flow to the regions of the heart. This condition causes permanent damage to the heart and poses a life-threatening risk. Electrocardiogram (ECG) signals, which can be obtained easily and cheaply, are used by experts for the detection of MI. However, MI-related abnormalities on some ECG signals may be overlooked or even interpreted differently. Work continues on automatic MI detection with artificial intelligence-based decision support systems in order to solve the problems encountered. In this study, lead-II derivation from 12-lead ECG signals of 52 normal and 148 MI individuals was analyzed. By using the features obtained by five different methods, namely Shannon entropy, Renyi entropy, Wavelet entropy, Kolmogorov-Sinai entropy and Fuzzy entropy, the performances in detecting healthy and MI were investigated. The performances of each entropy measure on noisy and noiseless ECG signals are compared. Performances on MI detection were analyzed using k-nearest neighbor (kNN), Naive Bayes and Ensemble classifier algorithms. As a result of the classification of the features obtained from five different methods, the highest accuracy value belongs to Fuzzy entropy with 87.72%. This value is obtained as a result of using kNN classifier on noisy signals. By classifying all features together, 90.99% overall accuracy, 95.58% sensitivity and 71.55% specificity values were obtained. This highest value was obtained as a result of the use of noisy signal and Ensemble classifier.

Keywords

Myocardial Infarction, Electrocardiogram, Feature extraction, Ensemble classifier

Miyokard Enfarktüsü Hastalarının Tespitinde Doğrusal Olmayan Özniteliklerin Performans Analizi

Abstract

Kalp rahatsızlıklarından biri olan Miyokard enfarktüsü (ME), kalbin bölgelerine kısmen veya tamamen kan akışının kesilmesi sonucunda kalp kaslarına zarar vermesi durumudur. Bu durum kalbe kalıcı hasar vermekte ve hayati risk oluşturmaktadır. ME tespiti için kolay ve ucuz elde edilebilen elektrokardiyogram (EKG) sinyalleri uzmanlar tarafından kullanılmaktadır. Fakat, bazı EKG sinyalleri üzerinde ME ile ilişkili anormallikler gözden kaçırılabilmekte hatta farklı yorumlanabilmektedir. Karşılaşılan problemlere çözüm olması amacıyla yapay zekâ tabanlı karar destek sistemleri ile otomatik ME tespiti üzerinde çalışmalar devam etmektedir. Bu çalışmada 52 sağlıklı ve 148 ME bireye ait 12 derivasyonlu EKG sinyallerinden lead-II derivasyonu analiz edilmiştir. Shannon entropi, Renyi entropi, Dalgacık entropi, Kolmogorov-Sinai entropi ve Bulanık entropi olmak üzere beş farklı yöntem ile elde edilen öznitelikler kullanılarak sağlıklı ve ME tespitindeki başarımlar araştırılmıştır. Her bir entropi ölçümünün gürültülü ve gürültüsüz EKG sinyalleri üzerinde performansları karşılaştırılmıştır. K-en yakın komşu (kNN), Naive Bayes ve Topluluk sınıflandırıcı algoritmaları kullanılarak ME tespiti üzerinde performansları analiz edilmiştir. Beş farklı yöntemden elde edilen özniteliklerin sınıflandırılması sonucu en yüksek doğruluk değeri %87,72 ile Bulanık entropi kullanılarak elde edilmiştir. Bu değer, gürültülü sinyallerin üzerinde kNN sınıflandırıcısının kullanılması sonucunda elde edilmiştir. Tüm özniteliklerin birlikte kullanılarak sınıflandırılması ile %90,99 genel doğruluk, %95,58 hassasiyet, %71,55 özgünlük değerleri elde edilmiştir. En yüksek bu değer, gürültülü sinyal ve Topluluk sınıflandırıcı kullanımı sonucunda elde edilmiştir.

Keywords

Miyokard enfarktüsü, Elektrokardiyogram, Öznitelik çıkarma, Topluluk sınıflandırıcı

References

• Acharya, U. R., Fujita, H., Sudarshan, V. K., Oh, S. L., Adam, M., Koh, J. E. and San Tan, R., 2016. Automated detection and localization of myocardial infarction using electrocardiogram: a comparative study of different leads. Knowledge-Based Systems, 99, 146-156. https://doi.org/10.1016/j.knosys.2016.01.040
• Arif, M., Malagore, I. A., and Afsar, F. A. 2012. Detection and localization of myocardial infarction using k-nearest neighbor classifier. Journal of Medical Systems, 36, 279-289. https://doi.org/10.1007/s10916-010-9474-3
• Attallah, O., and Ragab, D. A. 2023. Auto-MyIn: Automatic diagnosis of myocardial infarction via multiple GLCMs, CNNs, and SVMs. Biomedical Signal Processing and Control, 80, 104273. https://doi.org/10.1016/j.bspc.2022.104273
• Benjamin, E. J., Muntner, P., Alonso, A., Bittencourt, M. S., Callaway, C. W., Carson, A. P., and American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee., 2019. Heart disease and stroke statistics—2019 update: a report from the American Heart Association. Circulation, 139(10), e56-e528. https://doi.org/10.1161/CIR.0000000000000659
• Bousseljot, R., Kreiseler, D. and Schnabel, A., 2004. The PTB diagnostic ECG database. physionet. org.
• Chang, P. C., Lin, J. J., Hsieh, J. C. and Weng, J. 2012. Myocardial infarction classification with multi-lead ECG using hidden Markov models and Gaussian mixture models. Applied Soft Computing, 12(10), 3165-3175. https://doi.org/10.1016/j.asoc.2012.06.004 Chen, Z., Lalande, A., Salomon, M., Decourselle, T., Pommier, T., Qayyum, A. and Couturier, R. 2022. Automatic deep learning-based myocardial infarction segmentation from delayed enhancement MRI. Computerized Medical Imaging and Graphics, 95, 102014. https://doi.org/10.1016/j.compmedimag.2021.10204
• Degerli, A., Kiranyaz, S., Hamid, T., Mazhar, R., & Gabbouj, M. (2024). Early myocardial infarction detection over multi-view echocardiography. Biomedical Signal Processing and Control, 87, 105448. https://doi.org/10.1016/j.bspc.2023.105448
• Diker, A., Cömert, Z., Avci, E. and Velappan, S., 2018. Intelligent system based on Genetic Algorithm and support vector machine for detection of myocardial infarction from ECG signals. In 2018 26th Signal processing and communications applications conference (SIU) (pp. 1-4). IEEE. https://doi.org/10.1109/SIU.2018.8404299
• Gong, M., Liang, D., Xu, D., Jin, Y., Wang, G. and Shan, P., 2024. Analyzing predictors of in-hospital mortality in patients with acute ST-segment elevation myocardial infarction using an evolved machine learning approach. Computers in Biology and Medicine, 107950. https://doi.org/10.1016/j.compbiomed.2024.107950
• Han, C., and Shi, L., 2019. Automated interpretable detection of myocardial infarction fusing energy entropy and morphological features. Computer Methods and Programs in Biomedicine, 175, 9-23. https://doi.org/10.1016/j.cmpb.2019.03.012
• Hasbullah, S., Mohd Zahid, M. S., and Mandala, S., 2023. Detection of Myocardial Infarction Using Hybrid Models of Convolutional Neural Network and Recurrent Neural Network. BioMedInformatics, 3(2), 478-492. https://doi.org/10.3390/biomedinformatics3020033
• Kumar, M., Pachori, R. B., & Acharya, U. R., 2017. Automated diagnosis of myocardial infarction ECG signals using sample entropy in flexible analytic wavelet transform framework. Entropy, 19(9), 488. https://doi.org/10.3390/e19090488
• Liu, B., Liu, J., Wang, G., Huang, K., Li, F., Zheng, Y. and Zhou, F., 2015. A novel electrocardiogram parameterization algorithm and its application in myocardial infarction detection. Computers in Biology and Medicine, 61, 178-184. https://doi.org/10.1016/j.compbiomed.2014.08.010
• Miranda, D. F., Lobo, A. S., Walsh, B., Sandoval, Y., and Smith, S. W., 2018. New insights into the use of the 12-lead electrocardiogram for diagnosing acute myocardial infarction in the emergency department. Canadian Journal of Cardiology, 34(2), 132-145. https://doi.org/10.1016/j.cjca.2017.11.011
• Narin, A., 2022. Detection of focal and non-focal epileptic seizure using continuous wavelet transform-based scalogram images and pre-trained deep neural networks. IRBM, 43(1), 22–31. https://doi.org/10.1016/j.irbm.2020.11.002
• Padhy, S., and Dandapat, S., 2017. Third-order tensor based analysis of multilead ECG for classification of myocardial infarction. Biomedical Signal Processing and Control, 31, 71-78. https://doi.org/10.1016/j.bspc.2016.07.007
• Papaloukas, C., Fotiadis, D. I., Likas, A., & Michalis, L. K., 2002. An ischemia detection method based on artificial neural networks. Artificial Intelligence in Medicine, 24(2), 167-178. https://doi.org/10.1016/S0933-3657(01)00100-2
• Ramer, A., 1990. Concepts of fuzzy information measures on continuous domains. International Journal Of General System, 17(3), 241-248. https://doi.org/10.1080/03081079008935109 Rényi, A., 1961. On measures of entropy and information. In Proceedings of the Fourth Berkeley Symposium on Mathematical Statistics and Probability, Volume 1: Contributions to the Theory of Statistics. University of California Press, 4, 547-562.
• Shannon, C. E. 1948. A mathematical theory of communication. The Bell System Technical Journal, 27(3), 379-423. https://doi.org/10.1002/j.15387305.1948.tb0138.x
• Sharma, L. N., Tripathy, R. K. and Dandapat, S., 2015. Multiscale energy and eigenspace approach to detection and localization of myocardial infarction. IEEE Transactions on Biomedical Engineering, 62(7), 1827-1837. https://doi.org/10.1109/TBME.2015.2405134
• Subha, D. P., Joseph, P. K., Acharya U, R. And Lim, C. M., 2010. EEG signal analysis: a survey. Journal of Medical Systems, 34, 195-212. https://doi.org/10.1007/s10916-008-9231-z
• Savaré, G., & Toscani, G., 2014. The concavity of Rényi entropy power. IEEE Transactions on Information Theory, 60(5), 2687-2693. https://doi.org/10.1109/TIT.2014.2309341
• Sharma, L. D. and 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. https://doi.org/10.1007/s11760-017-1146-z
• Sopic, D., Aminifar, A., Aminifar, A., and Atienza, D., 2018. Real-time event-driven classification technique for early detection and prevention of myocardial infarction on wearable systems. IEEE Transactions on Biomedical Circuits and Systems, 12(5), 982-992. https://doi.org/10.1109/TBCAS.2018.2848477