Detection of Different Cardiac Conditions with Machine Learning Using Wavelet Transform and GLCM Feature Fusion in ECG Images

Öz

Cardiovascular diseases (CVDs) are the leading cause of mortality worldwide, accounting for 32% of global deaths. Electrocardiography (ECG) is a widely used, cost-effective, and non-invasive diagnostic tool for detecting cardiac abnormalities. However, ECG interpretation remains challenging due to noise interference, physiological variations, and the need for expert evaluation. This study proposes a machine learning-based approach for automatic classification of cardiac conditions using ECG images. The methodology involves feature extraction using Wavelet Transform (WT) and Gray-Level Co-occurrence Matrix (GLCM), followed by feature fusion to enhance classification. A total of 928 ECG images from four categories—Myocardial Infarction (MI), Abnormal Heartbeat (ABH), History of MI (HMI), and Normal—were analyzed. The extracted features were classified using XGBoost, Random Forest, Support Vector Machine, K-Nearest Neighbors, Decision Tree, and Logistic Regression. Results showed that XGBoost achieved the highest accuracy (93.55%), followed by Random Forest (93.01%), outperforming conventional methods. The findings suggest that feature fusion enhances classification and offers an interpretable, computationally efficient alternative to deep learning. This study contributes to automated cardiac diagnostics by providing a robust framework suitable for clinical applications and wearable ECG systems.

Anahtar Kelimeler

• Cardiac detection
• ECG classification
• GLCM
• Machine learning
• Wavelet transform

EKG Görüntülerinde Dalgacık Dönüşümü ve GLCM Özellik Füzyonu Kullanarak Farklı Kardiyak Durumların Makine Öğrenmesi ile Tespiti

Öz

Kardiyovasküler hastalıklar (KVH), küresel ölümlerin %32’sinden sorumlu olup, en yaygın ölüm nedenidir. Elektrokardiyografi (EKG), kardiyak anormalliklerin tespitinde yaygın kullanılan, düşük maliyetli ve non-invaziv bir yöntemdir. Ancak gürültü, bireysel fizyolojik farklılıklar ve uzman değerlendirme gereksinimi EKG yorumlamada zorluk yaratmaktadır. Bu çalışmada, EKG görüntüleriyle farklı kardiyak durumları otomatik sınıflandıran yeni bir makine öğrenmesi yöntemi önerilmektedir. Yöntem kapsamında Dalgacık Dönüşümü (WT) ve Gri-Seviye Eş-Oluşum Matrisi (GLCM) ile özellik çıkarımı yapılmış, özellik füzyonu gerçekleştirilmiştir. 928 EKG görüntüsü dört kategori Miyokard Enfarktüsü – (MI), Anormal Kalp Atışı – (ABH), Geçmiş MI – (HMI), Normal içinde analiz edilmiştir. Çıkarılan özellikler XGBoost, Random Forest, Destek Vektör Makineleri, K-En Yakın Komşu, Karar Ağacı ve Lojistik Regresyon ile sınıflandırılmıştır. Sonuçlar, XGBoost’un %93.55 doğrulukla en iyi performansı sergilediğini, onu %93.01 ile Random Forest modelinin takip ettiğini göstermiştir. Bulgular, önerilen özellik füzyonunun sınıflandırma başarısını artırdığını ve derin öğrenmeye kıyasla daha yorumlanabilir, hesaplama açısından verimli bir alternatif sunduğunu göstermektedir. Çalışma, otomatik kardiyak tanı sistemlerine katkıda bulunarak klinik uygulamalara ve taşınabilir EKG cihazlarına entegre edilebilir bir makine öğrenmesi çerçevesi sunmaktadır.

Anahtar Kelimeler

• Dalgacık dönüşümü
• EKG sınıflandırma
• GLCM
• Kardiyak tespit
• Makine öğrenmesi

Kaynakça

1. Timmis, A., Group on behalf of the AW, Vardas, P., et al. (2022). European Society of Cardiology: cardiovascular disease statistics 2021. Eur Heart J., 43(8), 716–799. https://doi.org/10.1093/EURHEARTJ/EHAB892
2. Zanchi, B., Monachino, G., Fiorillo, L., et al. (2025). Synthetic ECG signals generation: A scoping review. Comput Biol Med, 184, 109453. https://doi.org/10.1016/J.COMPBIOMED.2024.109453
3. Kaplan Berkaya, S., Uysal, A.K., Sora Gunal, E., et al. (2018). A survey on ECG analysis. Biomed Signal Process Control, 43, 216–235. https://doi.org/10.1016/J.BSPC.2018.03.003
4. Lopez-Jimenez, F., Attia, Z., Arruda-Olson, A.M., et al. (2020). Artificial Intelligence in Cardiology: Present and Future. Mayo Clin Proc, 95(5), 1015–1039. https://doi.org/10.1016/J.MAYOCP.2020.01.038
5. Oke, O.A., Cavus, N. (2025). Electrocardiogram image classification for six classes of heart diseases. Iran Journal of Computer Science, 2025, 1–21. https://doi.org/10.1007/S42044-025-00227-X
6. Mhamdi, L., Dammak, O., Cottin, F., Dhaou, I. (2022). Artificial Intelligence for Cardiac Diseases Diagnosis and Prediction Using ECG Images on Embedded Systems. Biomedicines, 10, 2013. https://doi.org/10.3390/BIOMEDICINES10082013
7. Sadad, T., Safran, M., Khan, I., et al. (2023). Efficient Classification of ECG Images Using a Lightweight CNN with Attention Module and IoT. Sensors, 23, 7697. https://doi.org/10.3390/S23187697
8. Ashtaiwi, A.A., Khalifa, T., Alirr, O. (2024). Enhancing heart disease diagnosis through ECG image vectorization-based classification. Heliyon, 10(18), e37574. https://doi.org/10.1016/j.heliyon.2024.e37574

9. Sattar, S., Mumtaz, R., Qadir, M., et al. (2024). Cardiac Arrhythmia Classification Using Advanced Deep Learning Techniques on Digitized ECG Datasets. Sensors, 24, 2484. https://doi.org/10.3390/S24082484
10. Aversano, L., Bernardi, M.L., Cimitile, M., et al. (2023). Early Diagnosis of Cardiac Diseases using ECG Images and CNN-2D. Procedia Comput Sci, 225, 2866–2875. https://doi.org/10.1016/J.PROCS.2023.10.279
11. Aversano, L., Bernardi, M.L., Cimitile, M., et al. (2024). Characterization of Heart Diseases per Single Lead Using ECG Images and CNN-2D. Sensors, 24(11), 3485. https://doi.org/10.3390/S24113485
12. Mohanty, M.N., Baliarsingh, S., Panda, P.K. (2025). An Ensemble Technique for Cardiac Data Compression in Smart Healthcare System. SN Comput Sci, 6(1), 78. https://doi.org/10.1007/s42979-024-03605-7
13. Venkataiah, Dr.V., Mamatha, B. (2024). Detection of Cardiovascular Diseases in ECG Images Using Machine Learning and Deep Learning Methods. International Journal of Information Technology and Computer Engineering, 12(3), 478–494.
14. Attallah, O. (2022). An Intelligent ECG-Based Tool for Diagnosing COVID-19 via Ensemble Deep Learning Techniques. Biosensors (Basel), 12(5), 299. https://doi.org/10.3390/BIOS12050299/S1
15. Rahman, T., Akinbi, A., Chowdhury, M.E.H., et al. (2022). COV-ECGNET: COVID-19 detection using ECG trace images with deep convolutional neural network. Health Inf Sci Syst, 10(1), 1–16. https://doi.org/10.1007/S13755-021-00169-1/FIGURES/12
16. Khan, A.H., Hussain, M., Malik, M.K. (2021). ECG Images dataset of Cardiac and COVID-19 Patients. Data Brief, 34, 106762. https://doi.org/10.1016/J.DIB.2021.106762
17. Khan, A.H., Hussain, M. (2021). ECG Images dataset of Cardiac Patients. 2. https://doi.org/10.17632/GWBZ3FSGP8.2
18. Donoho, D.L., Johnstone, J.M. (1994). Ideal spatial adaptation by wavelet shrinkage. Biometrika, 81(3), 425–455. https://doi.org/10.1093/BIOMET/81.3.425
19. Lee, H., Yoon, T., Yeo, C., et al. (2021). Cardiac Arrhythmia Classification Based on One-Dimensional Morphological Features. Applied Sciences, 11(20), 9460. https://doi.org/10.3390/APP11209460
20. Mallat, S. (2008). A Wavelet Tour of Signal Processing.
21. Pedregosa, F., Weiss, R., Brucher, M., et al. (2011). Scikit-learn: Machine Learning in Python. Journal of Machine Learning Research, 12, 2825–2830.