DERİN ÖĞRENME VE MAKİNE ÖĞRENİMİ İLE EKG SİNYALİ SINIFLANDIRMASI: İKİ BOYUTLU ÇERÇEVELERDE QRS KOMPLEKSLERİNİN TEMSİLİ VE SMOTE İLE VERİ DENGELEME
Öz
Bu çalışma, bilgisayar destekli analiz yöntemleri aracılığıyla insan kaynaklı değişkenliği azaltarak güvenilir ECG sinyali yorumlamasını desteklemeyi amaçlamaktadır. Makine öğrenmesi ve derin öğrenme yöntemleri, ECG sınıflandırmasında iki boyutlu ECG temsillerini ve Synthetic Minority Over-sampling Technique (SMOTE) tabanlı veri dengelemenin etkisini incelemek amacıyla kullanılmıştır. Mevcut ECG sınıflandırma çalışmalarında genellikle sinyal temsili ve sınıf dengesizliği ayrı ayrı ele alınırken, bu çalışmada iki boyutlu QRS temsili ile SMOTE tabanlı veri dengelemenin etkileşimi tek bir deneysel çerçeve içerisinde birlikte incelenmiş ve bu yaklaşımların sınıflandırma performansı üzerindeki birleşik etkisi sistematik olarak analiz edilmiştir. Yapay Sinir Ağları (ANN), Evrişimsel Sinir Ağları (CNN) ve K-En Yakın Komşu (KNN) algoritmaları uygulanmış ve karşılaştırmalı olarak analiz edilmiştir. MIT-BIH Aritmi veri setinin 108 numaralı kaydından elde edilen ECG atımları, sınıflandırma amacıyla görsel tabanlı bir biçimde temsil edilmiştir. Ciddi sınıf dengesizliğini gidermek amacıyla SMOTE yalnızca eğitim verisine uygulanmış ve iki boyutlu ECG temsilleri üzerindeki etkisi açıkça incelenmiştir. Normal ve Anormal kalp atımları, tabakalı 5 katlı çapraz doğrulama stratejisi kullanılarak sınıflandırılmıştır. Deneysel sonuçlar, SMOTE uygulandıktan sonra CNN modelinin %99,82 ± 0,002 ağırlıklı ortalama F1-skoru ile en başarılı performansı elde ettiğini göstermiş; bu durum, iki boyutlu QRS temsili ile veri dengelemenin birlikte kullanımının otomatik ECG sınıflandırma performansını artırmadaki etkinliğini ortaya koymuştur.
Anahtar Kelimeler
Kaynakça
- Abbaskhah, A., Sedighi, H., & Marvi, H. (2023). Infant cry classification by MFCC feature extraction with MLP and CNN structures. Biomedical Signal Processing and Control, 86, 105261. doi:https://doi.org/10.1016/j.bspc.2023.105261
- Chamseddine, E., Mansouri, N., Soui, M., & Abed, M. (2022). Handling class imbalance in COVID-19 chest X-ray images classification: Using SMOTE and weighted loss. Applied Soft Computing, 129, 109588. doi:https://doi.org/10.1016/j.asoc.2022.109588
- Chawla, N. V., Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2002). SMOTE: synthetic minority over-sampling technique. Journal of artificial intelligence research, 16, 321-357. doi:https://doi.org/10.1613/jair.953
- Chen, Y., Chang, R., & Guo, J. (2021). Effects of data augmentation method borderline-SMOTE on emotion recognition of EEG signals based on convolutional neural network. IEEE Access, 9, 47491-47502. doi:https://doi.org/10.1109/access.2021.3068316
- Gonzalez-Huitron, V., León-Borges, J. A., Rodriguez-Mata, A., Amabilis-Sosa, L. E., Ramírez-Pereda, B., & Rodriguez, H. (2021). Disease detection in tomato leaves via CNN with lightweight architectures implemented in Raspberry Pi 4. Computers and Electronics in Agriculture, 181, 105951. doi:https://doi.org/10.1016/j.compag.2020.105951
- Goodfellow, I., Bengio, Y., Courville, A., & Bengio, Y. (2016). Deep learning. (Vol. 1): MIT press Cambridge. doi: https://doi.org/10.1007/s10710-017-9314-z.
- Guo, G., Wang, H., Bell, D., Bi, Y., & Greer, K. (2003). KNN model-based approach in classification. Paper presented at the OTM Confederated International Conferences" On the Move to Meaningful Internet Systems". doi: https://doi.org/10.1007/978-3-540-39964-3_62
- Kumar, S., & Panda, K. (2023). SDIF-CNN: Stacking deep image features using fine-tuned convolution neural network models for real-world malware detection and classification. Applied Soft Computing, 146, 110676. doi:https://doi.org/10.1016/j.asoc.2023.110676
- Nahiduzzaman, M., Goni, M. O. F., Hassan, R., Islam, M. R., Syfullah, M. K., Shahriar, S. M., . . . Kowalski, M. (2023). Parallel CNN-ELM: A multiclass classification of chest X-ray images to identify seventeen lung diseases including COVID-19. Expert Systems with Applications, 229, 120528. doi:https://doi.org/10.1016/j.eswa.2023.120528
- Onat, A., Şurdumavcı, G., Şenocak, M., Örnek, E., Gözükara, Y., Karaaslan, Y., . . . Özcan, R. (1991). Türkiye'de erişkinlerde kalp hastalığı ve risk faktörleri sıklığı taraması: 3. Kalp hastalıkları prevalansı. Türk Kardiyoloji Derneği Arşivi, 19(1), 26-33.
- Özcan, H. (2014). Çok düşük çözünürlüklü yüz imgelerinde derin öğrenme uygulamaları. Deniz Harp Okulu/Deniz Bilimleri ve Mühendisliği Enstitüsü, İstanbul, Erişim adresi: https://tez.yok.gov.tr/UlusalTezMerkezi/tezDetay.jsp.
- Peng, H. (2015). Air quality prediction by machine learning methods. University of British Columbia. doi: https://doi.org/10.14288/1.0166787,
- Ruuska, S., Hämäläinen, W., Kajava, S., Mughal, M., Matilainen, P., & Mononen, J. (2018). Evaluation of the confusion matrix method in the validation of an automated system for measuring feeding behaviour of cattle. Behavioural processes, 148, 56-62. doi:https://doi.org/10.1016/j.beproc.2018.01.004
- Si, T., Bagchi, J., & Miranda, P. B. (2022). Artificial neural network training using metaheuristics for medical data classification: an experimental study. Expert Systems with Applications, 193, 116423. doi:https://doi.org/10.1016/j.eswa.2021.116423
- Sun, P., Wang, Z., Jia, L., & Xu, Z. (2024). SMOTE-kTLNN: A hybrid re-sampling method based on SMOTE and a two-layer nearest neighbor classifier. Expert Systems with Applications, 238, 121848. doi:https://doi.org/10.1016/j.eswa.2023.121848
- Tuncer, S. A., Ayyıldız, H., Kalaycı, M., & Tuncer, T. (2021). Scat-NET: COVID-19 diagnosis with a CNN model using scattergram images. Computers in Biology and Medicine, 135, 104579. doi:https://doi.org/10.1016/j.compbiomed.2021.104579
- Ullah, H., Bin Heyat, M. B., Akhtar, F., Sumbul, Muaad, A. Y., Islam, M. S., . . . Lin, Y. (2022). An End‐to‐End Cardiac Arrhythmia Recognition Method with an Effective DenseNet Model on Imbalanced Datasets Using ECG Signal. Computational Intelligence and Neuroscience, 2022(1), 9475162. doi:https://doi.org/10.1155/2022/9475162
- Wu, Y.-c., & Feng, J.-w. (2018). Development and application of artificial neural network. Wireless Personal Communications, 102(2), 1645-1656. doi:https://doi.org/10.1007/s11277-017-5224-x
- Zeraatkar, S., & Afsari, F. (2021). Interval–valued fuzzy and intuitionistic fuzzy–KNN for imbalanced data classification. Expert Systems with Applications, 184, 115510. doi:https://doi.org/10.1016/j.eswa.2021.115510 Zhang, P., Zhang, X., & Liu, A. (2022). Effects of Data Augmentation with the BNNSMOTE Algorithm in Seizure Detection Using 1D‐MobileNet. Journal of Healthcare Engineering, 2022(1), 4114178. doi:https://doi.org/10.1155/2022/4114178
- Zhang, S., Cheng, D., Deng, Z., Zong, M., & Deng, X. (2018). A novel kNN algorithm with data-driven k parameter computation. Pattern Recognition Letters, 109, 44-54. doi:https://doi.org/10.1016/j.patrec.2017.09.036
ECG SIGNAL CLASSIFICATION WITH DEEP LEARNING AND MACHINE LEARNING: REPRESENTATION OF QRS COMPLEXES IN TWO-DIMENSIONAL FRAMES AND DATA BALANCING WITH SMOTE
Öz
This study aims to support reliable ECG signal interpretation by reducing human-dependent variability through computer-aided analysis methods. Machine learning and deep learning methods were employed to examine 2D ECG representations and Synthetic Minority Over-Sampling Technique (SMOTE)-based balancing in ECG classification. Unlike existing ECG classification studies that typically address signal representation and class imbalance separately, this study jointly investigates the interaction between two-dimensional QRS representation and SMOTE-based data balancing within a unified experimental framework, thereby providing a systematic analysis of their combined impact on classification performance. Artificial Neural Networks (ANN), Convolutional Neural Networks (CNN), and K-Nearest Neighbors (KNN) algorithms were implemented and comparatively analyzed. ECG beats from record 108 of the MIT-BIH Arrhythmia dataset were represented in a vision-based form for classification. To address severe class imbalance, SMOTE was applied only to the training data, and its effect on two-dimensional ECG representations was explicitly examined. Normal and Abnormal heartbeats were classified using a stratified 5-fold cross-validation strategy. Experimental results demonstrated that the CNN model achieved the most successful performance after applying SMOTE, reaching a weighted average F1-score of 99.82% ± 0.002, highlighting the combined effectiveness of two-dimensional QRS representation and data balancing in improving automated ECG classification.
Anahtar Kelimeler
Kaynakça
- Abbaskhah, A., Sedighi, H., & Marvi, H. (2023). Infant cry classification by MFCC feature extraction with MLP and CNN structures. Biomedical Signal Processing and Control, 86, 105261. doi:https://doi.org/10.1016/j.bspc.2023.105261
- Chamseddine, E., Mansouri, N., Soui, M., & Abed, M. (2022). Handling class imbalance in COVID-19 chest X-ray images classification: Using SMOTE and weighted loss. Applied Soft Computing, 129, 109588. doi:https://doi.org/10.1016/j.asoc.2022.109588
- Chawla, N. V., Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2002). SMOTE: synthetic minority over-sampling technique. Journal of artificial intelligence research, 16, 321-357. doi:https://doi.org/10.1613/jair.953
- Chen, Y., Chang, R., & Guo, J. (2021). Effects of data augmentation method borderline-SMOTE on emotion recognition of EEG signals based on convolutional neural network. IEEE Access, 9, 47491-47502. doi:https://doi.org/10.1109/access.2021.3068316
- Gonzalez-Huitron, V., León-Borges, J. A., Rodriguez-Mata, A., Amabilis-Sosa, L. E., Ramírez-Pereda, B., & Rodriguez, H. (2021). Disease detection in tomato leaves via CNN with lightweight architectures implemented in Raspberry Pi 4. Computers and Electronics in Agriculture, 181, 105951. doi:https://doi.org/10.1016/j.compag.2020.105951
- Goodfellow, I., Bengio, Y., Courville, A., & Bengio, Y. (2016). Deep learning. (Vol. 1): MIT press Cambridge. doi: https://doi.org/10.1007/s10710-017-9314-z.
- Guo, G., Wang, H., Bell, D., Bi, Y., & Greer, K. (2003). KNN model-based approach in classification. Paper presented at the OTM Confederated International Conferences" On the Move to Meaningful Internet Systems". doi: https://doi.org/10.1007/978-3-540-39964-3_62
- Kumar, S., & Panda, K. (2023). SDIF-CNN: Stacking deep image features using fine-tuned convolution neural network models for real-world malware detection and classification. Applied Soft Computing, 146, 110676. doi:https://doi.org/10.1016/j.asoc.2023.110676
- Nahiduzzaman, M., Goni, M. O. F., Hassan, R., Islam, M. R., Syfullah, M. K., Shahriar, S. M., . . . Kowalski, M. (2023). Parallel CNN-ELM: A multiclass classification of chest X-ray images to identify seventeen lung diseases including COVID-19. Expert Systems with Applications, 229, 120528. doi:https://doi.org/10.1016/j.eswa.2023.120528
- Onat, A., Şurdumavcı, G., Şenocak, M., Örnek, E., Gözükara, Y., Karaaslan, Y., . . . Özcan, R. (1991). Türkiye'de erişkinlerde kalp hastalığı ve risk faktörleri sıklığı taraması: 3. Kalp hastalıkları prevalansı. Türk Kardiyoloji Derneği Arşivi, 19(1), 26-33.
- Özcan, H. (2014). Çok düşük çözünürlüklü yüz imgelerinde derin öğrenme uygulamaları. Deniz Harp Okulu/Deniz Bilimleri ve Mühendisliği Enstitüsü, İstanbul, Erişim adresi: https://tez.yok.gov.tr/UlusalTezMerkezi/tezDetay.jsp.
- Peng, H. (2015). Air quality prediction by machine learning methods. University of British Columbia. doi: https://doi.org/10.14288/1.0166787,
- Ruuska, S., Hämäläinen, W., Kajava, S., Mughal, M., Matilainen, P., & Mononen, J. (2018). Evaluation of the confusion matrix method in the validation of an automated system for measuring feeding behaviour of cattle. Behavioural processes, 148, 56-62. doi:https://doi.org/10.1016/j.beproc.2018.01.004
- Si, T., Bagchi, J., & Miranda, P. B. (2022). Artificial neural network training using metaheuristics for medical data classification: an experimental study. Expert Systems with Applications, 193, 116423. doi:https://doi.org/10.1016/j.eswa.2021.116423
- Sun, P., Wang, Z., Jia, L., & Xu, Z. (2024). SMOTE-kTLNN: A hybrid re-sampling method based on SMOTE and a two-layer nearest neighbor classifier. Expert Systems with Applications, 238, 121848. doi:https://doi.org/10.1016/j.eswa.2023.121848
- Tuncer, S. A., Ayyıldız, H., Kalaycı, M., & Tuncer, T. (2021). Scat-NET: COVID-19 diagnosis with a CNN model using scattergram images. Computers in Biology and Medicine, 135, 104579. doi:https://doi.org/10.1016/j.compbiomed.2021.104579
- Ullah, H., Bin Heyat, M. B., Akhtar, F., Sumbul, Muaad, A. Y., Islam, M. S., . . . Lin, Y. (2022). An End‐to‐End Cardiac Arrhythmia Recognition Method with an Effective DenseNet Model on Imbalanced Datasets Using ECG Signal. Computational Intelligence and Neuroscience, 2022(1), 9475162. doi:https://doi.org/10.1155/2022/9475162
- Wu, Y.-c., & Feng, J.-w. (2018). Development and application of artificial neural network. Wireless Personal Communications, 102(2), 1645-1656. doi:https://doi.org/10.1007/s11277-017-5224-x
- Zeraatkar, S., & Afsari, F. (2021). Interval–valued fuzzy and intuitionistic fuzzy–KNN for imbalanced data classification. Expert Systems with Applications, 184, 115510. doi:https://doi.org/10.1016/j.eswa.2021.115510 Zhang, P., Zhang, X., & Liu, A. (2022). Effects of Data Augmentation with the BNNSMOTE Algorithm in Seizure Detection Using 1D‐MobileNet. Journal of Healthcare Engineering, 2022(1), 4114178. doi:https://doi.org/10.1155/2022/4114178
- Zhang, S., Cheng, D., Deng, Z., Zong, M., & Deng, X. (2018). A novel kNN algorithm with data-driven k parameter computation. Pattern Recognition Letters, 109, 44-54. doi:https://doi.org/10.1016/j.patrec.2017.09.036
Ayrıntılar
| Birincil Dil | İngilizce |
|---|---|
| Konular | Derin Öğrenme, Yapay Zeka (Diğer) |
| Bölüm | Araştırma Makalesi |
| Yazarlar | |
| Gönderilme Tarihi | 30 Aralık 2025 |
| Kabul Tarihi | 10 Şubat 2026 |
| Yayımlanma Tarihi | 3 Mart 2026 |
| IZ | https://izlik.org/JA95BD94UY |
| Yayımlandığı Sayı | Yıl 2026 Cilt: 29 Sayı: 1 |