Research Article PENDETEKSIAN PERUBAHAN SINYAL DINAMIS ECG PADA ARITMIA JANTUNG MENGGUNAKAN ANALISIS RÉNYI ENTROPI DAN XGBOOST
Keywords:
ECG, Aritmia, Rényi Entropy, Discrete Wavelet Transform, XGBoost, Signal-to-Noise RatioAbstract
Deteksi aritmia jantung secara akurat dan cepat sangat penting untuk mencegah komplikasi kardiovaskular yang fatal. Penelitian ini mengusulkan metode deteksi berbasis kombinasi Discrete Wavelet Transform (DWT) multi-level, analisis Rényi Entropy, dan algoritma XGBoost. Dataset yang digunakan adalah MIT-BIH Arrhythmia Database dengan dua kanal utama, MLII dan V1. Tahapan penelitian mencakup filtering sinyal untuk peningkatan Signal-to-Noise Ratio (SNR), segmentasi beat presisi berbasis deteksi R-peak, pelabelan menggunakan anotasi MIT-BIH (.atr), serta ekstraksi fitur Rényi Entropy pada level DWT 1 hingga 4. Untuk mengatasi ketidakseimbangan kelas antara beat normal dan aritmia, dilakukan undersampling moderat dengan rasio 2:1 pada data latih. Hasil pengujian menunjukkan bahwa model XGBoost dengan balancing data mencapai akurasi 88,33% dan ROC AUC 0,9410, sedangkan tanpa balancing menghasilkan akurasi 88,61% dan ROC AUC 0,9365. Analisis SHAP mengungkap bahwa fitur entropy pada level DWT rendah memiliki kontribusi signifikan dalam klasifikasi. Peningkatan nilai SNR pasca-filtering menunjukkan keberhasilan tahap preprocessing dalam mengurangi noise dan mempertahankan morfologi PQRST. Metode yang diusulkan memberikan kinerja tinggi, sensitivitas yang baik terhadap kelas minoritas, dan potensi implementasi pada sistem pemantauan ECG waktu nyata.
Kata kunci: ECG, Aritmia, Rényi Entropy, Discrete Wavelet Transform, XGBoost, Signal-to-Noise Ratio
References
World Health Organization, "Cardiovascular diseases (CVDs) fact sheet," WHO, 2023. [Online]. Available: https://www.who.int/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds)
Y. Ansari, O. Mourad, K. Qaraqe, and E. Serpedin, "Deep learning for ECG arrhythmia detection and classification: An overview of progress for 2017–2023," Frontiers in Physiology, vol. 14, 2023, doi: 10.3389/fphys.2023.1246746.
A. Antzelevitch and M. Burashnikov, "Overview of basic mechanisms of arrhythmias," Circulation Research, vol. 109, no. 5, pp. 648–666, 2011.
D. P. Zipes and J. Jalife, Cardiac Electrophysiology: From Cell to Bedside, 6th ed. Philadelphia, PA, USA: Elsevier, 2014.
D. S. Levine, J. H. Rosenthal, and S. S. Podrid, "Cardiac arrhythmias: Diagnosis and treatment," in Heart Disease and Arrhythmias, 4th ed., pp. 455–470, 2019.
D. P. Goldberger, Clinical Electrocardiography: A Simplified Approach, 8th ed. Philadelphia, PA, USA: Elsevier, 2018.
G. D. Clifford, F. Azuaje, and P. E. McSharry, Advanced Methods and Tools for ECG Data Analysis. Norwood, MA, USA: Artech House, 2006.
G. B. Moody and R. G. Mark, "The impact of the MIT-BIH arrhythmia database," IEEE Engineering in Medicine and Biology Magazine, vol. 20, no. 3, pp. 45–50, 2001, doi: 10.1109/51.932724.
S. Mallat, A Wavelet Tour of Signal Processing: The Sparse Way, 3rd ed. Burlington, MA, USA: Academic Press, 2008.
P. Jizba and T. Arimitsu, "The world according to Rényi: Thermodynamics of multifractal systems," Annals of Physics, vol. 312, no. 1, pp. 17–59, Nov. 2004, doi: 10.1016/j.aop.2004.03.016.
M. Shi et al., "Rényi distribution entropy analysis of short-term heart rate variability signals and its application in coronary artery disease detection," Frontiers in Physiology, vol. 10, 2019, doi: 10.3389/fphys.2019.00809.
M. Ribeiro et al., "The entropy universe," Entropy, vol. 23, no. 2, p. 222, 2021, doi: 10.3390/e23020222.
Y. Sun, Y. Zhang, and M. Wang, "A novel deep entropy learning approach for ECG arrhythmia detection using hybrid wavelet-domain features," Computers in Biology and Medicine, vol. 157, p. 106760, 2023, doi: 10.1016/j.compbiomed.2023.106760.
S. Śmigiel, K. Pałczyński, and D. Ledziński, "ECG signal classification using deep learning techniques based on the PTB-XL dataset," Entropy, vol. 23, no. 9, p. 1121, 2021, doi: 10.3390/e23091121.
Y. Xin et al., "Paroxysmal atrial fibrillation recognition based on multi-scale Rényi entropy of ECG," Technology and Health Care, vol. 25, pp. 189–196, 2017, doi: 10.3233/THC-171321.
K. D. Shiva and V. V. R. Kondreddy, "Multiscale Rényi entropy analysis for classification of ECG signals," in Proc. Computing in Cardiology, vol. 50, Granada, Spain, 2023, pp. 1–4. [Online]. Available: https://www.cinc.org/archives/2023/pdf/CinC2023-447.pdf
A. Kumar, S. Pal, and R. Balasubramanian, "Hybrid entropy features and machine learning for ECG-based arrhythmia classification," IEEE Access, vol. 10, pp. 18102–18113, 2022, doi: 10.1109/ACCESS.2022.3148654.
M. Wang, C. Zhang, Q. Sun, J. Zhao, and Y. Zhang, "A multiscale Renyi entropy-based method for atrial fibrillation detection," Frontiers in Physiology, vol. 13, p. 846158, 2022. [Online]. Available: https://pubmed.ncbi.nlm.nih.gov/35492618/
M. Qodri, N. A. Rahman, and D. C. E. Wirawan, "Optimasi klasifikasi aritmia jantung berbasis ECG menggunakan hybrid entropy dan CNN," in Proc. Applied Computer and Engineering (ACE), 2023. [Online]. Available: https://www.ewadirect.com/proceedings/ACE/article/view/15901
I. A. Fadhlurrahman, A. Rachmatillah, and F. A. Nugroho, "Pendeteksian aritmia jantung berdasarkan citra sinyal ECG dengan metode convolutional neural network," J. Teknol. Inform., vol. 7, no. 1, pp. 1–10, 2023. [Online]. Available: https://e-journals.dinamika.ac.id/joti/article/view/792
Y.-H. Lin, H.-Y. Hsieh, and C.-Y. Lin, "QRS complex detection using wavelet transform and neural networks," IEEE Access, vol. 7, pp. 60441–60449, 2019, doi: 10.1109/ACCESS.2019.2914782.
L. Sharma and S. Kaushik, "ECG arrhythmia classification using ensemble learning with entropy-based features," Biomedical Signal Processing and Control, vol. 83, p. 104577, 2023, doi: 10.1016/j.bspc.2023.104577.
S. M. Lundberg and S.-I. Lee, "A unified approach to interpreting model predictions," Advances in Neural Information Processing Systems, vol. 30, pp. 4765–4774, 2017. [Online]. Available: https://arxiv.org/abs/1705.07874
T. Chen and C. Guestrin, "XGBoost: A scalable tree boosting system," in Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, 2016, pp. 785–794, doi: 10.1145/2939672.2939785.
M. Arif, M. A. Ghayour, and A. El-Rafei, "Comparative analysis of wavelet families for ECG feature extraction and classification," Journal of Medical Systems, vol. 46, no. 7, pp. 1–15, 2022, doi: 10.1007/s10916-022-01822-0.
A. Bouaziz, M. Deriche, and A. Kardouchi, "ECG signal denoising and features extraction using discrete wavelet transform (DWT)," Int. J. Comput. Appl., vol. 96, no. 20, pp. 36–42, June 2014, doi: 10.5120/16720-6401.
T. Phukpattaranont, "Comparison of wavelet functions for QRS detection in ECG signal," BioMedical Engineering OnLine, vol. 14, no. 1, p. 59, 2015, doi: 10.1186/s12938-015-0048-8.
K. A. Basha and S. S. Kumar, "ECG signal analysis using wavelet transform," Int. J. Eng. Trends Technol. (IJETT), vol. 17, no. 4, pp. 168–170, Nov. 2014. [Online]. Available: https://ijettjournal.org/volume-17/number-4/IJETT-V17P234.pdf
M. U. Zahid et al., "Robust R-peak detection in low-quality Holter ECGs using 1D convolutional neural network," arXiv preprint, arXiv:2101.01666, Jan. 2021. [Online]. Available: https://arxiv.org/abs/2101.01666
S. M. Lundberg and S.-I. Lee, "A unified approach to interpreting model predictions," Advances in Neural Information Processing Systems, vol. 30, pp. 4765–4774, 2017. [Online]. Available: https://arxiv.org/abs/1705.07874
Downloads
Published
Issue
Section
License
Authors who publish with this journal agree to the following terms:
(1) Copyright of the published articles will be transferred to the journal as the publisher of the manuscripts. Therefore, the author confirms that the copyright has been managed by the publisher.
(2) Publisher of TELKATTKA is Openlibrary Telkom University
(3) The copyright follows Creative Commons Attribution–ShareAlike License (CC BY SA): This license allows to Share — copy and redistribute the material in any medium or format, Adapt — remix, transform, and build upon the material, for any purpose, even commercially.
