Kombinasi Ekstraksi Ciri untuk  Klasifikasi Ventricular Fibrillation  menggunakan Support Vector Machine

Raden Danisworo Rivianto Wicaksono; Novie Theresia Pasaribu; Jo Suherman

doi:10.28932/jste.v1i2.14024

Authors

Raden Danisworo Rivianto Wicaksono Universitas Kristen Maranatha
Novie Theresia Pasaribu Universitas Kristen Maranatha
Jo Suherman Universitas Kristen Maranatha

DOI:

https://doi.org/10.28932/jste.v1i2.14024

Keywords:

Area Calculation, ECG, Spectral Analysis, Support Vector Machine, Ventricular Fibrillation

Abstract

Ventricular Fibrillation (VF) is a life-threatening heart rhythm disorder characterised by irregular and uncoordinated electrical activity of the heart that causes the heart to stop suddenly. An Electrocardiogram (ECG) is a medical test that detects heart abnormalities by measuring the electrical activity of the heart during contraction. The ECG in the VF shows very different characteristics from the normal heart rhythm, with loss of P waves and a regular QRS complex, replaced by rapid, irregular, and variable fibrillation waves of variable amplitude. A Support Vector Machine (SVM) is a type of Machine Learning that seeks the best hyperplane to separate classes. The kernel used in this study is best obtained by using the Quadratic Kernel. This study aims to detect Ventricular Fibrillation (VF) or Non-VF from ECG signals using Support Vector Machine (SVM). Preprocessing in this study: window size of ECG signals (5 seconds and 10 seconds), followed by a High Pass Filter, a Second Order Butterworth Low Pass Filter, and a Notch Filter. The characteristics used for extraction are Area Calculation (in this study, proposes using Ratio Area) and Spectral Analysis (FSMN, A1, A2, A3). Combinations of one to five of these trait extracts were trained and tested using SVM. The results obtained showed a combination of three characteristic extractions: FSMN-A1-A2 achieved the highest performance with 97% accuracy, 100% sensitivity, 94% specificity and the FSMN-A2-R characteristic extraction combination. The Area achieves 97% accuracy, 98% sensitivity, and 96% specificity. Adding trait extraction from three to four did not significantly improve performance.

References

A. Mjahad and A. Rosado-Muñoz, “Hybrid CNN-Fuzzy Approach for Automatic Identification of Ventricular Fibrillation and Tachycardia,” Applied Sciences, vol. 15, no. 17, p. 9289, Aug. 2025, doi: 10.3390/app15179289.

G. Salama and B.-R. Choi, “Imaging Ventricular Fibrillation,” Journal of Electrocardiology, vol. 40, no. 6, pp. S56–S61, Nov. 2007, doi: 10.1016/j.jelectrocard.2007.06.021.

M. F. Pérez-Gutiérrez et al., “Spectral Analysis and Mutual Information Estimation of Left and Right Intracardiac Electrograms during Ventricular Fibrillation,” Sensors, vol. 20, no. 15, p. 4162, Jul. 2020, doi: 10.3390/s20154162.

M. J. Reed, G. R. Clegg, and C. E. Robertson, “Analysing the Ventricular Fibrillation waveform,” Resuscitation, vol. 57, no. 1, pp. 11–20, Apr. 2003, doi: 10.1016/S0300-9572(02)00441-0.

A. Amann, R. Tratnig, and K. Unterkofler, “Reliability of old and new Ventricular Fibrillation detection algorithms for Automated External Defibrillators,” BioMed Eng OnLine, vol. 4, no. 1, p. 60, Dec. 2005, doi: 10.1186/1475-925X-4-60.

Qiao Li, C. Rajagopalan, and G. D. Clifford, “Ventricular Fibrillation and Tachycardia Classification Using a Machine Learning Approach,” IEEE Trans. Biomed. Eng., vol. 61, no. 6, pp. 1607–1613, Jun. 2014, doi: 10.1109/TBME.2013.2275000.

F. Alonso-Atienza, J. L. Rojo-Álvarez, A. Rosado-Muñoz, J. J. Vinagre, A. García-Alberola, and G. Camps-Valls, “Feature selection using Support Vector Machines and bootstrap methods for Ventricular Fibrillation detection,” Expert Systems with Applications, vol. 39, no. 2, pp. 1956–1967, Feb. 2012, doi: 10.1016/j.eswa.2011.08.051.

A. Picon et al., “Mixed convolutional and long short-term memory network for the detection of lethal ventricular arrhythmia,” PLoS ONE, vol. 14, no. 5, p. e0216756, May 2019, doi: 10.1371/journal.pone.0216756.

U. R. Acharya et al., “Automated identification of shockable and non-shockable life-threatening ventricular arrhythmias using convolutional neural network,” Future GeneRation Computer Systems, vol. 79, pp. 952–959, Feb. 2018, doi: 10.1016/j.future.2017.08.039.

Md. A. Awal, S. S. Mostafa, M. Ahmad, and M. A. Rashid, “An adaptive level dependent wavelet thresholding for ECG denoising,” Biocybernetics and Biomedical Engineering, vol. 34, no. 4, pp. 238–249, 2014, doi: 10.1016/j.bbe.2014.03.002.

G. B. Moody and R. G. Mark, “The impact of the MIT-BIH Arrhythmia Database,” IEEE Eng. Med. Biol. Mag., vol. 20, no. 3, pp. 45–50, Jun. 2001, doi: 10.1109/51.932724.

“The Only EKG Book You’ll Ever Need, 9e -- Malcolm S_ Thaler, M_D.”

Z. F. Issa, J. M. Miller, and D. P. Zipes, Clinical arrhythmology and electrophysiology: a companion to Braunwald’s heart disease, Third edition. Philadelphia, PA: Elsevier, 2019.

R. H. Clayton, A. Murray, and R. W. F. Campbell, “Recognition of Ventricular Fibrillation using neural networks,” Med. Biol. Eng. Comput., vol. 32, no. 2, pp. 217–220, Mar. 1994, doi: 10.1007/BF02518922.

Xu-Sheng Zhang, Yi-Sheng Zhu, N. V. Thakor, and Zhi-Zhong Wang, “Detecting ventricular tachycardia and fibrillation by complexity measure,” IEEE Trans. Biomed. Eng., vol. 46, no. 5, pp. 548–555, May 1999, doi: 10.1109/10.759055.

S. Barro, R. Ruiz, D. Cabello, and J. Mira, “Algorithmic sequential decision-making in the frequency domain for life threatening ventricular arrhythmias and imitative artefacts: a diagnostic system,” Journal of Biomedical Engineering, vol. 11, no. 4, pp. 320–328, Jul. 1989, doi: 10.1016/0141-5425(89)90067-8.

C. Cortes and V. Vapnik, “Support-vector networks,” Mach Learn, vol. 20, no. 3, pp. 273–297, Sep. 1995, doi: 10.1007/BF00994018.