EKSTRASI FITUR SINYAL EKG MYOCARDIAL INFARCTIN MENGGUNAKAN DISCRETE WAVELET TRANSFORMATION

Siti Agrippina Alodia Yusuf; Nani Sulistianingsih; Helmi Imaduddin

doi:10.46764/teknimedia.v4i1.96

Siti Agrippina Alodia Yusuf Universitas Muhammadiyah Mataram
Nani Sulistianingsih Universitas Muhammadiyah Mataram
Helmi Imaduddin Universitas Muhammadiyah Surakarta

DOI: https://doi.org/10.46764/teknimedia.v4i1.96

Keywords: DWT, ECG, KNN, K-Fold

Abstract

One important step in the process of identifying EKG signals is feature extraction, where the obtained features characterize the condition of the heart. The condition of the heart can be observed based on the waves produced in the EKG signal, which are generated by the electrical activity of the heart. In this study, two types of mother wavelets will be compared to determine which type is most suitable for extracting features from EKG signals. The types of mother wavelets to be compared are Daubechies and Symlet with orders of 5, 6, and 7 for Daubechies, and 6, 7, and 8 for Symlet. EKG signals with MI and normal heart conditions that have been improved in quality and have undergone signal segmentation are extracted using Discrete Wavelet Transformation (DWT) with Daubechies and Symlet mother wavelets at the two-level decomposition, and statistical features such as mean, median, standard deviation, kurtosis, and skewness are taken. Features are extracted from the D2 and D1 sub-bands, resulting in a total of 10 features obtained. The EKG signals are then classified using the KNN method, and to obtain generalized results, K-fold cross-validation is also applied. Based on the experiments conducted, the highest accuracy obtained was 94% with sensitivity and specificity of 82% and 91% by applying the Daubechies mother wavelet of order 7.

References

[1] “Cardiovascular diseases (CVDs).pdf.”
[2] World Health Organization, Noncommunicable diseases country profiles 2018. Geneva: World Health Organization, 2018. Accessed: Nov. 08, 2022. [Online]. Available: https://apps.who.int/iris/handle/10665/274512
[3] N. A. Khadse, A. M. Wankhade, and A. G. Gaiki, “Myocardial Infraction: Etiology, Risk Factors, Pathophysiology, Diagnosis and Management,” Am. J. PharmTech Res., vol. 10, no. 1, pp. 173–190, Feb. 2020, doi: 10.46624/ajptr.2020.v10.i1.014.
[4] M. S. Al-Ani, “ECG Waveform Classification Based on P-QRS-T Wave Recognition,” UHD J. Sci. Technol., vol. 2, no. 2, pp. 7–14, Jul. 2018, doi: 10.21928/uhdjst.v2n2y2018.pp7-14.
[5] P. Madona, R. I. Basti, and M. M. Zain, “PQRST wave detection on ECG signals,” Gac. Sanit., vol. 35, pp. S364–S369, 2021, doi: 10.1016/j.gaceta.2021.10.052.
[6] S. H. Jambukia, V. K. Dabhi, and H. B. Prajapati, “Classification of ECG signals using machine learning techniques: A survey,” in 2015 International Conference on Advances in Computer Engineering and Applications, Ghaziabad, India: IEEE, Mar. 2015, pp. 714–721. doi: 10.1109/ICACEA.2015.7164783.
[7] Y. Arpitha, G. L. Madhumathi, and N. Balaji, “Spectrogram analysis of ECG signal and classification efficiency using MFCC feature extraction technique,” J. Ambient Intell. Humaniz. Comput., vol. 13, no. 2, pp. 757–767, Feb. 2022, doi: 10.1007/s12652-021-02926-2.
[8] S. Mian Qaisar and S. Fawad Hussain, “Arrhythmia Diagnosis by Using Level-Crossing ECG Sampling and Sub-Bands Features Extraction for Mobile Healthcare,” Sensors, vol. 20, no. 8, p. 2252, Apr. 2020, doi: 10.3390/s20082252.
[9] S. A. Alodia Yusuf and R. Hidayat, “MFCC Feature Extraction and KNN Classification in ECG Signals,” in 2019 6th International Conference on Information Technology, Computer and Electrical Engineering (ICITACEE), Semarang, Indonesia: IEEE, Sep. 2019, pp. 1–5. doi: 10.1109/ICITACEE.2019.8904285.
[10] S. A. A. Yusuf and R. Hidayat, “Feature Extraction of ECG Signals using Discrete Wavelet Transform and MFCC,” in 2019 5th International Conference on Science in Information Technology (ICSITech), Yogyakarta, Indonesia: IEEE, Oct. 2019, pp. 167–170. doi: 10.1109/ICSITech46713.2019.8987544.
[11] Y. Toulni, N. Benayad, and B. D. Taoufiq, “Electrocardiogram signals classification using discrete wavelet transform and support vector machine classifier,” IAES Int. J. Artif. Intell. IJ-AI, vol. 10, no. 4, p. 960, Dec. 2021, doi: 10.11591/ijai.v10.i4.pp960-970.
[12] Y. Toulni, T. Belhoussine Drissi, and B. Nsiri, “ECG signal diagnosis using Discrete Wavelet Transform and K-Nearest Neighbor classifier.,” in The 4th International Conference on Networking, Information Systems amp Security., KENITRA AA Morocco: ACM, Apr. 2021, pp. 1–6. doi: 10.1145/3454127.3457628.
[13] A. Velayudhan and S. Peter, “Noise Analysis and Different Denoising Techniques of ECG Signal - A Survey,” IOSR J. Electron. Commun. Eng., p. 5, 2016.
[14] I. Fahruzi, “Mengurangi Pengaruh Noise Baseline Wander pada Sinyal Electrocardiogram(ECG),” vol. 5, no. 1, p. 5, Mar. 2013.
[15] J. H. Choi, H. K. Jung, and T. Kim, “A New Action Potential Detector Using the MTEO and Its Effects on Spike Sorting Systems at Low Signal-to-Noise Ratios,” IEEE Trans. Biomed. Eng., vol. 53, no. 4, pp. 738–746, Apr. 2006, doi: 10.1109/TBME.2006.870239.
[16] J. H. Choi and T. Kim, “Neural action potential detector using multi-resolution TEO,” Electron. Lett., vol. 38, no. 12, p. 541, 2002, doi: 10.1049/el:20020386.
[17] A. Almumri, E. Balakrishnan, and Narasimman, Sundararajan, “Discrete Wavelet Transform Based Feature Extraction in Electrocardiogram Signals,” vol. 17, no. 1, pp. 63–67, 2021.
[18] I. Saini, D. Singh, and A. Khosla, “QRS detection using K-Nearest Neighbor algorithm (KNN) and evaluation on standard ECG databases,” J. Adv. Res., vol. 4, no. 4, pp. 331–344, Jul. 2013, doi: 10.1016/j.jare.2012.05.007.