Klasifikasi Aritmia Pada Sinyal Elektrokardiogram Menggunakan Long-Short Term Memory Dan Variannya
Abstract
Penyakit jantung merupakan penyebab kematian tertinggi di dunia, termasuk Aritmia. Pengecekan Aritmia dilakukan menggunakan alat Elektrokardiogram (EKG), analisis dilakuk- an oleh para profesional medis. Namun, sering kali membu- tuhkan waktu subjektif dan rentan terhadap kesalahan. Peneli- tian ini mengusulkan sistem klasifikasi Aritmia menggunakan Long-Short Term Memory (LSTM) dan Variannya. Klasifikasi dilakukan terhadap delapan jenis Aritmia menggunakan data- set MIT-BIH Arrhythmia. Peneltian ini menggunakan teknik sliding window berdasarkan jumlah (PQRST) dalam setiap episode. Hasilnya digunakan sebagai masukan untuk klasifi- kasi Aritmia menggunakan arsitektur LSTM, BI-LSTM dan NLSTM dengan kombinasi optimizer (Adam, RMSprop, SGD) dan batch size (16,32, 64). Melalui penelitian ini ditemukan bahwa hasil klasifikasi Aritmia dengan kombinasi arsitektur LSTM dengan ukuran window 10 kompleks PQRST, optimizer RMSprop dan batch size 32 memberikan performa terbaik dibandingkan kombinasi lainnya. Hasil yang diperoleh adalah akurasi accuracy 96.51%, precision 96.77%, recall 96.51% dan F1-score 96.56%.
Kata kunci: Aritmia, BILSTM, LSTSM, NLSTM, Klasifikasi, Ritme jantung, RNN, Sinyal ECG.
References
A. Ali, E. Ifadah, and N. Hidayah, Keperawatan Gawat Darurat: Teori dan Implementasi. Jakarta: PT. Softpedia Publishing Indonesia, 2024.
World Health Organization, “Cardiovascular diseases (CVDs),” Jun. 11, 2021. [Online]. Available: https://www.who.int/news-room/fact- sheets/detail/cardiovascular-diseases-(cvds)
A. Sharma, N. Garg, S. Patidar, R. S. Tan, and U. R. Acharya, “Automated pre-screening of arrhythmia using hybrid combination of Fourier-Bessel expansion and LSTM,” Computers in Biology and Medicine, vol. 120, p. 103753, Apr. 2020, doi: ht-
S. of C. S. A. Engineering, “ECGTransForm: empowering adap- tive ECG arrhythmia classification framework with bidirectio- nal transformer,” NTU Singapore, 2024. [Online]. Available: ht- tps://dr.ntu.edu.sg/handle/10356/171854
A. H. Kashou et al., “ECG Interpretation Proficiency of healthcare professionals,” Current Problems in Cardi- ology, vol. 48, no. 10, p. 101924, Jul. 2023, doi: ht-
tps://doi.org/10.1016/j.cpcardiol.2023.10192410.1016/j.cpcardiol.2023.101924.
I. Ismail, D. Purnamawati, W. Jumaiyah, and F. Rayasari, “Pe- ningkatan Kemampuan Perawat dalam Interpretasi EKG Normal dan Aritmia dengan Metode Angka ‘3,’” Jurnal Keperawatan Silampari, vol. 4, no. 2, pp. 405–414, Mar. 2021, doi: ht- tps://doi.org/10.31539/jks.v4i2.192410.31539/jks.v4i2.1924.
D. O. Dantas, “Cardiac Arrhythmia Detection in Electrocardiogram Signals with CNN-LSTM,” UFS-BR, Sep. 2024. [Online].
Available:https://www.academia.edu/116202197/CardiacArrythmiaDetectioninElectrocardiogramSignalswithC NNLSTM
“LSTM-Based Auto-Encoder Model for ECG Arrhythmias Classifica- tion,” IEEE Journals & Magazine, Apr. 1, 2020. [Online]. Available: https://ieeexplore.iee.org/document/8688435
“MIT-BIH Arrhythmia Database V1.0.0,” Feb. 24, 2005. [Online].
Available: https://www.physionet.org/content/mitdb/1.0.0/
“View of noise suppression of ECG signal using optimized digital Butterworth Bandpass filter.” ht- tp://ijcs.net/ijcs/index.php/ijcs/article/view/4312/684
A. Nahar, “Adaptive Symlet filter based on ECG baseline wander removal, “Serbian Journal of Electrical Engineering, vol. 17, no. 2, pp. 187-197, Jan. 2020, doi: 10.2298/sjee2002188n.
I. Sadiq, A. M. Zuberi, I. Zaman, A. Hassan, and T. Zaidi, “Adaptive removal of power-line interference from high resolution ECG,” Advan- ces in Bioscience and Biotechnology, vol. 03, no. 04, pp. 324-326, Jan. 2012, doi:10.4236/abb.2012.34047.
“filtfilt — SciPy v1.15.3 Manual.” ht- tps://docs.scipy.org/doc/scipy/reference/generated/scipy.signal.filtfilt.html
Mandala, S., Rizal, A., Adiwijaya, N., Nurmaini, S., Amini, S. S., Sudarisman, G. A., Hau, Y. W., Abdullah, A. H. (2024b). An improved method to detect arrhythmia using ensemble learning-based model in multi lead electrocardiogram (ECG). PLoS ONE, 19(4), e0297551. https://doi.org/10.1371/journal.pone.0297551
“qrs.py - Waveform Database Software Package (WF- DB) for Python.” https://www.physionet.org/content/wfdb python/3.3.0/wfdb/processing/qrs.py
“ECG — NeuroKit2 0.2.12 documentation.” ht- tps://neuropsychology.github.io/NeuroKit/functions/ecg.html
“View of Sun Position Forecasting menggunakan metode RNN – LSTM sebagai referensi pengendalian daya solar cell.” [Online]. Available: https://journal.fortei7.org/index.php/sinarFe7/article/view/307/275
A. Sherstinsky, “Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) network,” Physica D: Nonlinear Phenomena, [n.d.].
L. D. Sharma, J. Rahul, A. Aggarwal, and V. K. Bohat, “An improved cardiac arrhythmia classification using stationary wavelet transform decomposed short duration QRS segment and Bi-LSTM network,” Multidimensional Systems and Signal Processing, vol. 34, no. 2, pp. 503–520, Mar. 2023, doi: https://doi.org/10.1007/s11045-023-00875- x10.1007/s11045-023-00875-x.
S. M. Hashemi, R. M. Botez, and G. Ghazi, “Bidirectional Long Short-Term Memory Development for Aircraft Traje- ctory Prediction Applications to the UAS-S4 EHE´ CATL,” Aerospace, vol. 11, no. 8, p. 625, Jul. 2024, doi: ht- tps://doi.org/10.3390/aerospace1108062510.3390/aerospace11080625.
M. Yang et al., “Design and implementation of an explainable bidirectio- nal LSTM model based on transition system approach for cooperative AI-Workers,” Applied Sciences, vol. 12, no. 13, p. 6390, 2022. doi: https://doi.org/10.3390/app1213639010.3390/app12136390.
K. Wang, K. Zhang, B. Liu, W. Chen, and M. Ham, “Early prediction of sudden cardiac death risk with Nested LSTM based on electroca- rdiogram sequential feature,” BMC Medical Informatics and Decision Making, vol. 24, no. 1, Apr. 2024, doi: https://doi.org/10.1186/s12911- 024-02493-410.1186/s12911-024-02493-4.tps://doi.org/10.1016/j.compbiomed.2020.10375310.1016/j.compbiomed.2020.103753.
