Prediksi Detak Jantung Berbasis LSTM pada Raspberry Pi untuk Pemantauan Kesehatan Portabel

Penulis

  • Ahmad Foresta Azhar Zen Universitas Brawijaya, Malang
  • Eko Sakti Pramukantoro Universitas Brawijaya, Malang
  • Kasyful Amron Universitas Brawijaya, Malang
  • Viera Wardhani Universitas Brawijaya, Malang
  • Putri Annisa Kamila Universitas Brawijaya, Malang

DOI:

https://doi.org/10.25126/jtiik.1078015

Kata Kunci:

prediksi detak jantung, LSTM, raspberry pi, ECG, rr-interval

Abstrak

Penyakit kardiovaskular atau cardiovascular disease (CVD) menduduki peringkat teratas penyebab kematian di dunia. Diperkirakan sekitar 17,9 juta jiwa meninggal akibat CVD pada tahun 2019, yang menyumbang sebanyak 32% dari seluruh kematian global. Penting untuk mendeteksi kelainan pada jantung sedini mungkin untuk mencegah kematian karena CVD. Peningkatan kesadaran tentang pentingnya pemantauan kesehatan diri sendiri telah mendorong perkembangan teknologi pemantauan kesehatan portabel. Dalam penelitian ini, kami mengusulkan model prediksi detak jantung berbasis Long Short-Term Memory (LSTM) dengan menggunakan fitur RR-Interval dan mengimplementasikan pada perangkat Raspberry Pi. Model berbasis LSTM merupakan salah satu jenis arsitektur jaringan saraf tiruan yang mampu menangani data berurutan dengan baik, sehingga sangat cocok untuk pemantauan dan prediksi detak jantung yang bersifat sekuensial. Raspberry Pi dikenal karena ukurannya yang kecil, harga yang terjangkau, kinerja yang andal, dan efisiensi komputasi yang baik. Raspberry Pi juga memungkinkan integrasi yang mudah dengan berbagai sensor, menjadikannya solusi yang cocok untuk pemantauan kesehatan yang portabel. Hasil penelitian ini menunjukkan bahwa model klasifikasi yang diusulkan memiliki kinerja yang baik dengan tingkat akurasi mencapai 96,66%. Implementasi inferensi pada Raspberry Pi juga menunjukkan performa yang baik, dengan waktu 4,82 detik untuk melakukan inferensi data sepanjang 100 detik, serta penggunaan memori sebesar 134,8MB.

 

Abstract

Cardiovascular diseases (CVDs) rank as the top cause of global death. An estimated 17.9 million people succumbed to CVDs in 2019, constituting 32% of all global deaths. Detecting heart abnormalities as early as possible is crucial to prevent CVD-related fatalities. The growing awareness of the importance of self-health monitoring has driven the development of portable health monitoring technologies. In this study, we propose a Long Short-Term Memory (LSTM)-based heart beat prediction model using RR-Interval as features  and implement it on the Raspberry Pi device. LSTM models are a type of artificial neural network architecture known for their ability to handle sequential data effectively, making them highly suitable for sequential heart rate monitoring and prediction. The Raspberry Pi is renowned for its compact size, affordability, reliable performance, and efficient computational capabilities. It also enables seamless integration with various sensors, making it an ideal solution for portable health monitoring. This research show that the proposed classification model performs well, achieving an accuracy rate of 96.66%. The implementation of inference on the Raspberry Pi also demonstrates good performance, with an average inference time of 4.82 seconds for processing 100 data points and a memory usage of 134.8MB.

Downloads

Download data is not yet available.

Referensi

AHSANUZZAMAN, S.M., AHMED, T. dan RAHMAN, MD.A., 2020. Low Cost, Portable ECG Monitoring and Alarming System Based on Deep Learning. In: 2020 IEEE Region 10 Symposium (TENSYMP). 2020 IEEE Region 10 Symposium (TENSYMP). pp.316–319. https://doi.org/10.1109/TENSYMP50017.2020.9231005.

ASSOCIATION FOR THE ADVANCEMENT OF MEDICAL INSTRUMENTATION, 1998. Testing and reporting performance results of cardiac rhythm and ST segment measurement algorithms. ANSI/AAMI EC57.

CHAWLA, N.V., BOWYER, K.W., HALL, L.O. dan KEGELMEYER, W.P., 2002. SMOTE: synthetic minority over-sampling technique. Journal of artificial intelligence research, 16, pp.321–357.

ESSA, E. dan XIE, X., 2021. An Ensemble of Deep Learning-Based Multi-Model for ECG Heartbeats Arrhythmia Classification. IEEE Access, 9, pp.103452–103464. https://doi.org/10.1109/ACCESS.2021.3098986.

HE, H., BAI, Y., GARCIA, E.A. dan LI, S., 2008. ADASYN: Adaptive synthetic sampling approach for imbalanced learning. In: 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence). pp.1322–1328. https://doi.org/10.1109/IJCNN.2008.4633969.

HOCHREITER, S. dan SCHMIDHUBER, J., 1997. Long short-term memory. Neural computation, 9(8), pp.1735–1780.

KARIM, F., MAJUMDAR, S., DARABI, H. dan CHEN, S., 2018. LSTM Fully Convolutional Networks for Time Series Classification. IEEE Access, 6, pp.1662–1669. https://doi.org/10.1109/ACCESS.2017.2779939.

LENIS, G., PILIA, N., LOEWE, A., SCHULZE, W.H.W. dan DÖSSEL, O., 2017. Comparison of Baseline Wander Removal Techniques considering the Preservation of ST Changes in the Ischemic ECG: A Simulation Study. Computational and Mathematical Methods in Medicine, 2017, p.9295029. https://doi.org/10.1155/2017/9295029.

MOHAMMED, R., RAWASHDEH, J. dan ABDULLAH, M., 2020. Machine Learning with Oversampling and Undersampling Techniques: Overview Study and Experimental Results. In: 2020 11th International Conference on Information and Communication Systems (ICICS). pp.243–248. https://doi.org/10.1109/ICICS49469.2020.239556.

MONDÉJAR-GUERRA, V., NOVO, J., ROUCO, J., PENEDO, M.G. dan ORTEGA, M., 2019. Heartbeat classification fusing temporal and morphological information of ECGs via ensemble of classifiers. Biomedical Signal Processing and Control, 47, pp.41–48. https://doi.org/10.1016/j.bspc.2018.08.007.

MOODY, G.B. dan MARK, R.G., 2001. The impact of the MIT-BIH arrhythmia database. IEEE engineering in medicine and biology magazine, 20(3), pp.45–50.

PAN, J. dan TOMPKINS, W.J., 1985. A real-time QRS detection algorithm. IEEE transactions on bio-medical engineering, 32(3), pp.230–236. https://doi.org/10.1109/TBME.1985.325532.

PANDEY, S.K., JANGHEL, R.R. dan VANI, V., 2020. Patient Specific Machine Learning Models for ECG Signal Classification. Procedia Computer Science, 167, pp.2181–2190. https://doi.org/10.1016/j.procs.2020.03.269.

PRAMUKANTORO, E.S. dan GOFUKU, A., 2022. A Heartbeat Classifier for Continuous Prediction Using a Wearable Device. Sensors, 22(14), p.5080. https://doi.org/10.3390/s22145080.

SHCHETININ, E. dan GLUSHKOVA, A., 2022. Arrhythmia detection using resampling and deep learning methods on unbalanced data. Computer Optics, 46(6), pp.980–987. https://doi.org/10.18287/2412-6179-CO-1112.

WORLD HEALTH ORGANIZATION, 2021. Cardiovascular Diseases (CVDs). [online] Tersedia di: <https://www.who.int/en/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds)> [Diakses 21 September 2023].

YU, Y., SI, X., HU, C. dan ZHANG, J., 2019. A review of recurrent neural networks: LSTM cells and network architectures. Neural computation, 31(7), pp.1235–1270.

Diterbitkan

30-12-2023

Cara Mengutip

Prediksi Detak Jantung Berbasis LSTM pada Raspberry Pi untuk Pemantauan Kesehatan Portabel. (2023). Jurnal Teknologi Informasi Dan Ilmu Komputer, 10(7), 1555-1562. https://doi.org/10.25126/jtiik.1078015