ResNet and BiLSTM Networks for Myocardial Infarction Detection
International Journal of Scientific Research in Computer Science, Engineering and Information Technology · 20.11.2025 · 11 (6) · 722-728
DOI: 10.32628/cseit26123378 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Scientific Research in Computer Science, Engineering and Information Technology, online erschienen am 20.11.2025. Verfasst von Rashedur Rahman und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
This study evaluates two deep learning models, ResNet1D and CNN-BiLSTM-Attention, for automated myocardial infarction detection using 12-lead ECG signals from the PTB-XL dataset. Both models were trained using class-weighted learning and data augmentation. ResNet1D achieved 96.50% accuracy, 92.88% precision, 92.63% recall, and 92.75% F1-score, while CNN-BiLSTM-Attention achieved 95.40% accuracy, 88.85% precision, 93.77% recall, and 91.08% F1-score. The results demonstrate that deep learning can accurately identify myocardial infarction from ECG recordings, with ResNet1D providing the best overall performance and CNN-BiLSTM-Attention offering a compact and effective alternative.
Zitieren
APA
Rahman, R., & Naim, S. A. S. M. (2025). ResNet and BiLSTM Networks for Myocardial Infarction Detection. International Journal of Scientific Research in Computer Science, Engineering and Information Technology, 11(6), 722-728. https://doi.org/10.32628/cseit26123378
Vancouver
Rahman R, Naim SASM. ResNet and BiLSTM Networks for Myocardial Infarction Detection. International Journal of Scientific Research in Computer Science, Engineering and Information Technology. 2025;11(6):722-728. doi:10.32628/cseit26123378
BibTeX
@article{Rahman2025,
author = {Rahman, Rashedur and Naim, S A Sabbirul Mohosin},
title = {ResNet and BiLSTM Networks for Myocardial Infarction Detection},
journal = {International Journal of Scientific Research in Computer Science, Engineering and Information Technology},
year = {2025},
volume = {11},
number = {6},
pages = {722--728},
doi = {10.32628/cseit26123378}
}
EUVIMED