Detection of cardiac arrhythmias using RNA-MLP with PSO

• Augusto P. Zatt UFRGS
• Juliano C. Machado IFSUL
• Alexandre Balbinot UFRGS

DOI: https://doi.org/10.5753/eramiars.2025.16719

Resumo

This work proposes and evaluates a cardiac arrhythmia classifier based on a Multilayer Perceptron (MLP) Artificial Neural Network (ANN) optimized by the Particle Swarm Optimization (PSO) algorithm. The model was trained and validated using a 10-fold cross-validation scheme, achieving remarkable performance metrics, with an average F1-score of 99.17%, precision of 99.2%, and sensitivity of 99.16%. The analysis of the learning curves, which show the training and validation loss converging in parallel, refutes the overfitting hypothesis. Although PSO is computationally more expensive than traditional gradient-based optimizers, this study demonstrates that this cost is a strategic investment. The meta-heuristic approach promotes robust convergence and the ability to escape local minima, resulting in a superior solution with a high generalization capability. The results confirm the effectiveness and robustness of the MLP-PSO combination for high-performance classification.

Referências

Acharya, U. R., Krishnan, S. M., Spaan, J. A., and Suri, J. S. (2007). Advances in cardiac signal processing. Springer.

Chen, J., Zhang, R., and Liu, Y. (2023). An adam-enhanced particle swarm optimizer for latent factor analysis.

Duda, R. O., Hart, P. E., et al. (2006). Pattern classification. John Wiley & Sons.

HRS (2017). Heart diseases and disorders. Technical report, Heart Rhythm Society.

Mansour, R., Amraoui, A., Nouaouri, I., García Díaz, V., Gupta, D., and Kumar, S. (2021). Artificial intelligence and internet of things enabled disease diagnosis model for smart healthcare systems. IEEE Access, PP:1–1.

Martis, R. J., Acharya, U. R., and Adeli, H. (2014). Current methods in electrocardiogram characterization. Computers in Biology and Medicine, 48:133–149.

Moody, G. and Mark, R. (1990). The mit-bih arrhythmia database on cd-rom and software for use with it. In [1990] Proceedings Computers in Cardiology, pages 185–188.

WHO (2021). Cardiovascular diseases (cvds). Technical report, World Health Organizaton.