Application of Deep Learning Networks Recurrent Neural Network, Long Short-Term Memory, and Gated Recurrent Unit in Predicting COVID-19 in the Post-Vaccination Scenario

• Rafaella Silva Ferreira UNESP
• Wallace Casaca UNESP
• Marilaine Colnago UNESP

DOI: https://doi.org/10.5753/semish.2024.2562

Abstract

This work proposes the adaptation of three artificial neural networks to predict COVID-19 time series in Brazil, considering the current scenario of data scarcity and recurrent waves of the disease, though of lesser magnitude compared to 2020 and 2021. The main goal is to assess the performance of different neural network architectures in forecasting daily COVID-19 cases in the state of São Paulo. The evaluated architectures are as follows: Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU). The performance of each model was analyzed in terms of adherence to real data and the ability to capture complex temporal patterns, in a context of sudden increases and declines in cases of lesser severity due to vaccination. The results demonstrated high accuracy of the trained networks and provided insights to improve the quality of the predictions, which are essential for pandemic control strategies, especially during periods of disease resurgence. Thus, this work aims to contribute to the advancement of predictive neural network applications for COVID-19, particularly in the post-vaccination context.

References

Amaral, F., Casaca, W., Oishi, C. M., and Cuminato, J. A. (2021a). Simulating immunization campaigns and vaccine protection against COVID-19 pandemic in Brazil. IEEE Access, 9:126011–126022.

Amaral, F., Casaca, W., Oishi, C. M., and Cuminato, J. A. (2021b). Towards providing effective data-driven responses to predict the covid-19 in São Paulo and Brazil. Sensors, 21(2):540.

Capomaccio, S. (2024). Casos de covid-19 registram aumento e vacinação ainda é importante. Jornal da USP.

Chandra, R., Jain, A., and Singh Chauhan, D. (2022). Deep learning via lstm models for covid-19 infection forecasting in india. PloS one, 17(1):e0262708.

Cho, K., van Merriënboer, B., Bahdanau, D., and Bengio, Y. (2014). On the properties of neural machine translation: Encoder–decoder approaches. In Wu, D., Carpuat, M., Carreras, X., and Vecchi, E. M., editors, Proceedings of SSST-8, Eighth Workshop on Syntax, Semantics and Structure in Statistical Translation, pages 103–111, Doha, Qatar. Association for Computational Linguistics.

Chung, J., Gulcehre, C., Cho, K., and Bengio, Y. (2014). Empirical evaluation of gated recurrent neural networks on sequence modeling. arXiv preprint arXiv:1412.3555.

Davahli, M., Fiok, K., Karwowski, W., Aljuaid, A., and Taiar, R. (2021). Predicting the dynamics of the covid-19 in the united states using graph theory-based neural networks. International Journal of Environmental Research and Public Health, 18(7).

Frazier, P. I. (2018). A tutorial on bayesian optimization. arXiv preprint arXiv:1807.02811.

Gomez-Cravioto, D. A., Diaz-Ramos, R. E., Cantu-Ortiz, F. J., and Ceballos, H. G. (2021). Data analysis and forecasting of the covid-19 spread: A comparison of recurrent neural networks and time series models. Cognitive Computation, pages 1–12.

Hawas, M. (2020). Generated time-series prediction data of covid-19’s daily infections in brazil by using recurrent neural networks. Data in Brief, 32:106175.

Hewamalage, H., Bergmeir, C., and Bandara, K. (2021). Recurrent neural networks for time series forecasting: Current status and future directions. International Journal of Forecasting, 37(1):388–427.

Linge, S. and Langtangen, H. P. (2020). Programming for Computations-Python: A Gentle Introduction to Numerical Simulations with Python 3.6. Springer Nature.

Ma, J., Zhao, H., and Schuck, P. (2015). A histogram approach to the quality of fit in sedimentation velocity analyses. Analytical Biochemistry, 483:1–3.

Mendes, V. (2024). Por que covid-19 ainda mata tanta gente no brasil. BBC News Brasil.

Microsoft (2023). Microsoft. [link]. Acesso em 20 de julho de 2023.

Ministério da Saúde (2024). Relatório da situação de saúde diante da pandemia de covid19. Disponível em: [link]. Acesso em: 14 mar. 2024.

Oliveira, C. (2024). Controle sobre a covid-19 ainda é instável no país após quatro anos do 1º caso. Brasil de Fato.

Roosa, K., Lee, Y., Luo, R., Kirpich, A., Rothenberg, R., Hyman, J. M., Yan, P., and Chowell, G. (2020). Short-term forecasts of the COVID-19 epidemic in Guangdong and Zhejiang, China: February 13–23, 2020. Journal of Clinical Medicine, 9:596.

Shi, X., Chen, Z., Wang, H., Yeung, D.-Y., Wong, W.-K., and Woo, W.-c. (2015). Convolutional lstm network: A machine learning approach for precipitation nowcasting. Advances in neural information processing systems, 28.

Snoek, J., Larochelle, H., and Adams, R. P. (2012). Practical bayesian optimization of machine learning algorithms. In Advances in Neural Information Processing Systems, volume 25. Curran Associates, Inc.

Snoek, J., Rippel, O., Swersky, K., Kiros, R., Satish, N., Sundaram, N., Patwary, M., Prabhat, M., and Adams, R. (2015). Scalable bayesian optimization using deep neural networks. In International conference on machine learning, pages 2171–2180. PMLR.

Worldometer (2024). Covid-19 Coronavirus pandemic. Disponível em: [link]. Acesso em: 14 mar. 2024.

Worobey, M. (2021). Dissecting the early COVID-19 cases in Wuhan. Science, 374(6572):1202–1204.

Zeroual, A., Harrou, F., Dairi, A., and Sun, Y. (2020). Deep learning methods for forecasting covid-19 time-series data: A comparative study. Chaos, Solitons & Fractals, 140:110121.