Using Recurrent Neural Networks to Predict Ischemic Heart Diseases from Time Series

• Davi Matos de Carvalho IFMA
• Omar Andres Carmona Cortes IFMA

DOI: https://doi.org/10.5753/sbcas_estendido.2025.7338

Abstract

The present investigation aims to evaluate the efficiency of using recurrent neural networks to analyze a time series of deaths due to ischemic heart disease in the city of São Luís, Maranhão. It employs two models, LSTM, BiLSTM, and GRU, seeking to identify the combination of parameters that yields the best results. These RNNs are compared with the traditional ARIMA model in an experiment using k-fold. Results indicate that LSTM achieves an RMSE of 0.70, the BiLSTM of 0.45, and the GRU of 0.46, while ARIMA reaches a higher RMSE of 7.5.

References

BARBOZA, M. F. X. (2021). Análise e predição de incidência de casos de malária no tempo e no espaço utilizando modelos deep learning resumo estendido. CTDSI/CTCCSI.

CARVALHO, D. L. d. (2023). Aplicação de Redes Neurais Recorrentes para Previsão de Séries Temporais. Tese (doutorado), Universidade Estadual da Paraíba, Campina Grande, PB.

GOODFELLOW, I., BENGIO, Y., and COURVILLE, A. (2016). Deep Learning. MIT Press.

HYNDMAN, R. J. and ATHANASOPOULOS, G. (2018). Forecasting: Principles and Practice. OTexts.

SHAHID, F., KHAN, M., AHMED, M., ALI, M., REHMAN, M., and KHAN, M. (2020). Predictions for covid-19 with deep learning models of lstm, gru and bi-lstm. Chaos, Solitons and Fractals, 140:111178.

Siami-Namini, S., Tavakoli, N., and Namin, A. S. (2019). The performance of lstm and bilstm in forecasting time series. In 2019 IEEE International Conference on Big Data (Big Data), pages 3285–3292.

Soyiri, I. N. and Reidpath, D. D. (2013). An overview of health forecasting. Environmental Health and Preventive Medicine, 18(1):1–9.

World Health Organization (2024). The top 10 causes of death. Acesso em: 18 fev. 2025.