Forecasts of cardiac and respiratory mortality in Tehran, Iran, using ARIMAX and CNN-LSTM models.

Researchers

Journal

Modalities

Models

Abstract

Cardiovascular diseases belong to the leading causes of disability and premature death worldwide, including in Iran. It is predicted that the burden of the disease in Iran in 2025 will be more than doubled compared to 2005. Therefore, many forecasting models have been used to predict disease progression, estimate mortality rates, and assess risk factors. Our study focused on two time series prediction on models: autoregressive integrated moving average with exogenous variable (ARIMAX) and Convolutional neural network-long short-term memory network (CNN-LSTM). ARIMAX (6,1,6) had the best MSE of 0.655 among time series regression models. The prediction of this model shows a significant association in lag 4 and lag 6. Nitrogen dioxide (NO2) was also significant in lag 6, while CNN-LSTM had a much better MSE of 0.21. For the time series analysis and forecasts studied in this paper, deep learning models provided more accurate results than classical methods such as ARIMAX.© 2022. The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.