Artificial neural networks for short-term forecasting of cases, deaths, and hospital beds occupancy in the COVID-19 pandemic at the Brazilian Amazon

Abstract

The first case of the novel coronavirus in Brazil was notified on February 26, 2020. After 21 days, the first case was reported in the second largest State of the Brazilian Amazon. The State of Pará presented difficulties in combating the pandemic, ranging from underreporting and a low number of tests to a large territorial distance between cities with installed hospital capacity. Due to these factors, mathematical data-driven short-term forecasting models can be a promising initiative to assist government officials in more agile and reliable actions. This study presents an approach based on artificial neural networks for the daily and cumulative forecasts of cases and deaths caused by COVID-19, and the forecast of demand for hospital beds. Six scenarios with different periods were used to identify the quality of the generated forecasting and the period in which they start to deteriorate. Results indicated that the computational model adapted capably to the training period and was able to make consistent short-term forecasts, especially for the cumulative variables and for demand hospital beds.

Use of Artificial Intelligence for Predicting COVID-19 Outcomes: A Scoping Review

“…The purpose of using the artificial intelligence method was to predict cases, death, mortality, and severity. Since one paper could predict more than one variable, among the 17 papers, seven papers forecasted mortality [1,7,11,[13][14][15]19], five papers forecasted daily report COVID-19 test-positive cases and death cases [4][5][6]12,16], four papers forecasted severity [3,9,17,18], and one paper predicted all of them [20]. Tree-based model was the most common choice, which was used seven times [9,11,12,13,15,16,19].…”

“…Tree-based model was the most common choice, which was used seven times [9,11,12,13,15,16,19]. The neural network method was a new trend, and five papers applied this method [3,4,5,7,20]. Three papers used the regression method [1,6,17].…”

Forecasting COVID-19 recovered cases with Artificial Neural Networks to enable designing an effective blood supply chain

“… Using data around the world and not separating regions or countries by region (??) 17 Wieczorek et al [ 31 ] Forecasting number of cases each day worldwide Information from the last 12 days plus geolocation coordinates of latitude and longitude have an impact on the cases correlations between neighboring countries Proposing a model, which can work as a part of an online system as a real-time predictor to help in the estimation of COVID-19 spread worldwide Using a limited number of data 18 de Barros Braga et al [ 32 ] Estimating the daily and cumulative cases and deaths caused by COVID-19 and demand for hospital beds Cumulative cases Cumulative deaths Municipal demography Occurrence date Name of State's municipalities Names of the health regions Training ANN with data from 6 different moments for providing the ability to evaluate the forecasting quality Using a limited number of data 19 Shetty and Pai [ 33 ] Forecasting the number of infected cases Daily reported cases Applying a fast training algorithm that is Extreme Learning machine to reduce the training time and using cuckoo search (CS) algorithm to select the parameters Not conducting a comparative analysis 20 Tamang et al [ 34 ] Estimating the number of rising cases and deaths in India, the USA, France, and the UK The number of days Presenting intelligent based optimum curve fitting and forecasting for different non-linear models Not conducting a comparative analysis 21 Melin et al [ 35 ] Predicting confirmed and COVID-19 deaths for 26 countries The confirmed and deaths Adopting the firefly algorithm for ensemble neural network optimization to COVID-19 time series prediction with type-2 fuzzy logic in a weighted average integration method Not conducting a comparison with different fuzzy extensions to measure the performance of the adopted model 22 Ardabili et al [ 36 ] ...…”

Multiple Output and Multi-Steps Prediction of COVID-19 Spread Using Weather and Vaccination Data

“…The suggested model can be used to perform multistage forecasting for the following days. Braga et al [8] have proposed an approach based on artificial neural networks to predict daily and cumulative cases and deaths caused by COVID-19, as well as forecasting the demand for hospital beds in the Brazilian Amazon. It has been shown that this last model was able to make consistent short-term predictions.…”