Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 May 13;12(1):7969.
doi: 10.1038/s41598-022-11693-9.

BeCaked: An Explainable Artificial Intelligence Model for COVID-19 Forecasting

Duc Q Nguyen  1   2 , Nghia Q Vo  3   2 , Thinh T Nguyen  1   2 , Khuong Nguyen-An  1   2 , Quang H Nguyen  1   2 , Dang N Tran  4 , Tho T Quan  5   6
Affiliations

BeCaked: An Explainable Artificial Intelligence Model for COVID-19 Forecasting

Duc Q Nguyen et al. Sci Rep. .

Abstract

From the end of 2019, one of the most serious and largest spread pandemics occurred in Wuhan (China) named Coronavirus (COVID-19). As reported by the World Health Organization, there are currently more than 100 million infectious cases with an average mortality rate of about five percent all over the world. To avoid serious consequences on people's lives and the economy, policies and actions need to be suitably made in time. To do that, the authorities need to know the future trend in the development process of this pandemic. This is the reason why forecasting models play an important role in controlling the pandemic situation. However, the behavior of this pandemic is extremely complicated and difficult to be analyzed, so that an effective model is not only considered on accurate forecasting results but also the explainable capability for human experts to take action pro-actively. With the recent advancement of Artificial Intelligence (AI) techniques, the emerging Deep Learning (DL) models have been proving highly effective when forecasting this pandemic future from the huge historical data. However, the main weakness of DL models is lacking the explanation capabilities. To overcome this limitation, we introduce a novel combination of the Susceptible-Infectious-Recovered-Deceased (SIRD) compartmental model and Variational Autoencoder (VAE) neural network known as BeCaked. With pandemic data provided by the Johns Hopkins University Center for Systems Science and Engineering, our model achieves 0.98 [Formula: see text] and 0.012 MAPE at world level with 31-step forecast and up to 0.99 [Formula: see text] and 0.0026 MAPE at country level with 15-step forecast on predicting daily infectious cases. Not only enjoying high accuracy, but BeCaked also offers useful justifications for its results based on the parameters of the SIRD model. Therefore, BeCaked can be used as a reference for authorities or medical experts to make on time right decisions.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Figure 1
Figure 1
Illustration of the usage of deep learning and explainable model for COVID-19 forecast.
Figure 2
Figure 2
The physical architecture of an RNN.
Figure 3
Figure 3
An unfolded RNN.
Figure 4
Figure 4
The structure of an LSTM cell.
Figure 5
Figure 5
The concept of autoencoder.
Figure 6
Figure 6
The concept of the SIRD model.
Figure 7
Figure 7
The architecture of BeCaked model.
Figure 8
Figure 8
Comparison of the number of cases between real data and BeCaked forecasting results from 161st–191st day (Jul. 1st 2020–Jul. 31st 2020) of the world.
Figure 9
Figure 9
Predicted transition rates from 161st–191st day (Jul. 1st 2020–Jul. 31st 2020) of the world.
Figure 10
Figure 10
Forecasting results with transition rates of 161th–222nd day (Jul. 1st 2020–Aug. 31st 2020) using data of 151st–160th day (Jun. 21st 2020–Jun. 30th 2020).
Figure 11
Figure 11
Forecasting results with transition rates of 176th–222nd day (Jul. 16th 2020–Aug. 31st 2020) using data of 166th–175th day (Jul. 6th 2020–Jul. 15th 2020).
Figure 12
Figure 12
Forecasting results with transition rates of 192th–222nd day (Aug. 1st 2020–Aug. 31st 2020) using data of 182nd–191th day (Jul. 22nd 2020–Jul. 31st 2020).
Figure 13
Figure 13
Forecasting results with transition rates of 207th–222nd day (Aug. 16th 2020–Aug. 31st 2020) using data of 197th–206th day (Aug. 6th 2020–Aug. 15th 2020).
Figure 14
Figure 14
Homepage of the web user interface including COVID-19 map created using HERE Maps.
Figure 15
Figure 15
Long-time forecasting page.
See this image and copyright information in PMC

References

    1. Bengio Y, Courvill A, Vincent P. Representation learning: A review and new perspectives. IEEE Trans. Pattern Anal. Mach. Intell. 2013;35:1798–1828. doi: 10.1109/TPAMI.2013.50. - DOI - PubMed
    1. WHO. Coronavirus. https://www.who.int/health-topics/coronavirus (2021).
    1. Block P, et al. Social network-based distancing strategies to flatten the COVID-19 curve in a post-lockdown world. Nat. Hum. Behav. 2020;4:588–596. doi: 10.1038/s41562-020-0898-6. - DOI - PubMed
    1. Rumelhart D, Hinton G, Williams R. Learning representations by back-propagating errors. Nature. 1986;323:533–536. doi: 10.1038/323533a0. - DOI
    1. Hochreiter S, Schmidhuber J. Long short-term memory. Neural Comput. 1997;9:1735–1780. doi: 10.1162/neco.1997.9.8.1735. - DOI - PubMed

Publication types