An investigation of multivariate data-driven deep learning models for predicting COVID-19 variants DOI Creative Commons

Akhmad Dimitri Baihaqi,

Novanto Yudistira, Edy Santoso

et al.

Healthcare Analytics, Journal Year: 2024, Volume and Issue: 5, P. 100331 - 100331

Published: April 7, 2024

The Coronavirus Disease 2019 (COVID-19) pandemic has swept almost all parts of the world. With increasing number COVID-19 cases worldwide, Severe Acute Respiratory Syn- drome 2 (SARS-CoV-2) mutated into various variants. Given increasingly dangerous conditions pandemic, it is crucial to predict cases. Deep Learning and Long Short-Term Memory (LSTM) have predicted disease progress with reasonable accuracy small errors. LSTM training used confirmed based on variants identified using European Centre for Prevention Control (ECDC) dataset containing from 30 countries. Tests were conducted Bidirectional (BiLSTM) models addition Recurrent Neural Network (RNN) as comparisons hidden size layer size. obtained result showed that in testing sizes 25, 50, 75, 100, RNN model provided better results, minimum Mean Squared Error (MSE) value 0.01 Root Square (RMSE) 0.012 B.1.427/B.1.429 variant a 100. Further 2, 3, 4, 5 shows BiLSTM MSE an RMSE 100 2.

Language: Английский

Bifurcations and model fitting of a discrete epidemic system with incubation period and saturated contact rate DOI
Limin Zhang,

Jiaxin Gu,

Guangyuan Liao

et al.

The Journal of Difference Equations and Applications, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 42

Published: Jan. 29, 2025

Language: Английский

Citations

0
A multi-population approach to epidemiological modeling of listeriosis transmission dynamics incorporating food and environmental contamination DOI Creative Commons
S.Y. Tchoumi, C. W. Chukwu,

Windarto Windarto

et al.

Healthcare Analytics, Journal Year: 2024, Volume and Issue: 5, P. 100344 - 100344

Published: May 27, 2024

Listeriosis is a food-borne disease that mainly affects pregnant women and newborns. We propose analyze deterministic model of by considering three groups individuals: newborns, women, others. Mathematical analysis the performed, equilibrium points are determined. The has equilibria, namely, disease-free equilibrium, bacteria-free endemic equilibrium. use Castillo-Chavez to establish equilibrium's global stability when basic reproduction number less than 1. local asymptotic also established using sign eigenvalues Jacobian matrix. non-standard finite difference scheme carry out numerical simulations confirm theoretical result conjecture further show impact specific parameters on dynamics infectious individuals observe we need intervene in all sub-populations reducing contact rate vertical transmission reduce infectious.

Language: Английский

Citations

1
An investigation of multivariate data-driven deep learning models for predicting COVID-19 variants DOI Creative Commons

Akhmad Dimitri Baihaqi,

Novanto Yudistira, Edy Santoso

et al.

Healthcare Analytics, Journal Year: 2024, Volume and Issue: 5, P. 100331 - 100331

Published: April 7, 2024

The Coronavirus Disease 2019 (COVID-19) pandemic has swept almost all parts of the world. With increasing number COVID-19 cases worldwide, Severe Acute Respiratory Syn- drome 2 (SARS-CoV-2) mutated into various variants. Given increasingly dangerous conditions pandemic, it is crucial to predict cases. Deep Learning and Long Short-Term Memory (LSTM) have predicted disease progress with reasonable accuracy small errors. LSTM training used confirmed based on variants identified using European Centre for Prevention Control (ECDC) dataset containing from 30 countries. Tests were conducted Bidirectional (BiLSTM) models addition Recurrent Neural Network (RNN) as comparisons hidden size layer size. obtained result showed that in testing sizes 25, 50, 75, 100, RNN model provided better results, minimum Mean Squared Error (MSE) value 0.01 Root Square (RMSE) 0.012 B.1.427/B.1.429 variant a 100. Further 2, 3, 4, 5 shows BiLSTM MSE an RMSE 100 2.

Language: Английский

Citations

0