Deep learning in public health: Comparative predictive models for COVID-19 case forecasting

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(3), P. e0294289 - e0294289

Published: March 14, 2024

The COVID-19 pandemic has had a significant impact on both the United Arab Emirates (UAE) and Malaysia, emphasizing importance of developing accurate reliable forecasting mechanisms to guide public health responses policies. In this study, we compared several cutting-edge deep learning models, including Long Short-Term Memory (LSTM), bidirectional LSTM, Convolutional Neural Networks (CNN), hybrid CNN-LSTM, Multilayer Perceptron’s, Recurrent (RNN), project cases in aforementioned regions. These models were calibrated evaluated using comprehensive dataset that includes confirmed case counts, demographic data, relevant socioeconomic factors. To enhance performance these Bayesian optimization techniques employed. Subsequently, re-evaluated compare their effectiveness. Analytic approaches, predictive retrospective nature, used interpret data. Our primary objective was determine most effective model for predicting Malaysia. findings indicate selected algorithms proficient cases, although efficacy varied across different models. After thorough evaluation, architectures suitable specific conditions UAE Malaysia identified. study contributes significantly ongoing efforts combat pandemic, providing crucial insights into application sophisticated precise timely cases. hold substantial value shaping strategies, enabling authorities develop targeted evidence-based interventions manage virus spread its populations confirms usefulness methodologies efficiently processing complex datasets generating projections, skill great healthcare professional settings.

Language: Английский

---

COVID-19: a new deep learning computer-aided model for classification

PeerJ Computer Science, Journal Year: 2021, Volume and Issue: 7, P. e358 - e358

Published: Feb. 18, 2021

Chest X-ray (CXR) imaging is one of the most feasible diagnosis modalities for early detection infection COVID-19 viruses, which classified as a pandemic according to World Health Organization (WHO) report in December 2019. rapid natural mutual virus that belongs coronavirus family. CXR scans are vital tools detect monitor further and control its spread. Classification aims whether subject infected or not. In this article, model proposed analyzing evaluating grayscale images called X-Ray COVID Network (CXRVN) based on three different datasets. The CXRVN lightweight architecture depends single fully connected layer representing essential features thus reducing total memory usage processing time verse pre-trained models others. adopts two optimizers: mini-batch gradient descent Adam optimizer, has almost same performance. Besides, accepts perfect image representation consume less storage time. Hence, can analyze with high accuracy few milliseconds. consequences learning process focus decision making using scoring function SoftMax leads rate true-positive classification. trained datasets compared models: GoogleNet, ResNet AlexNet, fine-tuning transfer technologies evaluation process. To verify effectiveness model, it was evaluated terms well-known performance measures such precision, sensitivity, F 1-score accuracy. results recall, accuracy, F1 score demonstrated that, after GAN augmentation, reached 96.7% experiment 2 (Dataset-2) classes 93.07% experiment-3 (Dataset-3) classes, while average 94.5%.

Language: Английский

---

COVID-Net CXR-2: An Enhanced Deep Convolutional Neural Network Design for Detection of COVID-19 Cases From Chest X-ray Images

Frontiers in Medicine, Journal Year: 2022, Volume and Issue: 9

Published: June 10, 2022

As the COVID-19 pandemic devastates globally, use of chest X-ray (CXR) imaging as a complimentary screening strategy to RT-PCR testing continues grow given its routine clinical for respiratory complaint. part COVID-Net open source initiative, we introduce CXR-2, an enhanced deep convolutional neural network design detection from CXR images built using greater quantity and diversity patients than original COVID-Net. We also new benchmark dataset composed 19,203 multinational cohort 16,656 at least 51 countries, making it largest, most diverse in access form. The CXR-2 achieves sensitivity positive predictive value 95.5 97.0%, respectively, was audited transparent responsible manner. Explainability-driven performance validation used during auditing gain deeper insights decision-making behavior ensure clinically relevant factors are leveraged improving trust usage. Radiologist conducted, where select cases were reviewed reported on by two board-certified radiologists with over 10 19 years experience, showed that critical consistent radiologist interpretations.

Language: Английский

---

Forecasting Covid-19 Transmission with ARIMA and LSTM Techniques in Morocco

SN Computer Science, Journal Year: 2022, Volume and Issue: 3(2)

Published: Jan. 14, 2022

Language: Английский

---

SecureFed: federated learning empowered medical imaging technique to analyze lung abnormalities in chest X-rays

International Journal of Machine Learning and Cybernetics, Journal Year: 2023, Volume and Issue: 14(8), P. 2659 - 2670

Published: Feb. 14, 2023

Language: Английский

---

Wastewater network infrastructure in public health: Applications and learnings from the COVID-19 pandemic

PLOS Global Public Health, Journal Year: 2021, Volume and Issue: 1(12), P. e0000061 - e0000061

Published: Dec. 2, 2021

Accurate estimates of COVID-19 burden infections in communities can inform public health strategy for the current pandemic. Wastewater based epidemiology (WBE) leverages sewer infrastructure to provide insights on rates infection by measuring viral concentrations wastewater. By accessing network at various junctures, important regarding disease activity be gained. The analysis sewage wastewater treatment plant level enables population-level surveillance trends and virus mutations. At neighborhood level, WBE used describe community thereby facilitating local efforts targeted mitigation. Finally, building suggest presence prompt individual testing. In this critical review, we types data that obtained through varying levels analysis, concrete plans implementation, actions taken infectious diseases, using recent successful applications during pandemic illustration.

Language: Английский

---

Prediction of COVID-19 Trend in India and Its Four Worst-Affected States Using Modified SEIRD and LSTM Models

SN Computer Science, Journal Year: 2021, Volume and Issue: 2(3)

Published: April 20, 2021

Language: Английский

---

SIMLR: Machine Learning inside the SIR Model for COVID-19 Forecasting

Forecasting, Journal Year: 2022, Volume and Issue: 4(1), P. 72 - 94

Published: Jan. 13, 2022

Accurate forecasts of the number newly infected people during an epidemic are critical for making effective timely decisions. This paper addresses this challenge using SIMLR model, which incorporates machine learning (ML) into epidemiological SIR model. For each region, tracks changes in policies implemented at government level, it uses to estimate time-varying parameters model forecasting new infections one four weeks advance. It also probability those these future times, is essential longer-range forecasts. We applied data from Canada and United States, show that its mean average percentage error as good state-of-the-art models, with added advantage being interpretable expect approach will be useful not only COVID-19 infections, but predicting evolution other infectious diseases.

Language: Английский

---

High-resolution short-term prediction of the COVID-19 epidemic based on spatial-temporal model modified by historical meteorological data

Fundamental Research, Journal Year: 2024, Volume and Issue: 4(3), P. 527 - 539

Published: March 5, 2024

In the global challenge of Coronavirus disease 2019 (COVID-19) pandemic, accurate prediction daily new cases is crucial for epidemic prevention and socioeconomic planning. contrast to traditional local, one-dimensional time-series data-based infection models, study introduces an innovative approach by formulating short-term problem in a region as multidimensional, gridded time series both input targets. A spatial-temporal depth model COVID-19 (ConvLSTM) presented, further ConvLSTM integrating historical meteorological factors (Meteor-ConvLSTM) refined, considering influence on propagation COVID-19. The correlation between 10 dynamic progression was evaluated, employing spatial analysis techniques (spatial autocorrelation analysis, trend surface etc.) describe temporal characteristics epidemic. Leveraging original ConvLSTM, artificial neural network layer introduced learn how impact spread, providing 5-day forecast at 0.01° × pixel resolution. Simulation results using real dataset from 3.15 outbreak Shanghai demonstrate efficacy Meteor-ConvLSTM, with reduced RMSE 0.110 increased

Language: Английский

---

Comparative performance of hybrid model based on discrete wavelet transform and ARIMA models in prediction incidence of COVID-19

Heliyon, Journal Year: 2024, Volume and Issue: 10(13), P. e33848 - e33848

Published: June 28, 2024

Public health surveillance is an important aspect of outbreak early warning based on prediction models. The present study compares a hybrid model discrete wavelet transform (DWT) and ARIMA (Autoregressive Integrated Moving Average) for predicting incidence cases due to COVID-19.

Language: Английский

---