Abstract:The rapid spread of COVID-19, caused by the SARS-CoV-2 virus, has had and continues to pose a significant threat to global health. We propose a predictive model based on the gated recurrent unit (GRU) that investigates the influence of non-pharmaceutical interventions (NPIs) on the progression of COVID-19. The proposed model is validated by case studies for multiple states in the United States. It should be noted that the proposed model can be generalized to other regions of interest. The results show that the… Show more
“… GQ1, SQ1, SQ3, SQ4 Bi et al. ( 2022 ) USA Present a GRU-based hybrid model to predict the spread of COVID-19 across USA GQ1, GQ2, SQ1, SQ3 Bushira and Ongala ( 2021 ) Namibia Using geospatial technologies, identify and map COVID-19 risk zones and model future COVID-19 responses in Namibia. GQ1, GQ2, SQ1, SQ4 Casini and Roccetti ( 2020 ) Italy Present three computational models of increasing complexity (linear, negative binomial regression, and cognitive) to identify the one that better correlates the relationship between Italian tourist flows during the summer of 2020 and the resurgence of COVID-19 cases across the country.…”
Section: Resultsmentioning
confidence: 99%
“… 2022 ; Bi et al. 2022 ) Fig. 10 Question for domain 2 GRADE assessment: ML prediction compared to real geographic location for tracking COVID-19 pandemics critical geolocations: ( a ) Ozik et al 2021 have no outcomes for cumulative COVID-19 cases’ prediction.…”
This systematic review aims to study and classify machine learning models that predict pandemics’ evolution within affected regions or countries. The advantage of this systematic review is that it allows the health authorities to decide what prediction model fits best depending upon the region’s criticality and optimize hospitals’ approaches to preparing and anticipating patient care. We searched ACM Digital Library, Biomed Central, BioRxiv+MedRxiv, BMJ, Computers and Applied Sciences, IEEEXplore, JMIR Medical Informatics, Medline Daily Updates, Nature, Oxford Academic, PubMed, Sage Online, ScienceDirect, Scopus, SpringerLink, Web of Science, and Wiley Online Library between 1 January 2020 and 31 July 2022. We divided the interventions into similarities between cumulative COVID-19 real cases and machine learning prediction models’ ability to track pandemics trending. We included 45 studies that rated low to high risk of bias. The standardized mean differences (SMD) for the two groups were 0.18, 95% CI, with interval of [0.01, 0.35],
=0, and
p
value=0.04. We built a taxonomic analysis of the included studies and determined two domains: pandemics trending prediction models and geolocation tracking models. We performed the meta-analysis and data synthesis and got low publication bias because of missing results. The level of certainty varied from very low to high. By submitting the 45 studies on the risk of bias, the levels of certainty, the summary of findings, and the statistical analysis via the forest and funnel plots assessments, we could determine the satisfactory statistical significance homogeneity across the included studies to simulate the progress of the pandemics and help the healthcare authorities to take preventive decisions.
“… GQ1, SQ1, SQ3, SQ4 Bi et al. ( 2022 ) USA Present a GRU-based hybrid model to predict the spread of COVID-19 across USA GQ1, GQ2, SQ1, SQ3 Bushira and Ongala ( 2021 ) Namibia Using geospatial technologies, identify and map COVID-19 risk zones and model future COVID-19 responses in Namibia. GQ1, GQ2, SQ1, SQ4 Casini and Roccetti ( 2020 ) Italy Present three computational models of increasing complexity (linear, negative binomial regression, and cognitive) to identify the one that better correlates the relationship between Italian tourist flows during the summer of 2020 and the resurgence of COVID-19 cases across the country.…”
Section: Resultsmentioning
confidence: 99%
“… 2022 ; Bi et al. 2022 ) Fig. 10 Question for domain 2 GRADE assessment: ML prediction compared to real geographic location for tracking COVID-19 pandemics critical geolocations: ( a ) Ozik et al 2021 have no outcomes for cumulative COVID-19 cases’ prediction.…”
This systematic review aims to study and classify machine learning models that predict pandemics’ evolution within affected regions or countries. The advantage of this systematic review is that it allows the health authorities to decide what prediction model fits best depending upon the region’s criticality and optimize hospitals’ approaches to preparing and anticipating patient care. We searched ACM Digital Library, Biomed Central, BioRxiv+MedRxiv, BMJ, Computers and Applied Sciences, IEEEXplore, JMIR Medical Informatics, Medline Daily Updates, Nature, Oxford Academic, PubMed, Sage Online, ScienceDirect, Scopus, SpringerLink, Web of Science, and Wiley Online Library between 1 January 2020 and 31 July 2022. We divided the interventions into similarities between cumulative COVID-19 real cases and machine learning prediction models’ ability to track pandemics trending. We included 45 studies that rated low to high risk of bias. The standardized mean differences (SMD) for the two groups were 0.18, 95% CI, with interval of [0.01, 0.35],
=0, and
p
value=0.04. We built a taxonomic analysis of the included studies and determined two domains: pandemics trending prediction models and geolocation tracking models. We performed the meta-analysis and data synthesis and got low publication bias because of missing results. The level of certainty varied from very low to high. By submitting the 45 studies on the risk of bias, the levels of certainty, the summary of findings, and the statistical analysis via the forest and funnel plots assessments, we could determine the satisfactory statistical significance homogeneity across the included studies to simulate the progress of the pandemics and help the healthcare authorities to take preventive decisions.
“…In addition, Bi, Fili and Hu [10] tested the performance of GRU against other models for 6-, 12-, 18-, 24-and 30-day periods. It was found that GRU outperformed LASSO for all but the 6-day prediction with a MAE that ranged from 19.62% to 32.72% whilst LASSO ' s MAE was between 25.71% and 42.85%.…”
Section: Performance Of Deep Learning Modelsmentioning
The accurate forecasting of Covid-19 cases remains a challenge even for comprehensive statistical models due to the complex dynamics inherent in largely urbanised populations. This difficulty, coupled with recent developments in artificial intelligence (AI), has led to a burgeoning interest in implementing machine learning (ML) algorithms to aid such predictions. Ever since the outbreak, various ML-based forecasting approaches have been trained and tested to support the design of intervention and healthcare strategies to protect the population. It is thus vital to assess the performance of these algorithms such that their accuracies, along with strengths and weaknesses, are known. This paper compares the forecasting performance of six commonly used ML models for Covid-19 forecasting. A brief analysis of the algorithms suitability for forecasting is also provided as part of the review. It is found that different algorithms were able to predict with varying degrees of accuracies in different scenarios, and that there is yet to be a definitive algorithm that is best for forecasting the Covid-19 outbreak.
“…The approach to forecasting the number of patients involves various machine learning techniques to discover the pattern of epidemic data. Various techniques have been used, such as Support Vector Regression [6], Deep Learning models including Recurrent Neural Network (RNN), [7] Gated Recurrent Unit (GRU) [8], Long Short-Term Memory (LSTM) [9], and others. The goal of this study is to use machine learning for analyzing and predicting the number of COVID-19 patients.…”
COVID-19 is a situation that has spread worldwide since 2019. This study predicts the number of patients with COVID-19 in Thailand. Using data between January 22, 2020, and December 31, 2021, we collect confirmed cases from John Hopkins open data. Using the machine learning model to predict the number of patient cases in the country helps the government manage its policies and resources. In this study, the K-Means clustering algorithm performs to group the countries that have similar patterns of confirmed cases to Thailand. Clustering results show that Japan, Malaysia, the Philippines, Bangladesh, Cuba, Iraq, Mexico, and Vietnam are all in the same cluster as Thailand. Using Long Short-Term Memory (LSTM) to predict the confirmed case of Thailand by feeding the model pairs of countries in the same cluster as Thailand, the performance of LSTM shows that using pairs of countries between Bangladesh, Japan, and Mexico with Thailand has the lowest error on MAPE, respectively, when compared to using only Thailand data.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
