Short-term local predictions of COVID-19 in the United Kingdom using dynamic supervised machine learning algorithms. Wang, X., Dong, Y., Thompson, W. D., Nair, H., & Li, Y. Communications medicine, 2:119, 2022. Place: Englanddoi abstract bibtex BACKGROUND: Short-term prediction of COVID-19 epidemics is crucial to decision making. We aimed to develop supervised machine-learning algorithms on multiple digital metrics including symptom search trends, population mobility, and vaccination coverage to predict local-level COVID-19 growth rates in the UK. METHODS: Using dynamic supervised machine-learning algorithms based on log-linear regression, we explored optimal models for 1-week, 2-week, and 3-week ahead prediction of COVID-19 growth rate at lower tier local authority level over time. Model performance was assessed by calculating mean squared error (MSE) of prospective prediction, and naïve model and fixed-predictors model were used as reference models. We assessed real-time model performance for eight five-weeks-apart checkpoints between 1st March and 14th November 2021. We developed an online application (COVIDPredLTLA) that visualised the real-time predictions for the present week, and the next one and two weeks. RESULTS: Here we show that the median MSEs of the optimal models for 1-week, 2-week, and 3-week ahead prediction are 0.12 (IQR: 0.08-0.22), 0.29 (0.19-0.38), and 0.37 (0.25-0.47), respectively. Compared with naïve models, the optimal models maintain increased accuracy (reducing MSE by a range of 21-35%), including May-June 2021 when the delta variant spread across the UK. Compared with the fixed-predictors model, the advantage of dynamic models is observed after several iterations of update. CONCLUSIONS: With flexible data-driven predictors selection process, our dynamic modelling framework shows promises in predicting short-term changes in COVID-19 cases. The online application (COVIDPredLTLA) could assist decision-making for control measures and planning of healthcare capacity in future epidemic growths.
@article{wang_short-term_2022,
title = {Short-term local predictions of {COVID}-19 in the {United} {Kingdom} using dynamic supervised machine learning algorithms.},
volume = {2},
copyright = {© The Author(s) 2022.},
issn = {2730-664X},
doi = {10.1038/s43856-022-00184-7},
abstract = {BACKGROUND: Short-term prediction of COVID-19 epidemics is crucial to decision making. We aimed to develop supervised machine-learning algorithms on multiple digital metrics including symptom search trends, population mobility, and vaccination coverage to predict local-level COVID-19 growth rates in the UK. METHODS: Using dynamic supervised machine-learning algorithms based on log-linear regression, we explored optimal models for 1-week, 2-week, and 3-week ahead prediction of COVID-19 growth rate at lower tier local authority level over time. Model performance was assessed by calculating mean squared error (MSE) of prospective prediction, and naïve model and fixed-predictors model were used as reference models. We assessed real-time model performance for eight five-weeks-apart checkpoints between 1st March and 14th November 2021. We developed an online application (COVIDPredLTLA) that visualised the real-time predictions for the present week, and the next one and two weeks. RESULTS: Here we show that the median MSEs of the optimal models for 1-week, 2-week, and 3-week ahead prediction are 0.12 (IQR: 0.08-0.22), 0.29 (0.19-0.38), and 0.37 (0.25-0.47), respectively. Compared with naïve models, the optimal models maintain increased accuracy (reducing MSE by a range of 21-35\%), including May-June 2021 when the delta variant spread across the UK. Compared with the fixed-predictors model, the advantage of dynamic models is observed after several iterations of update. CONCLUSIONS: With flexible data-driven predictors selection process, our dynamic modelling framework shows promises in predicting short-term changes in COVID-19 cases. The online application (COVIDPredLTLA) could assist decision-making for control measures and planning of healthcare capacity in future epidemic growths.},
language = {eng},
journal = {Communications medicine},
author = {Wang, Xin and Dong, Yijia and Thompson, William David and Nair, Harish and Li, You},
year = {2022},
pmid = {36168444},
pmcid = {PMC9509378},
note = {Place: England},
keywords = {Disease prevention, Epidemiology},
pages = {119},
}
Downloads: 0
{"_id":"rtypeveAr5Z2bpncF","bibbaseid":"wang-dong-thompson-nair-li-shorttermlocalpredictionsofcovid19intheunitedkingdomusingdynamicsupervisedmachinelearningalgorithms-2022","author_short":["Wang, X.","Dong, Y.","Thompson, W. D.","Nair, H.","Li, Y."],"bibdata":{"bibtype":"article","type":"article","title":"Short-term local predictions of COVID-19 in the United Kingdom using dynamic supervised machine learning algorithms.","volume":"2","copyright":"© The Author(s) 2022.","issn":"2730-664X","doi":"10.1038/s43856-022-00184-7","abstract":"BACKGROUND: Short-term prediction of COVID-19 epidemics is crucial to decision making. We aimed to develop supervised machine-learning algorithms on multiple digital metrics including symptom search trends, population mobility, and vaccination coverage to predict local-level COVID-19 growth rates in the UK. METHODS: Using dynamic supervised machine-learning algorithms based on log-linear regression, we explored optimal models for 1-week, 2-week, and 3-week ahead prediction of COVID-19 growth rate at lower tier local authority level over time. Model performance was assessed by calculating mean squared error (MSE) of prospective prediction, and naïve model and fixed-predictors model were used as reference models. We assessed real-time model performance for eight five-weeks-apart checkpoints between 1st March and 14th November 2021. We developed an online application (COVIDPredLTLA) that visualised the real-time predictions for the present week, and the next one and two weeks. RESULTS: Here we show that the median MSEs of the optimal models for 1-week, 2-week, and 3-week ahead prediction are 0.12 (IQR: 0.08-0.22), 0.29 (0.19-0.38), and 0.37 (0.25-0.47), respectively. Compared with naïve models, the optimal models maintain increased accuracy (reducing MSE by a range of 21-35%), including May-June 2021 when the delta variant spread across the UK. Compared with the fixed-predictors model, the advantage of dynamic models is observed after several iterations of update. CONCLUSIONS: With flexible data-driven predictors selection process, our dynamic modelling framework shows promises in predicting short-term changes in COVID-19 cases. The online application (COVIDPredLTLA) could assist decision-making for control measures and planning of healthcare capacity in future epidemic growths.","language":"eng","journal":"Communications medicine","author":[{"propositions":[],"lastnames":["Wang"],"firstnames":["Xin"],"suffixes":[]},{"propositions":[],"lastnames":["Dong"],"firstnames":["Yijia"],"suffixes":[]},{"propositions":[],"lastnames":["Thompson"],"firstnames":["William","David"],"suffixes":[]},{"propositions":[],"lastnames":["Nair"],"firstnames":["Harish"],"suffixes":[]},{"propositions":[],"lastnames":["Li"],"firstnames":["You"],"suffixes":[]}],"year":"2022","pmid":"36168444","pmcid":"PMC9509378","note":"Place: England","keywords":"Disease prevention, Epidemiology","pages":"119","bibtex":"@article{wang_short-term_2022,\n\ttitle = {Short-term local predictions of {COVID}-19 in the {United} {Kingdom} using dynamic supervised machine learning algorithms.},\n\tvolume = {2},\n\tcopyright = {© The Author(s) 2022.},\n\tissn = {2730-664X},\n\tdoi = {10.1038/s43856-022-00184-7},\n\tabstract = {BACKGROUND: Short-term prediction of COVID-19 epidemics is crucial to decision making. We aimed to develop supervised machine-learning algorithms on multiple digital metrics including symptom search trends, population mobility, and vaccination coverage to predict local-level COVID-19 growth rates in the UK. METHODS: Using dynamic supervised machine-learning algorithms based on log-linear regression, we explored optimal models for 1-week, 2-week, and 3-week ahead prediction of COVID-19 growth rate at lower tier local authority level over time. Model performance was assessed by calculating mean squared error (MSE) of prospective prediction, and naïve model and fixed-predictors model were used as reference models. We assessed real-time model performance for eight five-weeks-apart checkpoints between 1st March and 14th November 2021. We developed an online application (COVIDPredLTLA) that visualised the real-time predictions for the present week, and the next one and two weeks. RESULTS: Here we show that the median MSEs of the optimal models for 1-week, 2-week, and 3-week ahead prediction are 0.12 (IQR: 0.08-0.22), 0.29 (0.19-0.38), and 0.37 (0.25-0.47), respectively. Compared with naïve models, the optimal models maintain increased accuracy (reducing MSE by a range of 21-35\\%), including May-June 2021 when the delta variant spread across the UK. Compared with the fixed-predictors model, the advantage of dynamic models is observed after several iterations of update. CONCLUSIONS: With flexible data-driven predictors selection process, our dynamic modelling framework shows promises in predicting short-term changes in COVID-19 cases. The online application (COVIDPredLTLA) could assist decision-making for control measures and planning of healthcare capacity in future epidemic growths.},\n\tlanguage = {eng},\n\tjournal = {Communications medicine},\n\tauthor = {Wang, Xin and Dong, Yijia and Thompson, William David and Nair, Harish and Li, You},\n\tyear = {2022},\n\tpmid = {36168444},\n\tpmcid = {PMC9509378},\n\tnote = {Place: England},\n\tkeywords = {Disease prevention, Epidemiology},\n\tpages = {119},\n}\n\n","author_short":["Wang, X.","Dong, Y.","Thompson, W. D.","Nair, H.","Li, Y."],"key":"wang_short-term_2022","id":"wang_short-term_2022","bibbaseid":"wang-dong-thompson-nair-li-shorttermlocalpredictionsofcovid19intheunitedkingdomusingdynamicsupervisedmachinelearningalgorithms-2022","role":"author","urls":{},"keyword":["Disease prevention","Epidemiology"],"metadata":{"authorlinks":{}}},"bibtype":"article","biburl":"https://leoly2017.github.io/group/citations.bib","dataSources":["acjQJrDfd43dtXDph"],"keywords":["disease prevention","epidemiology"],"search_terms":["short","term","local","predictions","covid","united","kingdom","using","dynamic","supervised","machine","learning","algorithms","wang","dong","thompson","nair","li"],"title":"Short-term local predictions of COVID-19 in the United Kingdom using dynamic supervised machine learning algorithms.","year":2022}