Cargando…

A multi-layer model for the early detection of COVID-19

Current COVID-19 screening efforts mainly rely on reported symptoms and the potential exposure to infected individuals. Here, we developed a machine-learning model for COVID-19 detection that uses four layers of information: (i) sociodemographic characteristics of the individual, (ii) spatio-tempora...

Descripción completa

Detalles Bibliográficos
Autores principales: Shmueli, Erez, Mansuri, Ronen, Porcilan, Matan, Amir, Tamar, Yosha, Lior, Yechezkel, Matan, Patalon, Tal, Handelman-Gotlib, Sharon, Gazit, Sivan, Yamin, Dan
Formato: Online Artículo Texto
Lenguaje:English
Publicado: The Royal Society 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8331231/
https://www.ncbi.nlm.nih.gov/pubmed/34343454
http://dx.doi.org/10.1098/rsif.2021.0284
_version_ 1783732876267749376
author Shmueli, Erez
Mansuri, Ronen
Porcilan, Matan
Amir, Tamar
Yosha, Lior
Yechezkel, Matan
Patalon, Tal
Handelman-Gotlib, Sharon
Gazit, Sivan
Yamin, Dan
author_facet Shmueli, Erez
Mansuri, Ronen
Porcilan, Matan
Amir, Tamar
Yosha, Lior
Yechezkel, Matan
Patalon, Tal
Handelman-Gotlib, Sharon
Gazit, Sivan
Yamin, Dan
author_sort Shmueli, Erez
collection PubMed
description Current COVID-19 screening efforts mainly rely on reported symptoms and the potential exposure to infected individuals. Here, we developed a machine-learning model for COVID-19 detection that uses four layers of information: (i) sociodemographic characteristics of the individual, (ii) spatio-temporal patterns of the disease, (iii) medical condition and general health consumption of the individual and (iv) information reported by the individual during the testing episode. We evaluated our model on 140 682 members of Maccabi Health Services who were tested for COVID-19 at least once between February and October 2020. These individuals underwent, in total, 264 516 COVID-19 PCR tests, out of which 16 512 were positive. Our multi-layer model obtained an area under the curve (AUC) of 81.6% when evaluated over all the individuals in the dataset, and an AUC of 72.8% when only individuals who did not report any symptom were included. Furthermore, considering only information collected before the testing episode—i.e. before the individual had the chance to report on any symptom—our model could reach a considerably high AUC of 79.5%. Our ability to predict early on the outcomes of COVID-19 tests is pivotal for breaking transmission chains, and can be used for a more efficient testing policy.
format Online
Article
Text
id pubmed-8331231
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher The Royal Society
record_format MEDLINE/PubMed
spelling pubmed-83312312021-08-11 A multi-layer model for the early detection of COVID-19 Shmueli, Erez Mansuri, Ronen Porcilan, Matan Amir, Tamar Yosha, Lior Yechezkel, Matan Patalon, Tal Handelman-Gotlib, Sharon Gazit, Sivan Yamin, Dan J R Soc Interface Life Sciences–Mathematics interface Current COVID-19 screening efforts mainly rely on reported symptoms and the potential exposure to infected individuals. Here, we developed a machine-learning model for COVID-19 detection that uses four layers of information: (i) sociodemographic characteristics of the individual, (ii) spatio-temporal patterns of the disease, (iii) medical condition and general health consumption of the individual and (iv) information reported by the individual during the testing episode. We evaluated our model on 140 682 members of Maccabi Health Services who were tested for COVID-19 at least once between February and October 2020. These individuals underwent, in total, 264 516 COVID-19 PCR tests, out of which 16 512 were positive. Our multi-layer model obtained an area under the curve (AUC) of 81.6% when evaluated over all the individuals in the dataset, and an AUC of 72.8% when only individuals who did not report any symptom were included. Furthermore, considering only information collected before the testing episode—i.e. before the individual had the chance to report on any symptom—our model could reach a considerably high AUC of 79.5%. Our ability to predict early on the outcomes of COVID-19 tests is pivotal for breaking transmission chains, and can be used for a more efficient testing policy. The Royal Society 2021-08-04 /pmc/articles/PMC8331231/ /pubmed/34343454 http://dx.doi.org/10.1098/rsif.2021.0284 Text en © 2021 The Authors. https://creativecommons.org/licenses/by/4.0/Published by the Royal Society under the terms of the Creative Commons Attribution License http://creativecommons.org/licenses/by/4.0/ (https://creativecommons.org/licenses/by/4.0/) , which permits unrestricted use, provided the original author and source are credited.
spellingShingle Life Sciences–Mathematics interface
Shmueli, Erez
Mansuri, Ronen
Porcilan, Matan
Amir, Tamar
Yosha, Lior
Yechezkel, Matan
Patalon, Tal
Handelman-Gotlib, Sharon
Gazit, Sivan
Yamin, Dan
A multi-layer model for the early detection of COVID-19
title A multi-layer model for the early detection of COVID-19
title_full A multi-layer model for the early detection of COVID-19
title_fullStr A multi-layer model for the early detection of COVID-19
title_full_unstemmed A multi-layer model for the early detection of COVID-19
title_short A multi-layer model for the early detection of COVID-19
title_sort multi-layer model for the early detection of covid-19
topic Life Sciences–Mathematics interface
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8331231/
https://www.ncbi.nlm.nih.gov/pubmed/34343454
http://dx.doi.org/10.1098/rsif.2021.0284
work_keys_str_mv AT shmuelierez amultilayermodelfortheearlydetectionofcovid19
AT mansurironen amultilayermodelfortheearlydetectionofcovid19
AT porcilanmatan amultilayermodelfortheearlydetectionofcovid19
AT amirtamar amultilayermodelfortheearlydetectionofcovid19
AT yoshalior amultilayermodelfortheearlydetectionofcovid19
AT yechezkelmatan amultilayermodelfortheearlydetectionofcovid19
AT patalontal amultilayermodelfortheearlydetectionofcovid19
AT handelmangotlibsharon amultilayermodelfortheearlydetectionofcovid19
AT gazitsivan amultilayermodelfortheearlydetectionofcovid19
AT yamindan amultilayermodelfortheearlydetectionofcovid19
AT shmuelierez multilayermodelfortheearlydetectionofcovid19
AT mansurironen multilayermodelfortheearlydetectionofcovid19
AT porcilanmatan multilayermodelfortheearlydetectionofcovid19
AT amirtamar multilayermodelfortheearlydetectionofcovid19
AT yoshalior multilayermodelfortheearlydetectionofcovid19
AT yechezkelmatan multilayermodelfortheearlydetectionofcovid19
AT patalontal multilayermodelfortheearlydetectionofcovid19
AT handelmangotlibsharon multilayermodelfortheearlydetectionofcovid19
AT gazitsivan multilayermodelfortheearlydetectionofcovid19
AT yamindan multilayermodelfortheearlydetectionofcovid19