Personalized predictions of adverse side effects of the COVID-19 vaccines.
AI, artificial intelligence
AZD1222
Adverse side effects
Artificial intelligence
AstraZeneca
COVID-19
COVID-19, coronavirus disease of 2019
Covaxin
KNN, K Nearest Neighbors
LR, logistic regression
ML, machine learning
MLP, Multi-Layer Perceptron
Machine learning
Moderna
Pfizer
RF, random forest
ROC, receiver operating characteristic
Sinopharm
Sputnik V
Symptom
Vaccine
Journal
Heliyon
ISSN: 2405-8440
Titre abrégé: Heliyon
Pays: England
ID NLM: 101672560
Informations de publication
Date de publication:
Jan 2023
Jan 2023
Historique:
received:
01
11
2021
revised:
27
12
2022
accepted:
27
12
2022
pubmed:
5
1
2023
medline:
5
1
2023
entrez:
4
1
2023
Statut:
ppublish
Résumé
Misconceptions about adverse side effects are thought to influence public acceptance of the Coronavirus disease 2019 (COVID-19) vaccines negatively. To address such perceived disadvantages of vaccines, a novel machine learning (ML) approach was designed to generate personalized predictions of the most common adverse side effects following injection of six different COVID-19 vaccines based on personal and health-related characteristics. Prospective data of adverse side effects following COVID-19 vaccination in 19943 participants from Iran and Switzerland was utilized. Six vaccines were studied: The AZD1222, Sputnik V, BBIBP-CorV, COVAXIN, BNT162b2, and the mRNA-1273 vaccine. The eight side effects were considered as the model output: fever, fatigue, headache, nausea, chills, joint pain, muscle pain, and injection site reactions. The total input parameters for the first and second dose predictions were 46 and 54 features, respectively, including age, gender, lifestyle variables, and medical history. The performances of multiple ML models were compared using Area Under the Receiver Operating Characteristic Curve (ROC-AUC). The total number of people receiving the first dose of the AZD1222, Sputnik V, BBIBP-CorV, COVAXIN, BNT162b2, and mRNA-1273 were 6022, 7290, 5279, 802, 277, and 273, respectively. For the second dose, the numbers were 2851, 5587, 3841, 599, 242 and 228. The Logistic Regression model for predicting different side effects of the first dose achieved ROC-AUCs of 0.620-0.686, 0.685-0.716, 0.632-0.727, 0.527-0.598, 0.548-0.655, 0.545-0.712 for the AZD1222, Sputnik V, BBIBP-CorV, COVAXIN, BNT162b2 and mRNA-1273 vaccines, respectively. The second dose models yielded ROC-AUCs of 0.777-0.867, 0.795-0.848, 0.857-0.906, 0.788-0.875, 0.683-0.850, and 0.486-0.680, respectively. Using a large cohort of recipients vaccinated with COVID-19 vaccines, a novel and personalized strategy was established to predict the occurrence of the most common adverse side effects with high accuracy. This technique can serve as a tool to inform COVID-19 vaccine selection and generate personalized factsheets to curb concerns about adverse side effects.
Sections du résumé
Background
UNASSIGNED
Misconceptions about adverse side effects are thought to influence public acceptance of the Coronavirus disease 2019 (COVID-19) vaccines negatively. To address such perceived disadvantages of vaccines, a novel machine learning (ML) approach was designed to generate personalized predictions of the most common adverse side effects following injection of six different COVID-19 vaccines based on personal and health-related characteristics.
Methods
UNASSIGNED
Prospective data of adverse side effects following COVID-19 vaccination in 19943 participants from Iran and Switzerland was utilized. Six vaccines were studied: The AZD1222, Sputnik V, BBIBP-CorV, COVAXIN, BNT162b2, and the mRNA-1273 vaccine. The eight side effects were considered as the model output: fever, fatigue, headache, nausea, chills, joint pain, muscle pain, and injection site reactions. The total input parameters for the first and second dose predictions were 46 and 54 features, respectively, including age, gender, lifestyle variables, and medical history. The performances of multiple ML models were compared using Area Under the Receiver Operating Characteristic Curve (ROC-AUC).
Results
UNASSIGNED
The total number of people receiving the first dose of the AZD1222, Sputnik V, BBIBP-CorV, COVAXIN, BNT162b2, and mRNA-1273 were 6022, 7290, 5279, 802, 277, and 273, respectively. For the second dose, the numbers were 2851, 5587, 3841, 599, 242 and 228. The Logistic Regression model for predicting different side effects of the first dose achieved ROC-AUCs of 0.620-0.686, 0.685-0.716, 0.632-0.727, 0.527-0.598, 0.548-0.655, 0.545-0.712 for the AZD1222, Sputnik V, BBIBP-CorV, COVAXIN, BNT162b2 and mRNA-1273 vaccines, respectively. The second dose models yielded ROC-AUCs of 0.777-0.867, 0.795-0.848, 0.857-0.906, 0.788-0.875, 0.683-0.850, and 0.486-0.680, respectively.
Conclusions
UNASSIGNED
Using a large cohort of recipients vaccinated with COVID-19 vaccines, a novel and personalized strategy was established to predict the occurrence of the most common adverse side effects with high accuracy. This technique can serve as a tool to inform COVID-19 vaccine selection and generate personalized factsheets to curb concerns about adverse side effects.
Identifiants
pubmed: 36597482
doi: 10.1016/j.heliyon.2022.e12753
pii: S2405-8440(22)04041-5
pmc: PMC9800018
doi:
Types de publication
Journal Article
Langues
eng
Pagination
e12753Informations de copyright
© 2022 The Authors.
Références
Lancet Infect Dis. 2021 Jul;21(7):939-949
pubmed: 33930320
JAMA. 2022 Jan 25;327(4):331-340
pubmed: 35076665
BMC Med. 2015 Jan 06;13:1
pubmed: 25563062
Front Digit Health. 2022 Jan 13;3:681608
pubmed: 35098205
JAMA Netw Open. 2021 Dec 1;4(12):e2140364
pubmed: 34935921
BMJ. 2021 Aug 26;374:n1931
pubmed: 34446426
J Med Internet Res. 2020 Oct 27;22(10):e21476
pubmed: 32946413
BMC Med. 2021 Jul 28;19(1):173
pubmed: 34315454
Cochrane Database Syst Rev. 2018 Jun 26;6:CD002733
pubmed: 29943802
Front Public Health. 2021 Apr 21;9:626852
pubmed: 33968880
NPJ Vaccines. 2019 Sep 24;4:39
pubmed: 31583123
BMJ. 2021 Dec 16;375:e068665
pubmed: 34916207
Int Immunopharmacol. 2021 Aug;97:107724
pubmed: 33951558
Front Artif Intell. 2021 Jun 22;4:673527
pubmed: 34250465
Pers Individ Dif. 2021 Apr;172:110590
pubmed: 33518869
Comput Biol Med. 2021 May;132:104324
pubmed: 33774270
Int J Environ Res Public Health. 2021 Apr 06;18(7):
pubmed: 33917544
Front Artif Intell. 2020 Feb 28;3:4
pubmed: 33733124
Sci Rep. 2021 Feb 5;11(1):3238
pubmed: 33547334
Pharmacogenomics J. 2015 Jun;15(3):284-7
pubmed: 25287071
Acta Biomed. 2021 May 12;92(2):e2021102
pubmed: 33988147
PLoS One. 2022 Feb 18;17(2):e0263610
pubmed: 35180249
Int Rev Neurobiol. 2008;83:1-10
pubmed: 18929073
Indian J Clin Biochem. 2021 Oct;36(4):427-439
pubmed: 33814753
Vaccines (Basel). 2021 May 26;9(6):
pubmed: 34073382
J Allergy Clin Immunol Pract. 2020 Apr;8(4):1371-1386
pubmed: 32006723
BMJ. 2021 May 20;373:n1138
pubmed: 34016653
Bioorg Chem. 2021 Jan;106:104490
pubmed: 33261845
J Int Med Res. 2021 Mar;49(3):3000605211000157
pubmed: 33771068
BMC Cardiovasc Disord. 2021 Mar 9;21(1):129
pubmed: 33750304
Int J Health Serv. 2022 Jan;52(1):9-22
pubmed: 33686893
Health Policy. 2005 Mar;71(3):315-31
pubmed: 15694499
Int J Tuberc Lung Dis. 2021 Sep 1;25(9):701-707
pubmed: 34802491
Vaccine. 2018 Mar 20;36(13):1711-1719
pubmed: 29478755
Science. 2020 Nov 27;370(6520):1022
pubmed: 33243869