Using Machine Learning Models to Evaluate the Need for COVID-۱۹ Vaccination

Background and aims: In the wake of the ongoing COVID-۱۹ pandemic, Artificial Intelligence(AI) is gaining much attention, and one of its practical fields is Machine Learning. While todayvaccine adherence is high, there was a time at the start of the COVID-۱۹ pandemic when manypeople did not trust vaccines and believed that once they were infected with COVID-۱۹, there wasno need for vaccination. Still, there is evidence stating that COVID-۱۹ antibodies will not staypositive permanently, and there is a vital need for a booster vaccine. This study aims to developa pilot model using machine learning methods in order to predict if unvaccinated patients’ serumIgG antibodies are sufficient or if there is a need for a vaccine without a laboratory test.Method: This study used symptoms and demographic data of ۲۰۶ confirmed COVID-۱۹ patientswhose COVID-۱۹-specific serum IgG was measured, and months passed since COVID-۱۹ infectionwas recorded and added as a variable. Data was gathered from January to October ۲۰۲۱,before vaccination initiation in Iran. Data were preprocessed and cleaned, important features wereselected, and serum IgG amount was transformed into a binary variable based on the ۱.۲۰ cutoff.This variable was later used to be predicated. Data were randomly split into train and test groupswith proportions of ۲۰% and ۸۰%; ۵-fold cross-validation using models including Random Forest,Support Vector Machine, Neural Network, Naïve Bayes, and XGBoost was conducted, andthey were evaluated and compared; the one with the best results was selected. Models were deployedin R Studio software using packages including “randomForest”, “caret”, “e۱۰۷۱”, “neuralnet”,“naivebayes”, and “xgboost”, and evaluation metrics were recorded. The model was laterexported and uploaded to the GitHub repository for analysis reuse.Results: The train and test set included ۱۶۲ and ۴۴ samples, respectively. Features that had beenselected included Gender, Age, Hospitalization, time that had passed since infection, urban orrural living area, education level, occupation, chronic disease, fever, headache, cough, malaise,restlessness, sore throat, bone pain, conjunctivitis, anosmia, loss of taste sense, sweating, nausea,vomiting, stomachache, diarrhea, chest pain, dyspnea, history of covid infection in family members,and disease severity. The reported accuracy for the Random Forest, SVM, Decision Tree,Neural Network, Naïve Bayes, and XGBoost were ۰.۸۴۰۹, ۰.۷۹۵۵, ۰.۶۸۱۸, ۰.۷۰۴۵, ۰.۶۸۱۸ and۰.۷۷۲۷, respectively. The Random Forest model ۵-fold with its default settings (number of trees۵۰۰, features per split ۴) reported accuracy was ۰.۸۴۰۹ (۹۵% CI of ۰.۶۹۹۳, ۰.۹۳۳۶); model sensitivityand specificity were ۰.۷۶۹۲ and ۰.۸۷۱۰, respectively. The recall was ۰.۷۱۴۳, a negativepredictive value of ۰.۹۰۰۰ was reported, and a ROC plot was drawn.Conclusion: The Random Forest model showed satisfactory and exciting results, such as theimportance of occupation in the longevity of COVID-۱۹ serum sufficient presence. A model wasprovided for predicting the need for vaccination in unvaccinated individuals infected with COVID-۱۹; this study may serve as a stepping stone toward determining if vaccine booster doses needto be administered based on the time since the last vaccination or infection.

کلیدواژه ها:

Machine Learning ، COVID-۱۹ ، Immunoglobulin G ، Vaccine