Supervised Machine Learning Models for Prediction of COVID-19 Infection using Epidemiology Dataset

Area of Science:

• Medical Informatics
• Computational Biology
• Epidemiology

Background:

• COVID-19 (2019-nCoV) is now endemic, posing an ongoing challenge to global healthcare systems due to the lack of specific treatments or cures.
• The absence of effective antivirals or vaccines necessitates alternative strategies to manage the disease and alleviate healthcare burdens, particularly in developing nations.

Purpose of the Study:

• To develop and evaluate supervised machine learning models for diagnosing COVID-19 (2019-nCoV) infection.
• To assess the performance of various machine learning algorithms in classifying COVID-19 cases using epidemiological data.

Main Methods:

• Employed supervised machine learning algorithms including logistic regression, decision tree, support vector machine, naive Bayes, and artificial neural network.
• Utilized an epidemiology-labeled dataset of positive and negative COVID-19 cases from Mexico, with 80% for training and 20% for testing.
• Performed correlation coefficient analysis to understand feature relationships before model development.

Main Results:

• The decision tree model demonstrated the highest accuracy at 94.99%.
• The Support Vector Machine model achieved the highest sensitivity (93.34%).
• The Naïve Bayes model exhibited the highest specificity (94.30%).

Conclusions:

• Supervised machine learning models offer a viable approach for COVID-19 (2019-nCoV) diagnosis and prognosis.
• These AI-driven methods can help reduce the strain on healthcare systems and economic sectors impacted by the endemic disease.