Machine Learning Algorithms Show Promise in Predicting COVID-19 Outcomes

Strategic Insights -

Researchers at the K.N. Toosi University of Technology in Iran have utilized machine learning algorithms to analyze disease severity and mortality prediction in COVID-19 patients. The study found that the Logistic Regression model exhibited the highest accuracy for both class 0 (97%) and class 1 (80%) in diagnosing COVID-19 patients. This breakthrough could aid healthcare centers in prioritizing medical resource allocation based on patient condition and predict chances of survival.

Key Takeaways:

  • The study utilized public and clinical data from over one million patients to evaluate the effectiveness of machine learning algorithms in predicting COVID-19 outcomes.
  • The Logistic Regression model demonstrated the highest accuracy among 12 models tested, achieving 97% accuracy for class 0 (no COVID-19) and 80% accuracy for class 1 (COVID-19).
  • The study's findings suggest that machine learning algorithms can be used to tackle the challenges of COVID-19, including limitations, high costs, and time requirements for medical tests.
  • The research was conducted using public and clinical data from patients, identifying the most important features in predicting COVID-19 severity and mortality.
  • The study provides a comparative analysis of different models for diagnosis using statistical data.

Statistics:

  • Over one million patients were included in the study, providing a comprehensive dataset for analysis.
  • The Logistic Regression model achieved 97% accuracy for class 0 (no COVID-19) and 80% accuracy for class 1 (COVID-19).
  • The study evaluated 12 machine learning models, with the Logistic Regression model demonstrating the highest accuracy.
  • The study's findings have significant implications for healthcare centers, enabling them to prioritize medical resource allocation based on patient condition and predict chances of survival.

Sources:

  • "Analysis of disease severity and mortality prediction using machine learning during COVID-19" by Hodjat Hojatollah Hamidi, et al. Acta Psychologica, 2025;258:105136.
  • NewsRx. K.N. Toosi University of Technology Reports Findings in COVID-19 (Analysis of disease severity and mortality prediction using machine learning during COVID-19). Journal of Engineering. June 30, 2025; p 2.