A COMPREHENSIVE REVIEW OF MACHINE LEARNING MODELS FOR COVID-19 PREDICTION

In recent years, the Coronavirus Disease (COVID-19) has emerged as a global public health emergency, spreading rapidly across countries and placing unprecedented pressure on healthcare systems worldwide. The sudden surge in infected individuals has highlighted significant limitations in traditional diagnostic and screening methods, particularly in terms of testing time, cost, and availability of medical resources. As a result, there is a growing need for intelligent, fast, and cost-effective computational approaches to support early diagnosis and clinical decision-making. Machine learning has gained considerable attention as an effective tool for medical data analysis due to its ability to learn patterns from large and complex datasets. With successful applications in healthcare, biosciences, finance, and security, machine learning techniques have demonstrated strong potential in disease detection and prediction tasks. By analyzing clinical features, laboratory test results, and patient health records, machine learning models can assist in identifying COVID-19 cases at an early stage and reduce dependency on time-consuming diagnostic procedures. This paper presents a comprehensive survey of existing machine learning models used for the prediction of COVID-19 in patients. The study reviews commonly employed supervised learning and classification algorithms, including their working principles, advantages, and limitations in the context of COVID-19 diagnosis. Furthermore, a comparative performance analysis of selected machine learning techniques is conducted based on evaluation metrics such as accuracy, precision, recall, and classification efficiency. The survey highlights key challenges such as class imbalance, data quality issues, and model generalization in real-world clinical environments. The findings of this review suggest that machine learning–based predictive systems can significantly reduce diagnostic delays and support healthcare professionals in delivering timely and effective medical treatment. By integrating machine learning techniques into clinical workflows, healthcare systems can enhance early detection capabilities, optimize resource utilization, and improve overall patient outcomes. This study aims to serve as a useful reference for researchers and practitioners working on intelligent healthcare solutions for COVID-19 and related infectious diseases.