Analysis of disease severity and mortality prediction using machine learning during COVID-19

Area of Science:

• Medical Informatics
• Computational Biology
• Epidemiology

Background:

• Traditional statistical methods for predicting COVID-19 progression have limitations.
• Challenges in disease containment include high costs and time-consuming medical tests.
• Machine learning (ML) offers advanced solutions for medical data analysis and prediction.

Purpose of the Study:

• To analyze the application of ML algorithms in predicting COVID-19 disease severity and mortality.
• To identify key features influencing disease outcomes.
• To comparatively evaluate different ML models for COVID-19 diagnosis and prognosis.

Main Methods:

• Preprocessing of a large COVID-19 patient dataset (>1 million records).
• Comparative analysis of over 12 machine learning models for disease classification.
• Evaluation of model accuracy for predicting disease severity (class 0) and mortality risk (class 1).

Main Results:

• The Logistic Regression model achieved the highest accuracy in predicting both disease severity (97%) and mortality risk (80%).
• Key features influencing patient outcomes were identified through model analysis.
• Comparative evaluation highlighted the superior performance of Logistic Regression over other models.

Conclusions:

• Machine learning, particularly Logistic Regression, provides accurate tools for COVID-19 severity and mortality prediction.
• Findings support improved healthcare resource allocation and patient survival prediction.
• This study validates ML's role in enhancing pandemic response strategies.