Computational Intelligence-Based Model for Mortality Rate Prediction in COVID-19 Patients.

Area of Science:

• Medical Informatics
• Artificial Intelligence in Healthcare
• Epidemiology

Background:

• The COVID-19 pandemic presents a significant global health challenge, with high mortality rates.
• Accurate early detection of severe COVID-19 cases is crucial for reducing mortality and complications.
• Machine Learning (ML) and Deep Learning (DL) show promise in disease diagnosis and prediction.

Purpose of the Study:

• To develop and evaluate ML and DL models for predicting COVID-19 patient mortality.
• To compare the performance of various ML algorithms against a proposed DL model.
• To identify the most effective model for early identification of severe COVID-19 cases.

Main Methods:

• Utilized ML algorithms: Decision Tree (DT), Logistic Regression (LR), Random Forest (RF), Extreme Gradient Boosting (XGBoost), and K-Nearest Neighbor (KNN).
• Developed a six-layer Deep Learning (DL) model with ReLU activation and a sigmoid output layer.
• Trained and tested models using data from confirmed COVID-19 patients across 146 countries with a reduced feature set.

Main Results:

• The proposed DL model achieved a high accuracy of 0.97 in predicting COVID-19 mortality.
• Comparative analysis demonstrated the superior performance of the DL model over traditional ML algorithms.
• The DL model significantly outperformed a baseline study using a reduced feature set.

Conclusions:

• The developed DL model is highly effective for predicting COVID-19 mortality rates.
• Early and accurate prediction of mortality can aid in clinical decision-making and resource allocation.
• This approach offers a significant advancement over existing methods for COVID-19 severity assessment.