Metaheuristic-based Deep COVID-19 Screening Model from Chest X-Ray Images.

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Computational Biology

Background:

• The COVID-19 pandemic necessitates rapid and accurate diagnostic tools.
• Deep learning models show promise for early COVID-19 detection from X-ray images.
• Existing deep learning models face challenges with overfitting and hyperparameter optimization.

Purpose of the Study:

• To propose a metaheuristic-based deep learning model for enhanced COVID-19 screening using X-ray images.
• To address overfitting and hyperparameter-tuning issues in deep learning models for COVID-19 diagnosis.
• To evaluate the performance of the proposed model on a multi-class dataset.

Main Methods:

• A modified AlexNet architecture was employed for feature extraction and classification.
• The Strength Pareto Evolutionary Algorithm II (SPEA-II) was utilized for hyperparameter tuning of AlexNet.
• The model was trained and validated on a four-class dataset including COVID-19, tuberculosis, pneumonia, and healthy individuals.

Main Results:

• The proposed metaheuristic-optimized deep learning model demonstrated effectiveness in screening COVID-19 from X-ray images.
• Hyperparameter optimization using SPEA-II improved the performance of the modified AlexNet architecture.
• Comparative analysis indicated the potential advantages of the proposed model over existing methods.

Conclusions:

• Metaheuristic optimization offers a viable solution to enhance deep learning model performance for COVID-19 detection.
• The developed model shows promise for accurate and efficient early screening of respiratory diseases, including COVID-19.
• Further validation on diverse datasets is recommended to solidify clinical applicability.