COVID-19 detection using deep learning models to exploit Social Mimic Optimization and structured chest X-ray images...

Area of Science:

• Biomedical imaging
• Artificial Intelligence in Medicine
• Computational Virology

Background:

• Coronaviruses are RNA viruses causing respiratory infections, including pneumonia, in humans and animals.
• Artificial intelligence (AI) shows promise for advanced biomedical data analysis.
• Accurate and early detection of coronavirus infections is crucial for public health.

Purpose of the Study:

• To develop and evaluate a deep learning-based model for detecting coronavirus infections using X-ray imagery.
• To investigate the effectiveness of AI in differentiating between coronavirus, pneumonia, and normal X-ray patterns.
• To enhance the accuracy of AI-driven diagnostic tools for respiratory diseases.

Main Methods:

• Utilized a dataset comprising X-ray images categorized as coronavirus, pneumonia, or normal.
• Applied the Fuzzy Color technique for data restructuring and image stacking as a preprocessing step.
• Trained deep learning models (MobileNetV2, SqueezeNet) and optimized feature sets using the Social Mimic algorithm.
• Classified features using Support Vector Machines (SVM) for final diagnosis.

Main Results:

• The proposed AI approach achieved an overall classification accuracy of 99.27%.
• The model demonstrated high efficacy in distinguishing between coronavirus, pneumonia, and normal chest X-ray images.
• Feature optimization using the Social Mimic method improved classification performance.

Conclusions:

• The developed deep learning model shows significant potential for efficient and accurate detection of coronavirus infections.
• AI-powered analysis of X-ray imagery can serve as a valuable tool in diagnosing COVID-19.
• This approach offers a promising avenue for improving diagnostic capabilities in respiratory disease management.