Metaheuristics based COVID-19 detection using medical images: A review

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Computational Biology

Background:

• The rapid global spread of COVID-19 necessitates efficient and cost-effective automated diagnostic tools to alleviate healthcare system burdens.
• Image classification using X-ray and CT scans is a key area of research for COVID-19 detection.
• Deep learning models are prominent for image classification but their performance is architecture-dependent.

Purpose of the Study:

• To review image classification techniques for chest imaging in COVID-19 detection.
• To investigate the application of metaheuristics in optimizing deep learning and machine learning models for COVID-19 diagnosis.
• To identify future research challenges in medical image-based COVID-19 detection.

Main Methods:

• Review of various image classification techniques for chest X-ray and CT images.
• Exploration of metaheuristic algorithms for optimizing deep neural network architectures.
• Analysis of metaheuristic applications in feature selection for machine learning models.

Main Results:

• Metaheuristics offer a flexible, simple, and problem-independent approach to optimizing complex non-linear problems like deep neural network architecture.
• The study emphasizes the potential of metaheuristics to enhance the performance of deep learning models in COVID-19 detection.
• A focus on metaheuristic applications for feature selection and model optimization is crucial for improving diagnostic accuracy.

Conclusions:

• Metaheuristics show significant promise for advancing automated COVID-19 detection systems through optimized deep learning and machine learning models.
• Further research into overlooked aspects of metaheuristic applications in medical image analysis is encouraged.
• Developing efficient and accessible diagnostic tools remains a critical objective in managing the COVID-19 pandemic.