Comprehensive Survey of Machine Learning Systems for COVID-19 Detection

Area of Science:

• Medical Imaging and Artificial Intelligence
• Infectious Disease Diagnostics

Background:

• The COVID-19 pandemic necessitated rapid and accurate diagnostic methods.
• Existing clinical diagnosis methods for COVID-19 present limitations and challenges.
• Automated detection systems are crucial for timely intervention and preventing virus propagation.

Purpose of the Study:

• To conduct a comprehensive review of Artificial Intelligence (AI)-based solutions for COVID-19 detection.
• To analyze AI mechanisms utilizing chest medical images for diagnosis.
• To summarize the advantages and shortcomings of proposed AI detection methods.

Main Methods:

• Systematic review and analysis of over 200 research papers.
• In-depth examination of 145 articles focusing on AI mechanisms for COVID-19 detection.
• Evaluation of machine learning applications in COVID-19 detection, segmentation, and classification.

Main Results:

• Identified and analyzed various AI-based approaches for automated COVID-19 diagnosis from chest images.
• Detailed examination of the strengths and weaknesses associated with different AI methodologies.
• Synthesis of findings from extensive literature on machine learning for COVID-19.

Conclusions:

• AI demonstrates significant potential in enhancing the accuracy and speed of COVID-19 diagnosis.
• Machine learning techniques are pivotal for developing effective automated detection, segmentation, and classification tools.
• Further research is warranted to address the limitations and optimize AI performance in clinical settings.