Multisource Smart Computer-Aided System for Mining COVID-19 Infection Data.

Area of Science:

• Medical Imaging Analysis
• Artificial Intelligence in Healthcare
• Infectious Disease Diagnostics

Background:

• The COVID-19 pandemic necessitated rapid and accurate diagnostic tools.
• Multisource scan images like CT and X-ray are crucial for assessing respiratory system damage.
• Existing diagnostic methods require enhancement for efficiency and accuracy.

Purpose of the Study:

• To develop and validate a computer-aided diagnosis (CAD) system for detecting and diagnosing COVID-19 infections.
• To enhance the healthcare system's capacity during the pandemic through automated image analysis.
• To compare the performance of proposed models against existing approaches.

Main Methods:

• Utilized multisource scan images (CT and X-ray) for analysis.
• Developed a hybrid deep learning and shallow machine learning CAD system.
• Optimized hyperparameters using Support Vector Machines (SVM), deep learning with mini-batch stochastic gradient descent, and Naïve Bayes for parameter selection.

Main Results:

• The proposed CAD system demonstrated superior performance across multiple datasets.
• Achieved high accuracy rates, including 99.94% for binary and 100% for three-class labels on CT scans.
• Validated effectiveness through six experiments comparing hybrid and end-to-end deep learning models.

Conclusions:

• The developed CAD system effectively detects and diagnoses COVID-19 using CT and X-ray images.
• The hybrid deep learning approach offers a robust solution for respiratory system damage assessment.
• The system provides a valuable tool for healthcare providers in managing the COVID-19 pandemic.