COVID-19 Data Analytics Using Extended Convolutional Technique.

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Infectious Diseases

Background:

• The COVID-19 pandemic significantly impacted global health systems, economies, and daily life.
• Early detection of COVID-19 remains challenging, with current methods like RT-PCR having limitations.
• There is a need for faster, more accurate diagnostic tools for COVID-19.

Purpose of the Study:

• To develop and evaluate a deep learning-based method for detecting COVID-19 using chest X-ray images.
• To improve the speed and accuracy of COVID-19 diagnosis.

Main Methods:

• Utilized deep learning models (GoogleNet, U-Net, ResNet50) integrated for analyzing chest X-ray images.
• Implemented a preprocessing phase involving lung segmentation and noise reduction.
• Employed transfer learning for model training and heatmap visualization for result interpretation.

Main Results:

• The proposed deep learning method achieved a COVID-19 detection accuracy of approximately 99%.
• The technique effectively categorizes patients as COVID-19 positive or negative based on X-ray imaging.

Conclusions:

• Deep learning models applied to chest X-rays show high potential for accurate and rapid COVID-19 diagnosis.
• This approach offers a promising alternative or supplement to existing diagnostic methods.