Multiclass Classification for Detection of COVID-19 Infection in Chest X-Rays Using CNN.

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Infectious Diseases

Background:

• Coronavirus (COVID-19) presents diagnostic challenges due to its novelty and high infectivity.
• Reverse Transcription Polymerase Chain Reaction (RT-PCR) is the standard COVID-19 test but is costly, invasive, and prone to errors.
• Radiographic imaging, specifically chest X-rays, offers a viable alternative for COVID-19 detection, leveraging radiologist expertise.

Purpose of the Study:

• To develop and evaluate a deep learning model for accurate COVID-19 detection using chest X-ray images.
• To investigate the potential of Convolutional Neural Networks (CNNs) in identifying COVID-19 patterns in radiographs.
• To provide an efficient and accurate diagnostic tool for COVID-19, complementing existing methods.

Main Methods:

• A deep learning model utilizing Convolutional Neural Networks (CNNs) was developed.
• The CNN model was trained on a large dataset of over 35,000 chest X-ray images.
• The dataset included images categorized as COVID-19 positive, normal, and pneumonia positive.

Main Results:

• The proposed deep learning model achieved a high accuracy of 99%.
• Performance metrics included a precision of 0.98, recall of 1.02, and an F1-score of 99.0%.
• The model demonstrated superior performance compared to other deep learning models in existing studies.

Conclusions:

• Deep learning, specifically CNNs, shows significant potential for accurate COVID-19 detection from chest X-rays.
• The developed model offers a highly accurate and efficient diagnostic approach, potentially reducing reliance on RT-PCR.
• This AI-driven method can aid in rapid screening and diagnosis of COVID-19 in clinical settings.