A CNN based coronavirus disease prediction system for chest X-rays

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Radiology

Background:

• The global spread of COVID-19 necessitates advanced diagnostic tools.
• Chest X-rays are vital for COVID-19 staging and diagnosis.
• Challenges include image variations and low contrast, complicating radiological report generation.

Purpose of the Study:

• To develop an automated system for COVID-19 prediction using chest X-ray images.
• To address limitations in existing diagnostic methods for COVID-19.

Main Methods:

• Introduction of CODISC-CNN (CNN based Coronavirus DIsease Prediction System for Chest X-rays).
• Application of image preprocessing for edge detection to identify infected regions.
• Utilizing data augmentation to overcome limited labeled datasets.

Main Results:

• Achieved 97% accuracy in classifying Normal vs. COVID-19.
• Reached 89% accuracy for Normal, COVID-19, and Virus Bacteria classification.
• Attained 84% accuracy for Normal, COVID-19, Virus Bacteria, and Virus Pneumonia classification.
• The model demonstrated superior performance compared to existing classification models.

Conclusions:

• CODISC-CNN effectively extracts features for automated COVID-19 prediction from chest X-rays.
• The proposed model shows significant potential in improving healthcare strategies for pandemic management.
• The system enhances state-of-the-art performance in medical image analysis for infectious diseases.