LBP-based information assisted intelligent system for COVID-19 identification.

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Radiology

Background:

• Real-time COVID-19 detection is crucial for managing the pandemic.
• Chest X-rays offer a potential imaging modality for COVID-19 diagnosis.
• Integrating AI with medical imaging can enhance diagnostic accuracy and efficiency.

Purpose of the Study:

• To develop an automated COVID-19 detection system using chest X-ray images.
• To investigate the efficacy of combining textural and Local Binary Pattern (LBP) features.
• To create a robust model for identifying COVID-19 infections from X-ray data.

Main Methods:

• Extraction of textural features from chest X-ray images.
• Generation of Local Binary Pattern (LBP) images and feature extraction.
• Joint investigation and selection of highly discriminatory features for classification.
• Training and validation of an automated model on a dataset of 2905 X-ray images.

Main Results:

• The developed method achieved 97.97% accuracy and 99.88% sensitivity for classifying COVID-19 vs. pneumonia and normal X-rays.
• For COVID-19 vs. normal X-ray classification, the system attained 98.91% accuracy and 99.33% sensitivity.
• The approach demonstrated robustness across various class combinations.

Conclusions:

• The proposed automated system effectively identifies COVID-19 from chest X-ray images.
• The combination of textural and LBP features enhances diagnostic performance.
• This AI-driven tool can assist radiologists in mass screening for rapid, accurate, and contact-free COVID-19 diagnosis.