RadientFusion-XR: A Hybrid LBP-HOG Model for COVID-19 Detection Using Machine Learning

Area of Science:

• Medical Imaging and Artificial Intelligence
• Radiology and Diagnostic Imaging

Background:

• Accurate COVID-19 detection from chest X-rays is crucial for timely patient management.
• Radiological image analysis presents challenges due to overlapping features between COVID-19, normal, and pneumonia cases.

Purpose of the Study:

• To investigate the efficacy of combining Local Binary Pattern (LBP) and Histogram of Oriented Gradients (HOG) features with machine learning for differentiating COVID-19 from normal and pneumonia chest X-rays.
• To introduce and evaluate the RadientFusion-XR hybrid fusion model for enhanced diagnostic accuracy.

Main Methods:

• Development of the RadientFusion-XR model, a hybrid fusion approach utilizing LBP and HOG features with shallow learning algorithms.
• Application of the model to chest X-ray images for classification of COVID-19, normal, and pneumonia cases.

Main Results:

• The RadientFusion-XR model achieved exceptional accuracy: 99% for binary classification (COVID-19 vs. normal) and 97% for multi-class classification (COVID-19, normal, pneumonia).
• The hybrid feature extraction and shallow learning approach significantly improved diagnostic accuracy in chest X-ray analysis.

Conclusions:

• The RadientFusion-XR model demonstrates a promising and efficient tool for early diagnosis of COVID-19 and pneumonia using chest X-rays.
• The model's interpretability and high performance make it a valuable asset for clinical applications and automated diagnostic systems.