Computer-aided diagnosis of COVID-19 from chest X-ray images using histogram-oriented gradient features and Random...

Area of Science:

• Medical Imaging
• Computer-Aided Diagnosis
• Machine Learning in Healthcare

Background:

• Accurate and rapid diagnosis of COVID-19 is critical for pandemic control.
• Manual interpretation of chest X-rays for COVID-19 is challenging due to image similarities with other respiratory diseases.
• Automated diagnostic tools are needed to assist healthcare professionals.

Purpose of the Study:

• To develop an automated system for COVID-19 detection using X-ray images.
• To evaluate the effectiveness of Histogram Oriented Gradient (HOG) features combined with an optimized Random Forest (RF) classifier.
• To compare HOG with other feature extraction techniques and RF with various classification algorithms.

Main Methods:

• Feature extraction using Histogram Oriented Gradient (HOG), Gray-Level Co-Occurrence Matrix (GLCM), and Hu moments.
• Training and optimization of a Random Forest (RF) classifier.
• Comparative analysis of RF against Linear Regression (LR), Linear Discriminant Analysis (LDA), K-nearest neighbor (kNN), Classification and Regression Trees (CART), Support Vector Machine (SVM), and Multi-layer perceptron neural network (MLP).

Main Results:

• HOG features effectively capture local edge descriptions and structural information for discriminating COVID-19.
• The optimized RF classifier trained with HOG features achieved a highest classification accuracy of 99.73%.
• HOG outperformed GLCM and Hu moments in feature extraction for COVID-19 diagnosis.

Conclusions:

• The proposed HOG feature extraction with RF classifier offers a highly accurate and efficient method for automated COVID-19 diagnosis from X-ray images.
• This approach can significantly assist radiologists and physicians in clinical decision-making.
• The study highlights the potential of machine learning in improving diagnostic speed and accuracy for infectious diseases.