Covid-19 detection from radiographs by feature-reinforced ensemble learning.

Area of Science:

• Computer Science
• Medical Imaging
• Artificial Intelligence

Background:

• The COVID-19 pandemic necessitates rapid and accurate diagnostic tools.
• Artificial intelligence (AI) shows promise in analyzing medical images for disease detection.

Purpose of the Study:

• To develop and evaluate ensemble learning models for SARS-CoV-2 detection using chest X-ray images.
• To enhance the classification performance of the Residual Convolutional Neural Network (ResCNN) by integrating machine learning algorithms.

Main Methods:

• Two ensemble learning models were proposed, combining deep learning (ResCNN) with machine learning algorithms.
• Models were trained and validated on 5228 chest X-ray images (Normal, Pneumonia, COVID-19).
• Image datasets were processed at various resolutions (32x32 to 256x256) and evaluated using 10-fold cross-validation.

Main Results:

• Both proposed ensemble models demonstrated superior classification ability compared to the standalone ResCNN.
• The second model achieved comparable classification scores with significantly reduced image resolution (32x32), indicating lower computational requirements.
• The ensemble approach improved diagnostic accuracy for COVID-19 detection from X-rays.

Conclusions:

• Ensemble learning models integrating ResCNN with machine learning algorithms offer an effective approach for COVID-19 detection from X-ray images.
• The proposed methods provide a computationally efficient alternative for AI-driven medical image analysis in pandemic scenarios.