COVID-19 Detection from Chest X-Ray Images Using GBM with Comparative Analysis

Effective disease management and control depend on the quick and precise diagnosis of COVID-19. Chest X-ray(CXR) imaging has gained prominence as a convenient and accessible diagnostic tool. This study presents a Gradient Boosting Machines(GBM) and Convolutional Neural Networks (CNNs)-based model of COVID-19 detection, along with a comprehensive comparative analysis of the performances of various other machine learning (ML) algorithms. We compiled collected and pre-processed datasets comprising CXR of healthy people and those with COVID-19 positive. The model was trained with diverse CNN architectures and hyperparameter tuning optimized their performance. Algorithm effectiveness was assessed using evaluation metrics including accuracy, precision, recall and F1-score. Experimental results revealed the strengths and weaknesses of each algorithm in COVID-19 detection and the benefit of using these algorithms. The CNN and GBM model successfully differentiate between COVID-19-positive cases and healthy people with a remarkable accuracy rate of 97.4% but GBM take less computation time as compared to CNN.