Analysis of CT scan medical images to diagnose COVID-19 disease by using machine learning and Image enhancement algorithms

Abstract

The study presents a novel Artificial Intelligence (AI) system created to distinguish between COVID-19, normal lung states, and other lung illnesses by analysing Computed Tomography (CT) scans. The research uses sophisticated machine learning algorithms to pinpoint distinct characteristics and trends in CT images to achieve precise classification. The AI system's performance is thoroughly assessed using important metrics to tackle the issue of false negatives in early COVID-19 detection in comparison to Polymerase Chain Reaction (PCR) assays. Deep learning is crucial for improving the accuracy and efficiency of diagnostics. This study delves into the system's processing and analysis of CT images, emphasising the significance of internal and external validation, and the use of primary algorithms for feature extraction and categorization. The study's contributions to early COVID-19 detection are highlighted through a comparative analysis with existing AI-assisted diagnostic techniques, demonstrating potential savings in labour and time requirements. The article discusses the importance of high-resolution imaging in CT scans for enhanced diagnostic accuracy and explores the potential of AI in medical imaging for respiratory diseases, underscoring the study's novelty and objectives.