Chapter 10 - Early detection of COVID-19 pneumonia based on ground-glass opacity (GGO) features of computerized tomography (CT) angiography

Abstract

The World Health Organization declared COVID-19 as a global pandemic in March 2020. As of January 2021, more than 96.2 million confirmed cases have been recorded, causing catastrophic damage around the world. To diagnose COVID-19 patients, PCR testing takes time, and the test findings tend to be of low accuracy relative to computerized tomography (CT) testing. The aim of this study is to develop a system for early detection of COVID-19 pneumonia using machine intelligence with CT angiography. Three forms of CT scan images from the Union Hospital and Liyuan Hospital were used as the dataset. We randomly collected 27% of positive CT image samples and 11% of negative CT image samples from the original dataset. The proposed system consists of a convolutional neural network and cascade classifier model for identifying ground-glass opacity (GGO) regions in the lungs. The proposed machine learning algorithm for COVID-19 pneumonia detection achieved 92.8% accuracy at 0.931 precision.