COVID-19 CT Image Classification and Pneumonia Lesions Segmentation Using Deep Learning

Abstract

In the clinical diagnosis and treatment of COVID-19, the classification of chest CT images and the accurate segmentation of novel coronavirus pneumonia (COVID-19) lesions in chest CT images are of great importance in diagnosing and assessing the severity of patients. In this paper, we used a deep learning model for classification and lesion segmentation of chest CT images, and in our experiments, we achieved better results with the approach we used. The results show that the use of deep learning is effective as an important indicator to assist in the assessment of whether one is infected with COVID-19.