SCOAT-Net: A Novel Network for Segmenting COVID-19 Lung Opacification from CT Images

The new coronavirus disease 2019 (COVID19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has spread worldwide at a rapid rate. There is no clinically automated tool to quantify the infection of COVID-19 patients. Automatic segmentation of lung opacification from computed tomography (CT) images provides excellent potential, which is of great significance for judging the disease development and treatment response of the patients. However, the segmentation of lung opacification from CT slices still faces some challenges, including the complexity and variability features of the opacity regions, the small difference between the infected and healthy tissues, and the noise of CT images. Besides, due to the limited medical resources, it is impractical to obtain a large amount of data in a short time, which further hinders the training of deep learning models. To answer these challenges, we proposed a novel spatial and channel-wise coarse-to-fine attention network (CAT-Net) inspired by the biological vision mechanism, which is for the segmentation of COVID-19 lung opacification from CT Images. CAT-Net has the spatial-wise attention module and the channel-wise attention module to attract the self-attention learning of the network, which serves to extract the practical features at the pixel and channel level successfully. Experiments show that our proposed CAT-Net achieves better results compared to state-of-the-art image segmentation networks.