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 (SCOAT-Net) inspired by the biological vision mechanism, which is for the segmentation of COVID-19 lung opacification from CT Images. SCOAT-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 SCOAT-Net achieves better results compared to state-of-the-art image segmentation networks.

Chest CT findings and clinical features of coronavirus disease 2019 in children / 中南大学学报(医学版)

Objective: To describe the CT features and clinical characteristics of pediatric patients with coronavirus disease 2019 (COVID-19).Method: A total of 9 COVID-19 infected pediatric patients were included in this study.Clinical history,laboratory examination,and detailed CT imaging features were analyzed.All patients underwent the first CT scanning on the same day of being diagnosed by realtime reverse-transcription polymerase chain reaction (rRT-PCR).A low-dose CT scan was performed during follow-up.Results: All the child patients had positive results.Four patients had cough and one patient had fever.One patient presented both cough and fever.Two children presented other symptoms like sore throat and stuffy nose.One child showed no clinical symptom.Five patients had positive initial CT findings with subtle lesions like ground-glass opacity (GGO) or spot-like mixed consolidation.Three patients were reported with negative results in the initial and follow-up CT examination.One patient was reported with initial negative CT findings but turning positive during the first follow-up.All patients had absorbed lesions on follow-up CT images after treatment.Conclusion: Pediatric COVID-19 patients have certain imaging and clinical features as well as disease prognosis.Children with COVID-19 tend to have normal or subtle CT findings and relatively better outcome.