ABSTRACT
Since computed tomography (CT) provides the most sensitive radiological technique for diagnosing COVID-19, CT has been used as an efficient and necessary aided diagnosis. However, the size and number of publicly available COVID-19 imaging datasets are limited and have problems such as low data volume, easy overfitting for training, and significant differences in the characteristics of lesions at different scales. Our work presents an image segmentation network, Pyramid-and-GAN-UNet (PGUNet), to support the segmentation of COVID-19 lesions by combining feature pyramid and generative adversarial network (GAN). Using GAN, the segmentation network can learn more abundant high-level features and increase the generalization ability. The module of the feature pyramid is used to solve the differences between image features at different levels. Compared with the current mainstream method, our experimental results show that the proposed network achieved more competitive performances on the CT slice datasets of the COVID-19 CT Segmentation dataset and CC-CCII dataset. © 2022 IEEE.
ABSTRACT
OBJECTIVES: To understand the sequelae of COVID-19. METHODS: We followed up 1174 patients with severe coronavirus disease 2019 (COVID-19)who were recovered and discharged for 6 months. RESULTS: There were 175 cases with clear IgG results 6 months after discharge, of which 82 (46.9%) were IgG (+) and 16 (9.1%) were IgG (dim+). Four hundred and forty-one participants (55.4%) had some kind of sequelae. The most common symptoms were fatigue (25.3%), sleep disorder (23.2%) and shortness of breath (20.4%). In those who had sequelae, 262 (59.4%) had more than one symptom. Critical cases were more likely to have cough (20.5% vs 11.6%, p = 0.023) and hypomnesis (15.1% vs 8.0%, p = 0.041) than severe cases. Furthermore, univariate and multivariate logistic regression analyses revealed that women are more likely to have multiple symptoms (p = 0.002), fatigue (p = 0.009) and sleep disorder (p = 0.008), whereas critical illness was found as independent risk factor for hypomnesis (p = 0.045). CONCLUSION: Our study demonstrated the duration of antibody and sequelae of COVID-19 and compared the differences amongst different populations.