CMM: A CNN-MLP Model for COVID-19 Lesion Segmentation and Severity Grading.

In this paper, a CNN-MLP model (CMM) is proposed for COVID-19 lesion segmentation and severity grading in CT images. The CMM starts by lung segmentation using UNet, and then segmenting the lesion from the lung region using a multi-scale deep supervised UNet (MDS-UNet), finally implementing the severity grading by a multi-layer preceptor (MLP). In MDS-UNet, shape prior information is fused with the input CT image to reduce the searching space of the potential segmentation outputs. The multi-scale input compensates for the loss of edge contour information in convolution operations. In order to enhance the learning of multiscale features, the multi-scale deep supervision extracts supervision signals from different upsampling points on the network. In addition, it is empirical that the lesion which has a whiter and denser appearance tends to be more severe in the COVID-19 CT image. So, the weighted mean gray-scale value (WMG) is proposed to depict this appearance, and together with the lung and lesion area to serve as input features for the severity grading in MLP. To improve the precision of lesion segmentation, a label refinement method based on the Frangi vessel filter is also proposed. Comparative experiments on COVID-19 public datasets show that our proposed CMM achieves high accuracy on COVID-19 lesion segmentation and severity grading. Source codes and datasets are available at our GitHub repository (https://github.com/RobotvisionLab/COVID-19-severity-grading.git).

FAM: focal attention module for lesion segmentation of COVID-19 CT images.

The novel coronavirus pneumonia (COVID-19) is the world's most serious public health crisis, posing a serious threat to public health. In clinical practice, automatic segmentation of the lesion from computed tomography (CT) images using deep learning methods provides an promising tool for identifying and diagnosing COVID-19. To improve the accuracy of image segmentation, an attention mechanism is adopted to highlight important features. However, existing attention methods are of weak performance or negative impact to the accuracy of convolutional neural networks (CNNs) due to various reasons (e.g. low contrast of the boundary between the lesion and the surrounding, the image noise). To address this issue, we propose a novel focal attention module (FAM) for lesion segmentation of CT images. FAM contains a channel attention module and a spatial attention module. In the spatial attention module, it first generates rough spatial attention, a shape prior of the lesion region obtained from the CT image using median filtering and distance transformation. The rough spatial attention is then input into two 7 × 7 convolution layers for correction, achieving refined spatial attention on the lesion region. FAM is individually integrated with six state-of-the-art segmentation networks (e.g. UNet, DeepLabV3+, etc.), and then we validated these six combinations on the public dataset including COVID-19 CT images. The results show that FAM improve the Dice Similarity Coefficient (DSC) of CNNs by 2%, and reduced the number of false negatives (FN) and false positives (FP) up to 17.6%, which are significantly higher than that using other attention modules such as CBAM and SENet. Furthermore, FAM significantly improve the convergence speed of the model training and achieve better real-time performance. The codes are available at GitHub (https://github.com/RobotvisionLab/FAM.git).

Release and Demand of Public Health Information in Social Media During the Outbreak of COVID-19 in China.

Information release is a key to the macro-economy during the outbreak of the Coronavirus Diosease-2019 (COVID-19). To explore the relationship between information supply by the government and public information demand in the pandemic, this study collected over 4,000 posts published on the most popular social media platform, i.e., WeChat. Many approaches, such as text mining, are employed to explore the information at different stages during the pandemic. According to the results, the government attached great importance to the information related to the pandemic. The main topics of information released by the government included the latest situation of the pandemic, announcements by the State Council, and prevention policies for COVID-19. Information mismatch between the public and Chinese governments contributed to the economic depression caused by the pandemic. Specifically, the topics of "the latest situation" and "popular scientific knowledge regarding the pandemic" have gained the most attention of the public. The information demand of the public has changed from the pandemic itself to the recovery of social life and industrial activities after the authority announced the control of the pandemic. However, during the recession phase, the information demand has shifted to asymptomatic infections and global pandemic trends. By contrast, some of the main topics provided by the government, such as "How beautiful you are," were excessive because the public demand is insufficient. Therefore, severe mismatches existed between information release of the government and public information demand during the pandemic, which impeded the recovery of the economy. The results in this study provide strategical suggestions of information release and opinion guidance for the authorities.

Influencing factors for public risk perception of COVID-19 --perspective of the pandemic whole life cycle.

The public's risk perception of public health emergencies will determine their behavior choices to a certain extent. Research on public risk perception of emergencies is an integral part of crisis management. From the perspective of the whole life cycle, this article takes the COVID-19 epidemic as an example. It conducts empirical analysis to study the influencing factors of public risk perception of public health emergencies. The results show that: (1) the public's risk perception is affected by individual factors, event characteristics, social influencing factors, and individual relationship factors. (2) The more the public is familiar with the epidemic, the lower the risk of the epidemic. The more the public can control the loss of the epidemic risk, the perceived epidemic risk will be reduced. The more the public trusts the supreme power of the government, the lower the risk of the epidemic in their hearts is. The higher the closeness of the risk and impact of the epidemic to individuals, the higher the level of risk perception is. (3)The public's risk perception will evolve with the development of the situation, and there are differences in recognition of government departments' control measures at different stages of public health emergencies. The relevant departments should effectively guide the public's risk response behavior in combination with the life cycle of public health emergencies. The research conclusions of this article clarify the dynamic evolution of risk perception and provide a specific reference for the emergency management of public health emergencies.

Clinical efficacy of methylprednisolone and the combined use of lopinavir/ritonavir with arbidol in treatment of coronavirus disease 2019.

This study aims to comparatively analyze the therapeutic efficacy upon multiple medication plans over lopinavir/ritonavir (LPV/r), arbidol (ARB), and methylprednisolone on patients with coronavirus disease 2019 (COVID-19). Totally, 75 COVID-19 patients admitted to The First Affiliated Hospital, Zhejiang University School of Medicine from January 22, 2020 to February 29, 2020 were recruited and grouped based on whether or not LPV/r and ARB were jointly used and whether or not methylprednisolone was used. Indexes including body temperature, time for nucleic acid negative conversion, hospital stays, and laboratory indexes were examined and compared. For all patients, there were no significant differences in the change of body temperature, the time for negative conversion, and hospital stays whether LPV/r and ARB were jointly used or not. While for severe and critically severe patients, methylprednisolone noticeably reduced the time for negative conversion. Meanwhile, the clinical efficacy was superior on patients receiving methylprednisolone within 3 days upon admission, and the duration of hospital stays was much shorter when methylprednisolone was given at a total dose of 0-400 mg than a higher dose of >400 mg if all patients received a similar dose per day. Nonetheless, no significant changes across hepatic, renal, and myocardial function indexes were observed. LPV/r combined with ARB produced no noticeably better effect on COVID-19 patients relative to the single-agent treatment. Additionally, methylprednisolone was efficient in severe and critically severe cases, and superior efficacy could be realized upon its early, appropriate, and short-term application.

Factors Associated With Prolonged Viral RNA Shedding in Patients with Coronavirus Disease 2019 (COVID-19).

BACKGROUND: An outbreak of coronavirus disease 2019 (COVID-19) is becoming a public health emergency. Data are limited on the duration and host factors related to viral shedding. METHODS: In this retrospective study, risk factors associated with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) RNA shedding were evaluated in a cohort of 113 symptomatic patients from 2 hospitals outside Wuhan. RESULTS: The median (interquartile range) duration of SARS-CoV-2 RNA detection was 17 (13-22) days as measured from illness onset. When comparing patients with early (<15 days) and late (≥15 days after illness onset) viral RNA clearance, prolonged SARS-CoV-2 RNA shedding was associated with male sex (P = .009), old age (P = .033), concomitant hypertension (P = .009), delayed admission to hospital after illness onset (P = .001), severe illness at admission (P = .049), invasive mechanical ventilation (P = .006), and corticosteroid treatment (P = .025). Patients with longer SARS-CoV-2 RNA shedding duration had slower recovery of body temperature (P < .001) and focal absorption on radiograph images (P < .001) than patients with early SARS-CoV-2 RNA clearance. Male sex (OR, 3.24; 95% CI, 1.31-8.02), delayed hospital admission (OR, 1.30; 95% CI, 1.10-1.54), and invasive mechanical ventilation (OR, 9.88; 95% CI, 1.11-88.02) were independent risk factors for prolonged SARS-CoV-2 RNA shedding. CONCLUSIONS: Male sex, delayed admission to hospital after illness onset, and invasive mechanical ventilation were associated with prolonged SARS-CoV-2 RNA shedding. Hospital admission and general treatments should be started as soon as possible in symptomatic COVID-19 patients, especially male patients.

Critically ill patients with COVID-19 with ECMO and artificial liver plasma exchange: A retrospective study.

COVID-19 is an emerging infectious disease capable of causing severe pneumonia. We aimed to characterize a group of critically ill patients in a single-center study.This was a retrospective case series of 23 patients with confirmed COVID-19-related critical illness in the intensive care unit (ICU) of a hospital in Hangzhou Zhejiang Province between January 22 and March 20, 2020.Of the 23 critically ill patients, the median age was 66 years (interquartile range [IQR] 59-80 years). The median time from disease onset to ICU admission was 10 days (IQR 6-11 days), to mechanical ventilation (MV) was 11 days (IQR 7.75-13 days), to artificial liver plasma exchange was 12 days (IQR 9.75-14.75 days), and to extracorporeal membrane oxygenation (ECMO) was 22 days (IQR 17.5-30 days). Nine patients required high flow oxygen. Fourteen patients received MV. Six required ECMO. Nine received artificial liver plasma exchange. Mortality was 0 at day 28.Mortality was 0 at day 28 in our single-center study. Extracorporeal membrane oxygenation reduced the requirements for ventilator support. Artificial liver plasma exchange significantly reduced inflammatory cytokine levels. These supportive therapies helped to extend the patients' survival times and increase the chance of follow-up treatments.

Epidemiological, clinical and virological characteristics of 74 cases of coronavirus-infected disease 2019 (COVID-19) with gastrointestinal symptoms.

OBJECTIVE: The SARS-CoV-2-infected disease (COVID-19) outbreak is a major threat to human beings. Previous studies mainly focused on Wuhan and typical symptoms. We analysed 74 confirmed COVID-19 cases with GI symptoms in the Zhejiang province to determine epidemiological, clinical and virological characteristics. DESIGN: COVID-19 hospital patients were admitted in the Zhejiang province from 17 January 2020 to 8 February 2020. Epidemiological, demographic, clinical, laboratory, management and outcome data of patients with GI symptoms were analysed using multivariate analysis for risk of severe/critical type. Bioinformatics were used to analyse features of SARS-CoV-2 from Zhejiang province. RESULTS: Among enrolled 651 patients, 74 (11.4%) presented with at least one GI symptom (nausea, vomiting or diarrhoea), average age of 46.14 years, 4-day incubation period and 10.8% had pre-existing liver disease. Of patients with COVID-19 with GI symptoms, 17 (22.97%) and 23 (31.08%) had severe/critical types and family clustering, respectively, significantly higher than those without GI symptoms, 47 (8.14%) and 118 (20.45%). Of patients with COVID-19 with GI symptoms, 29 (39.19%), 23 (31.08%), 8 (10.81%) and 16 (21.62%) had significantly higher rates of fever >38.5°C, fatigue, shortness of breath and headache, respectively. Low-dose glucocorticoids and antibiotics were administered to 14.86% and 41.89% of patients, respectively. Sputum production and increased lactate dehydrogenase/glucose levels were risk factors for severe/critical type. Bioinformatics showed sequence mutation of SARS-CoV-2 with m6A methylation and changed binding capacity with ACE2. CONCLUSION: We report COVID-19 cases with GI symptoms with novel features outside Wuhan. Attention to patients with COVID-19 with non-classic symptoms should increase to protect health providers.

Clinical findings in a group of patients infected with the 2019 novel coronavirus (SARS-Cov-2) outside of Wuhan, China: retrospective case series.

OBJECTIVE: To study the clinical characteristics of patients in Zhejiang province, China, infected with the 2019 severe acute respiratory syndrome coronavirus 2 (SARS-Cov-2) responsible for coronavirus disease 2019 (covid-2019). DESIGN: Retrospective case series. SETTING: Seven hospitals in Zhejiang province, China. PARTICIPANTS: 62 patients admitted to hospital with laboratory confirmed SARS-Cov-2 infection. Data were collected from 10 January 2020 to 26 January 2020. MAIN OUTCOME MEASURES: Clinical data, collected using a standardised case report form, such as temperature, history of exposure, incubation period. If information was not clear, the working group in Hangzhou contacted the doctor responsible for treating the patient for clarification. RESULTS: Of the 62 patients studied (median age 41 years), only one was admitted to an intensive care unit, and no patients died during the study. According to research, none of the infected patients in Zhejiang province were ever exposed to the Huanan seafood market, the original source of the virus; all studied cases were infected by human to human transmission. The most common symptoms at onset of illness were fever in 48 (77%) patients, cough in 50 (81%), expectoration in 35 (56%), headache in 21 (34%), myalgia or fatigue in 32 (52%), diarrhoea in 3 (8%), and haemoptysis in 2 (3%). Only two patients (3%) developed shortness of breath on admission. The median time from exposure to onset of illness was 4 days (interquartile range 3-5 days), and from onset of symptoms to first hospital admission was 2 (1-4) days. CONCLUSION: As of early February 2020, compared with patients initially infected with SARS-Cov-2 in Wuhan, the symptoms of patients in Zhejiang province are relatively mild.