Positive Rate of Serology and RT-PCR for COVID-19 among Community Residents and Healthcare Workers in Wuhan, China.

This study aimed to evaluate the infection rate of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) among different populations in Wuhan, China. This cross-sectional survey-based study examined the results of SARS-CoV-2-specific serological tests and RT-PCR tests for 4,454 community residents and 4,614 healthcare workers performed from May 15 to May 29, 2020. The healthcare workers were classified as administrative and logistical staff (n = 1,378), non-first-line healthcare workers (n = 2,630), or first-line healthcare workers (n = 606) according to their frequency of contact with coronavirus disease (COVID-19) patients. The positive rates of SARS-CoV-2-specific IgG, IgM, and RNA were 2.9%, 0.4%, and 0.1% for the community residents and 3.3%, 0.6%, and 0.2% for the healthcare workers, respectively. There were no statistically significant differences between the rates of the two groups. Spearman's correlation analysis showed that the frequency of contact with COVID-19 patients negatively correlated with the positive rates of RT-PCR (rs = -0.036, P = 0.016), but did not significantly correlate with the positive rates of IgM (rs = -0.006, P = 0.698) or IgG (rs = 0.017, P = 0.239). There was no statistically significant difference between the SARS-CoV-2-specific IgG, IgM, or RNA positive rates of the community residents and those of the healthcare workers. The positive rate of SARS-CoV-2 RNA was lower for the first-line healthcare workers than for the non-first-line healthcare workers and the administrative and logistical staff.

Improving Sentiment Classification of Restaurant Reviews with Attention-Based Bi-GRU Neural Network

In the era of Web 2.0, there is a huge amount of user-generated content, but the huge amount of unstructured data makes it difficult for merchants to provide personalized services and for users to extract information efficiently, so it is necessary to perform sentiment analysis for restaurant reviews. The significant advantage of Bi-GRU is the guaranteed symmetry of the hidden layer weight update, to take into account the context in online restaurant reviews and to obtain better results with fewer parameters, so we combined Word2vec, Bi-GRU, and Attention method to build a sentiment analysis model for online restaurant reviews. Restaurant reviews from Dianping.com were used to train and validate the model. With F1-score greater than 89%, we can conclude that the comprehensive performance of the Word2vec+Bi-GRU+Attention sentiment analysis model is better than the commonly used sentiment analysis models. We applied deep learning methods to review sentiment analysis in online food ordering platforms to improve the performance of sentiment analysis in the restaurant review domain.

Presence of SARS-CoV-2 RNA in isolation ward environment 28 days after exposure.

Recent studies have reported that surfaces and objects in the rooms of infected patients that are frequently touched by both medical staff and patients could be contaminated with SARS-CoV-2. In December 2019, Wuhan China suffered the earliest from this COVID-19 pandemic, and we took that opportunity to investigate whether the SARS-CoV-2 RNA exists in the ward environment after a long time from exposure. We found that on the 28th day following the discharge of COVID-19 patients, SARS-CoV-2 RNA could still be detected on the surfaces of pagers and in drawers in the isolation wards. Thorough disinfection of the ward environment was subsequently performed, after which these surfaces in the isolation wards tested negative for the presence of SARS-CoV-2 RNA. The findings remind us that the contaminated environment in the wards may become potential infectious resources and that despite a long time from exposure, the thorough disinfection in the COVID-10 units after is still necessary.