An Improved Lightweight YOLOv5 Model Based on Attention Mechanism for Face Mask Detection

Coronavirus 2019 has brought severe challenges to social stability and public health worldwide. One effective way of curbing the epidemic is to require people to wear masks in public places and monitor their mask-wearing states by suitable automatic detectors. However, existing models struggle to simultaneously achieve the requirements of both high precision and real-time performance. To solve this problem, we propose an improved lightweight face mask detector based on YOLOv5, which can achieve an excellent balance of precision and speed. Firstly, a novel backbone ShuffleCANet that combines ShuffleNetV2 network with Coordinate Attention mechanism is proposed as the backbone. Afterward, an efficient path aggression network BiFPN is applied as the feature fusion neck. Furthermore, the localization loss is replaced with alpha-CIoU in model training phase to obtain higher-quality anchors. Some valuable strategies such as data augmentation, adaptive image scaling, and anchor cluster operation are also utilized. Experimental results on AIZOO face mask dataset show the superiority of the proposed model. Compared with the original YOLOv5, the proposed model increases the inference speed by 28.3% while still improving the precision by 0.58%. It achieves the best mean average precision of 95.2% compared with other seven existing models, which is 4.4% higher than the baseline.

[Consideration on implementation of co-administration of Seasonal Influenza and COVID-19 vaccines during pandemic in China].

Influenza is an infectious respiratory disease caused by the influenza viruses. Older people, infants and people with underlying medical conditions could have a higher risk of severe influenza symptoms and complications. The co-infection of Coronavirus Diseases 2019 (COVID-19) with influenza viruses could lead to the complication of prevention, diagnosis, control, treatment, and recovery of COVID-19. Influenza vaccine and COVID-19 vaccine overlapped in target populations, vaccination time, and inoculation units. Although there was insufficient evidence on the immunogenicity and safety of co-administration of influenza vaccine and COVID-19 vaccine, World Health Organization and some countries recommended co-administration of inactivated influenza vaccine and COVID-19 vaccine. This review summarized domestic and international vaccination policies and research progress, and put forward corresponding suggestions in order to provide scientific support for the formulation of vaccination strategy on seasonal influenza vaccine and COVID-19 vaccine.

Leakage-Suppressed Range Query Scheme for Structured Data in IoT

The structured data collected by the Internet of Things can be encrypted for protecting the user's privacy. Range query can be used to get the expected data with some specific attributes among the encrypted data, that is, given the upper and lower limits $(x, y)$ of a certain attribute, the range query will get all the records whose corresponding attribute values are in $(x, y)$. However, in the structured encryption scheme with range query, there is a certain amount of information leakage, which will lead to the so-called inference attacks, i.e., the attacker can obtain the user's attribute values. To hide attribute values and their relationship, this article transformed the problem of the numerical comparison between two attribute values into the problem of the intersection of two sets. By using the Bloom filter, the elements in the attribute value collection are recorded and determined whether the intersection of the two sets is empty. This ensures that our scheme effectively resists inference attacks. Besides, by multiplying the endpoints of the range interval by a large number, we gave an improved scheme to hide the user's search pattern. In the query process, our scheme will not leak the upper and lower limits of the range value and will not leak the relationship between the attribute value and the range interval. This will prevent the attacker from inducing the relationship of attribute values by multiple range queries. Finally, we conducted a simulation evaluation of the scheme by using the published novel coronavirus pneumonia data, and the results show that our scheme has a better performance than the existing schemes.

A systematic review of prevention and intervention strategies for smartphone addiction in students: Applicability during the covid-19 pandemic

During this coronavirus (COVID-19) pandemic, smartphones play an important role in online classes, study, and entertainment. However, excessive use may lead to smartphone addiction (SPA). The incidence of SPA among students has increased with the spread of COVID-19 and threatens to impair home-based students’ learning efficiency and physical and mental health. This study aimed to provide a comprehensive overview of the latest achievements in SPA prevention and treatment, and a theoretical basis for future experimental research and clinical treatment, while considering their applicability during the current pandemic. We researched the core literature in Chinese, English, and Korean databases from 2000 to 2021;3208 articles were identified. After reading the titles, s, and full texts, 53 articles were selected. Research on SPA interventions was relatively limited;we identified six types of prevention and treatment measures: psychotherapies, cognitive training, behavioral intervention, application restriction, social intervention, and complementary and alternative medicine. They can be implemented by students, parents, or online experts. Future research should focus on developing early measures to identify and prevent SPA and enhance students’ change motivation. © 2021, Cluj University Press. All rights reserved.

Factors related to public COVID-19 vaccine hesitancy based on the "3Cs" model: a cross-sectional study

Objective: To understand the current situation of the hesitation of COVID-19 vaccines among Chinese residents, analyze the factors of vaccine hesitancy based on the "3Cs" model, and to provide reference for population intervention. Methods: From Dec 31, 2020 to Jan 11, 2021, a convenience sampling method was adopted to conduct an online survey of residents in 34 provinces, cities and autonomous regions across the country.The survey content included demographic characteristics, vaccine hesitancy, and the dimensions of the "3Cs" (confidence, complacency, and convenience) model.We analyzed the influence of demographic characteristics on vaccine hesitancy by χ2 test.Logistic regression was used to evaluate the effects of "3Cs" variables on vaccine hesitancy. Results: A total of 2 531 respondents were surveyed.Their average age was (33.9±8.9) years old, male to female ratio was 1:1.42, and vaccine hesitating respondents accounted for 44.3%.Logistic regression analysis found that gender (ORfemale=1.33, 95%CI: 1.12-1.58), monthly income (compared with<6 000 yuan, OR>10 000=1.57, 95%CI: 1.25-1.97), healthcare workers(ORno=1.39, 95%CI: 1.12-1.73), and confidence (OR=0.47, 95%CI: 0.40-0.56) and complacency (OR=2.49, 95%CI: 2.10-2.96) in the "3Cs" model showed statistically significant impacts on the hesitation of COVID-19 vaccines. Conclusion: The confidence and complacency dimensions in the "3Cs" model have an impact on the hesitation of COVID-19 vaccines.Future interventions can focus on improving the public's confidence and reducing complacency associated with COVID-19 vaccines to increase the vaccination rate. © 2021, Editorial Department of Fudan University Journal of Medical Sciences. All right reserved.

Towards a Realistic Model for Simulating Spread of Infectious COVID-19 Disease

The classical compartment model namely SEIR was not designed originally for COVID-19. There are significant numbers of people infected by COVID-19 did not get sick immediately but have become carriers of COVID-19. The patients might have certain length of incubation period. In order to cover the quarantine and asymptomatic variables, the existing SEIR model is extended to a multi-compartment infectious disease model. The contribution presented in this paper is a new model called SEAIRD which caters for the new characteristics of the 2019-nCoV, therefore applicable for the inclusion of asymptomatic population in the simulation. © 2020 ACM.