Masks, ventilation and exposure time: A web-based calculator of indoor COVID-19 infection risk

Two and half years into the COVID-19 pandemic, there is quite a lot of confusion over public health guidance necessary in order to reduce disease infection risks, from room air ventilation, the use of air cleaners, and type of mask and whether or not to wear a mask. This paper describes the development of a novel web-based calculator for use by the public to assess COVID-19 infection risks between a source and receiver in a typical room. The aim is to inform the disease infection risk in response to varying exposure times, mask-wearing, and viral variant in circulation. The calculator is based on the state-of-the-art research evidence, i.e., a room air ventilation model, mask infiltration efficiencies, room cleaner efficiencies, the quanta emission rates of various viral variants of COVID-19, and the modified Wells Riley equations. The results show that exposure times are critical in determining transmission risk. Masks are important and can reduce infection risk especially over shorter exposure times and for lower source emission quantum. N95 respirators are by far the most effective, especially for Omicron, and the results indicate that N95 respirators are necessary for the more infectious variants. Increasing fresh air ventilation rates from 2ac/h to 6ac/h can have a considerable impact in reducing transmission risk in a well-mixed space. Going from 6 ac/h to 12ac/h is less effective especially at lower exposure times. Venues can be classified in terms of risk, and appropriate high ventilation rates might be recommended for high-risk, speaking loudly and singing, such as classrooms and theatres. However, for low risk, quiet and speaking softly venues, such as offices and libraries, higher ventilation rates may not be required;instead, mechanical ventilation systems in combination with air cleaners can effectively remove small fraction size aerosol particles. The web-based calculator provides an easy-to-use and valuable tool for use in estimating infection risk.

[Course of disease and related epidemiological parameters of COVID-19: a prospective study based on contact tracing cohort].

Objective: To analyze the course of disease and epidemiological parameters of COVID-19 and provide evidence for making prevention and control strategies. Methods: To display the distribution of course of disease of the infectors who had close contacts with COVID-19 cases from January 1 to March 15, 2020 in Guangdong Provincial, the models of Lognormal, Weibull and gamma distribution were applied. A descriptive analysis was conducted on the basic characteristics and epidemiological parameters of course of disease. Results: In total, 515 of 11 580 close contacts were infected, with an attack rate about 4.4%, including 449 confirmed cases and 66 asymptomatic cases. Lognormal distribution was fitting best for latent period, incubation period, pre-symptomatic infection period of confirmed cases and infection period of asymptomatic cases; Gamma distribution was fitting best for infectious period and clinical symptom period of confirmed cases; Weibull distribution was fitting best for latent period of asymptomatic cases. The latent period, incubation period, pre-symptomatic infection period, infectious period and clinical symptoms period of confirmed cases were 4.50 (95%CI:3.86-5.13) days, 5.12 (95%CI:4.63-5.62) days, 0.87 (95%CI:0.67-1.07) days, 11.89 (95%CI:9.81-13.98) days and 22.00 (95%CI:21.24-22.77) days, respectively. The latent period and infectious period of asymptomatic cases were 8.88 (95%CI:6.89-10.86) days and 6.18 (95%CI:1.89-10.47) days, respectively. Conclusion: The estimated course of COVID-19 and related epidemiological parameters are similar to the existing data.

Enhancing Automated COVID-19 Chest X-ray Diagnosis by Image-to-Image GAN Translation

The severe pneumonia induced by the infection of the SARS-CoV-2 virus causes massive death in the ongoing COVID-19 pandemic. The early detection of the SARS-CoV-2 induced pneumonia relies on the unique patterns of the chest XRay images. Deep learning is a data-greedy algorithm to achieve high performance when adequately trained. A common challenge for machine learning in the medical domain is the accessibility to properly annotated data. In this study, we apply a conditional adversarial network (cGAN) to perform image to image (Pix2Pix) translation from the non-COVID-19 chest X-Ray domain to the COVID-19 chest X-Ray domain. The objective is to learn a mapping from the normal chest X-Ray visual patterns to the COVID-19 pneumonia chest X-ray patterns. The original dataset has a typical imbalanced issue because it contains only 219 COVID-19 positive images but has 1,341 images for normal chest X-Ray and 1,345 images for viral pneumonia. A U-Net based architecture is applied for the image-to-image translation to generate synthesized COVID-19 X-Ray chest images from the normal chest X-ray images. A 50-convolutional-layer residual net (ResNet) architecture is applied for the final classification task. After training the GAN model for 100 epochs, we use the GAN generator to translate 1,100 COVID-19 images from the normal X-Ray to form a balanced training dataset (3,762 images) for the classification task. The ResNet based classifier trained by the enhanced dataset achieves the classification accuracy of 97.8% compared to 96.1% in the transfer learning mode. When trained with the original imbalanced dataset, the model achieves an accuracy of 96.1% compared to 95.6% in the training from trainby-scratch model. In addition, the classifier trained by the enhanced dataset has more stable measures in precision, recall, and F1 scores across different image classes. We conclude that the GAN-based data enhancement strategy is applicable to most medical image pattern recognition tasks, and it provides an effective way to solve the common expertise dependence issue in the medical domain. © 2020 IEEE.