Using machine learning to predict COVID-19 infection and severity risk among 4510 aged adults: a UK Biobank cohort study.

Many risk factors have emerged for novel 2019 coronavirus disease (COVID-19). It is relatively unknown how these factors collectively predict COVID-19 infection risk, as well as risk for a severe infection (i.e., hospitalization). Among aged adults (69.3 ± 8.6 years) in UK Biobank, COVID-19 data was downloaded for 4510 participants with 7539 test cases. We downloaded baseline data from 10 to 14 years ago, including demographics, biochemistry, body mass, and other factors, as well as antibody titers for 20 common to rare infectious diseases in a subset of 80 participants with 124 test cases. Permutation-based linear discriminant analysis was used to predict COVID-19 risk and hospitalization risk. Probability and threshold metrics included receiver operating characteristic curves to derive area under the curve (AUC), specificity, sensitivity, and quadratic mean. Model predictions using the full cohort were marginal. The "best-fit" model for predicting COVID-19 risk was found in the subset of participants with antibody titers, which achieved excellent discrimination (AUC 0.969, 95% CI 0.934-1.000). Factors included age, immune markers, lipids, and serology titers to common pathogens like human cytomegalovirus. The hospitalization "best-fit" model was more modest (AUC 0.803, 95% CI 0.663-0.943) and included only serology titers, again in the subset group. Accurate risk profiles can be created using standard self-report and biomedical data collected in public health and medical settings. It is also worthwhile to further investigate if prior host immunity predicts current host immunity to COVID-19.

An optimised protocol for detection of SARS-CoV-2 in stool.

BACKGROUND: SARS-CoV-2 has been detected in stool samples of COVID-19 patients, with potential implications for faecal-oral transmission. Compared to nasopharyngeal swab samples, the complexity of the stool matrix poses a challenge in the detection of the virus that has not yet been solved. However, robust and reliable methods are needed to estimate the prevalence and persistence of SARS-CoV-2 in the gut and to ensure the safety of microbiome-based procedures such as faecal microbiota transplant (FMT). The aim of this study was to establish a sensitive and reliable method for detecting SARS-CoV-2 in stool samples. RESULTS: Stool samples from individuals free of SARS-CoV-2 were homogenised in saline buffer and spiked with a known titre of inactivated virus ranging from 50 to 750 viral particles per 100 mg stool. Viral particles were concentrated by ultrafiltration, RNA was extracted, and SARS-CoV-2 was detected via real-time reverse-transcription polymerase chain reaction (RT-qPCR) using the CDC primers and probes. The RNA extraction procedure we used allowed for the detection of SARS-CoV-2 via RT-qPCR in most of the stool samples tested. We could detect as few as 50 viral particles per 100 mg of stool. However, high variability was observed across samples at low viral titres. The primer set targeting the N1 region provided more reliable and precise results and for this primer set our method had a limit of detection of 1 viral particle per mg of stool. CONCLUSIONS: Here we describe a sensitive method for detecting SARS-CoV-2 in stool samples. This method can be used to establish the persistence of SARS-CoV-2 in stool and ensure the safety of clinical practices such as FMT.

Spatiotemporal Big Data for PM2.5 Exposure and Health Risk Assessment during COVID-19.

The coronavirus disease 2019 (COVID-19) first identified at the end of 2019, significantly impacts the regional environment and human health. This study assesses PM2.5 exposure and health risk during COVID-19, and its driving factors have been analyzed using spatiotemporal big data, including Tencent location-based services (LBS) data, place of interest (POI), and PM2.5 site monitoring data. Specifically, the empirical orthogonal function (EOF) is utilized to analyze the spatiotemporal variation of PM2.5 concentration firstly. Then, population exposure and health risks of PM2.5 during the COVID-19 epidemic have been assessed based on LBS data. To further understand the driving factors of PM2.5 pollution, the relationship between PM2.5 concentration and POI data has been quantitatively analyzed using geographically weighted regression (GWR). The results show the time series coefficients of monthly PM2.5 concentrations distributed with a U-shape, i.e., with a decrease followed by an increase from January to December. In terms of spatial distribution, the PM2.5 concentration shows a noteworthy decline over the Central and North China. The LBS-based population density distribution indicates that the health risk of PM2.5 in the west is significantly lower than that in the Middle East. Urban gross domestic product (GDP) and urban green area are negatively correlated with PM2.5; while, road area, urban taxis, urban buses, and urban factories are positive. Among them, the number of urban factories contributes the most to PM2.5 pollution. In terms of reducing the health risks and PM2.5 pollution, several pointed suggestions to improve the status has been proposed.

Spatiotemporal Big Data for PM2.5 Exposure and Health Risk Assessment during COVID-19

The coronavirus disease 2019 (COVID-19) first identified at the end of 2019, significantly impacts the regional environment and human health. This study assesses PM2.5 exposure and health risk during COVID-19, and its driving factors have been analyzed using spatiotemporal big data, including Tencent location-based services (LBS) data, place of interest (POI), and PM2.5 site monitoring data. Specifically, the empirical orthogonal function (EOF) is utilized to analyze the spatiotemporal variation of PM2.5 concentration firstly. Then, population exposure and health risks of PM2.5 during the COVID-19 epidemic have been assessed based on LBS data. To further understand the driving factors of PM2.5 pollution, the relationship between PM2.5 concentration and POI data has been quantitatively analyzed using geographically weighted regression (GWR). The results show the time series coefficients of monthly PM2.5 concentrations distributed with a U-shape, i.e., with a decrease followed by an increase from January to December. In terms of spatial distribution, the PM2.5 concentration shows a noteworthy decline over the Central and North China. The LBS-based population density distribution indicates that the health risk of PM2.5 in the west is significantly lower than that in the Middle East. Urban gross domestic product (GDP) and urban green area are negatively correlated with PM2.5;while, road area, urban taxis, urban buses, and urban factories are positive. Among them, the number of urban factories contributes the most to PM2.5 pollution. In terms of reducing the health risks and PM2.5 pollution, several pointed suggestions to improve the status has been proposed.