ObjectiveLimited information is available regarding spatiotemporal variations of particles with median aerodynamic diameter < 2.5 μm (PM2.5) at high resolutions, and their relationships with meteorological factors in Beijing, China. This study aimed to detect spatiotemporal change patterns of PM2.5 from August 2013 to July 2014 in Beijing, and to assess the relationship between PM2.5 and meteorological factors.MethodsDaily and hourly PM2.5 data from the Beijing Environmental Protection Bureau (BJEPB) were analyzed separately. Ordinary kriging (OK) interpolation, time-series graphs, Spearman correlation coefficient and coefficient of divergence (COD) were used to describe the spatiotemporal variations of PM2.5. The Kruskal-Wallis H test, Bonferroni correction, and Mann-Whitney U test were used to assess differences in PM2.5 levels associated with spatial and temporal factors including season, region, daytime and day of week. Relationships between daily PM2.5 and meteorological variables were analyzed using the generalized additive mixed model (GAMM).ResultsAnnual mean and median of PM2.5 concentrations were 88.07 μg/m3 and 71.00 μg/m3, respectively, from August 2013 to July 2014. PM2.5 concentration was significantly higher in winter (P < 0.0083) and in the southern part of the city (P < 0.0167). Day to day variation of PM2.5 showed a long-term trend of fluctuations, with 2–6 peaks each month. PM2.5 concentration was significantly higher in the night than day (P < 0.0167). Meteorological factors were associated with daily PM2.5 concentration using the GAMM model (R
2 = 0.59, AIC = 7373.84).ConclusionPM2.5 pollution in Beijing shows strong spatiotemporal variations. Meteorological factors influence the PM2.5 concentration with certain patterns. Generally, prior day wind speed, sunlight hours and precipitation are negatively correlated with PM2.5, whereas relative humidity and air pressure three days earlier are positively correlated with PM2.5.
Aims
Facial features were associated with increased risk of coronary artery disease (CAD). We developed and validated a deep learning algorithm for detecting CAD based on facial photos.
Methods and results
We conducted a multicentre cross-sectional study of patients undergoing coronary angiography or computed tomography angiography at nine Chinese sites to train and validate a deep convolutional neural network for the detection of CAD (at least one ≥50% stenosis) from patient facial photos. Between July 2017 and March 2019, 5796 patients from eight sites were consecutively enrolled and randomly divided into training (90%, n = 5216) and validation (10%, n = 580) groups for algorithm development. Between April 2019 and July 2019, 1013 patients from nine sites were enrolled in test group for algorithm test. Sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were calculated using radiologist diagnosis as the reference standard. Using an operating cut point with high sensitivity, the CAD detection algorithm had sensitivity of 0.80 and specificity of 0.54 in the test group; the AUC was 0.730 (95% confidence interval, 0.699–0.761). The AUC for the algorithm was higher than that for the Diamond–Forrester model (0.730 vs. 0.623, P < 0.001) and the CAD consortium clinical score (0.730 vs. 0.652, P < 0.001).
Conclusion
Our results suggested that a deep learning algorithm based on facial photos can assist in CAD detection in this Chinese cohort. This technique may hold promise for pre-test CAD probability assessment in outpatient clinics or CAD screening in community. Further studies to develop a clinical available tool are warranted.
(1) Background: Evidence regarding scarlet fever and its relationship with meteorological, including air pollution factors, is not very available. This study aimed to examine the relationship between ambient air pollutants and meteorological factors with scarlet fever occurrence in Beijing, China. (2) Methods: A retrospective ecological study was carried out to distinguish the epidemic characteristics of scarlet fever incidence in Beijing districts from 2013 to 2014. Daily incidence and corresponding air pollutant and meteorological data were used to develop the model. Global Moran’s I statistic and Anselin’s local Moran’s I (LISA) were applied to detect the spatial autocorrelation (spatial dependency) and clusters of scarlet fever incidence. The spatial lag model (SLM) and spatial error model (SEM) including ordinary least squares (OLS) models were then applied to probe the association between scarlet fever incidence and meteorological including air pollution factors. (3) Results: Among the 5491 cases, more than half (62%) were male, and more than one-third (37.8%) were female, with the annual average incidence rate 14.64 per 100,000 population. Spatial autocorrelation analysis exhibited the existence of spatial dependence; therefore, we applied spatial regression models. After comparing the values of R-square, log-likelihood and the Akaike information criterion (AIC) among the three models, the OLS model (R2 = 0.0741, log likelihood = −1819.69, AIC = 3665.38), SLM (R2 = 0.0786, log likelihood = −1819.04, AIC = 3665.08) and SEM (R2 = 0.0743, log likelihood = −1819.67, AIC = 3665.36), identified that the spatial lag model (SLM) was best for model fit for the regression model. There was a positive significant association between nitrogen oxide (p = 0.027), rainfall (p = 0.036) and sunshine hour (p = 0.048), while the relative humidity (p = 0.034) had an adverse association with scarlet fever incidence in SLM. (4) Conclusions: Our findings indicated that meteorological, as well as air pollutant factors may increase the incidence of scarlet fever; these findings may help to guide scarlet fever control programs and targeting the intervention.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.