Association of air pollutants and meteorological factors with tuberculosis: a national multicenter ecological study in China.

The impact of weather variability and air pollutants on tuberculosis (TB) has been a research hotspot. Previous studies have mostly been limited to a certain area or with a small sample size of cases, and multi-scale systematic studies are lacking. In this study, 14,816,329 TB cases were collected from 31 provinces in China between 2004 and 2018 to estimate the association between TB risk and meteorological factors and air pollutants using a two-stage time-series analysis. The impact and lagged time of meteorological factors and air pollutants on TB risk varied greatly in different provinces and regions. Overall cumulative exposure-response summary associations across 31 provinces suggested that high monthly mean relative humidity (RH) (66.8-82.4%, percentile56-100 (P56-100)), rainfall (316.5-331.1 mm, P96-100), PM2.5 exposure concentration (93.3-145.0 µg/m3, P58-100), and low monthly mean wind speed (1.6-2.1 m/s, P0-38) increased the risk of TB incidence, with a relative risk (RR) of 1.10 (95% CI: 1.04-1.16), 1.10 (95% CI: 1.03-1.16), 2.08 (95% CI: 1.18-3.65), and 2.06 (95% CI: 1.27-3.33), and attributable risk percent (AR%) of 9%, 9%, 52%, and 51%, respectively. Conversely, high monthly average wind speed (2.3-2.9 m/s, P54-100) and mean temperature (20.2-25.3 °C, P79-96), and low monthly average rainfall (2.4-25.2 mm, P0-7) and concentration of SO2 (8.1-21.2 µg/m3, P0-16) exposure decreased the risk of TB incidence, with an overall cumulative RR of 0.92 (95% CI: 0.87-0.98), 0.74 (95% CI: 0.59-0.94), 0.87 (95% CI: 0.79-0.95), and 0.72 (95% CI: 0.56-0.93), respectively. Our study provided insights into future planning of public health interventions for TB.

Large-Scale Person Re-Identification Based on Deep Hash Learning.

Person re-identification in the image processing domain has been a challenging research topic due to the influence of pedestrian posture, background, lighting, and other factors. In this paper, the method of harsh learning is applied in person re-identification, and we propose a person re-identification method based on deep hash learning. By improving the conventional method, the method proposed in this paper uses an easy-to-optimize shallow convolutional neural network to learn the inherent implicit relationship of the image and then extracts the deep features of the image. Then, a hash layer with three-step calculation is incorporated in the fully connected layer of the network. The hash function is learned and mapped into a hash code through the connection between the network layers. The generation of the hash code satisfies the requirements that minimize the error of the sum of quantization loss and Softmax regression cross-entropy loss, which achieve the end-to-end generation of hash code in the network. After obtaining the hash code through the network, the distance between the pedestrian image hash code to be retrieved and the pedestrian image hash code library is calculated to implement the person re-identification. Experiments conducted on multiple standard datasets show that our deep hashing network achieves the comparable performances and outperforms other hashing methods with large margins on Rank-1 and mAP value identification rates in pedestrian re-identification. Besides, our method is predominant in the efficiency of training and retrieval in contrast to other pedestrian re-identification algorithms.