ABSTRACT
Background Studies on the association between greenness exposure and allergic rhinitis (AR) in children are mostly conducted in developed countries, and the conclusion is not consistent. Objective Using street view data to explore the association between greenness exposure and allergic rhinitis (AR) prevalence in Chinese children. Methods A cross-sectional study was conducted among 40868 children aged 2-17 years in three cities of Northeast China from 2012 to 2013, which consisted of 20886 (51.1%) boys and 19982 (48.9%) girls. The information of AR prevalence was obtained through questionnaire. Based on downloaded street view images from Tencent Maps, a green view index (GVI) of green vegetation (trees and grass) within 800 m and 1000 m buffer of the participants' schools was calculated by using artificial intelligence, and it was used as a surrogate of the greenness exposure. A mixed-effect logistic regression model was used to estimate the odds ratio (OR) of AR prevalence in children for per increase of inter-quartile range (IQR) of GVI. In addition, according to ambient PM2.5 concentration, the participants were divided into a low PM2.5 exposure group (≤56.23 μg·m−3) and a high exposure group (>56.23 μg·m−3) to investigate whether PM2.5 was a modifier on the association between GVI and AR. Results The average age of the subjects was (10.40±3.68) years and 3 963 (9.7%) subjects reported diagnosed AR. Within 800 m buffer, an IQR increase in GVI for trees (IQR=0.031, OR=0.85, 95%CI: 0.81-0.90) and overall greenness (IQR=0.029, OR=0.86, 95%CI: 0.81-0.90) was associated with lower adjusted odds ratio of AR. The interaction between PM2.5 and GVI was statistically significant (P< 0.1), that is, the negative associations of trees and overall greenness with AR were observed only at low PM2.5 exposure levels. The sensitivity analysis results of GVI within 1000 m buffer was consistent with that within 800 m buffer. Conclusion Exposure to green vegetation, especially trees, may be associated with decreased risks of AR in children, and such associations may be more obvious in areas with a low PM2.5 concentration.
ABSTRACT
Objective:Topredict the three-dimensional dose distribution of regions of interest (ROI) with brachytherapy for cervical cancer based on U-Net fully convolutional network, and evaluate the accuracy of prediction model.Methods:First, 100 cases of cervical cancer intracavity combined with interstitial implantation were selected as the entire research data set, and divided into the training set ( n=72), validation set ( n=8), and test set ( n=20). Then the U-Net was used to construct two models based on whether the uterine tandem and the implantation needles were included as the distinguishing factors. Finally, dose distribution of 20 cases in the test set were predicted using the trained model, and comparative analysis was performed. The performance of the model was jointly evaluated by , and the mean absolute deviation (MAD). Results:Compared with the model without the uterine tandem and the implantation needles, the of the rectum was increased by (16.83±1.82) cGy ( P<0.05), and the or of the other ROI were not different significantly (all P>0.05). The MAD of the high-risk clinical target volume, rectum, sigmoid, small bowel, and bladder was increased by (11.96±3.78) cGy, (11.43±0.54) cGy, (24.08±1.65) cGy, (17.04±7.17) cGy and (9.52±4.35) cGy, respectively (all P<0.05). The MAD of the intermediate-risk clinical target volume was decreased by (120.85±29.78) cGy ( P<0.05). The mean value of MAD for all ROI was decreased by (7.8±53) cGy ( P<0.05), which was closer to the actual plan. Conclusions:U-Net fully convolutional network can be used to predict three-dimensional dose distribution of patients with cervical cancer undergoing brachytherapy. Combining the uterine tube with the implantation needles as the input parameters yields more accurate predictions than a single use of the ROI structure as the input.
ABSTRACT
Diseases such as diabetes and hypertension can lead to change the shape of the retinal blood vessels. Segmentation of fundus images is a key step in the process of quantitative analysis of the disease, which is instructive in the analysis and diagnosis of clinical diseases. In this paper, a method for the segmentation of retinal image vessels based on fully convolutional network (FCN) with depthwise separable convolution and channel weighting is presented. Firstly, CLAHE and Gamma correction of the green channel of the fundus image are used to enhance the contrast. Then, in order to adapt to network training, the enhanced image is divided into patches to expand the data. Finally, the depthwise separable convolution instead of the standard convolution method is used to increase the network width. Meanwhile, the channel weighting module is introduced to explicitly model the relationship between the characteristic channels in order to improve the distinguishability of the features. The combination of them is applied to the FCN and the results of expert manual identification are used to supervise the experiment on the DRIVE database. The results show that the segmentation accuracy of the proposed method in DRIVE database reached 0.963 0 and AUC reached 0.983 1. The segmentation accuracy in STARE database reached 0.962 0 and AUC achieved 0.983 0. To some extent, the proposed method has better feature resolution and better segmentation performance.