Multiple correspondence analysis of the influencing factors of injury among left behind primary school students in Pingliang / 中国学校卫生

Objective@#To explore the distribution characteristics of the influencing factors of injury among left behind primary school students, so as to provide a reference for identifying high risk injury groups and carrying out accurate injury intervention.@*Methods@#From August 2021 to July 2022, a questionnaire survey was conducted among 3 769 students from grades 4 to 6 from nine primary schools in three townships in Pingliang City by using the a random cluster sampling method. Multiple correspondence analysis was used to analyze the influencing factors of injury.@*Results@#The detection rate of injuries among non left behind pupils was 21.8%(573/2 631), whereas the detection rate of injuries among left behind pupils was 40.9%(466/1 138). In addition, a statistically significant difference was observed in the detection rate of injuries between left behind pupils and non left behind pupils ( χ 2=146.21, P <0.01). Among the injuries of left behind pupils, 263 had fall related injuries, accounting for the highest proportion ( 56.4 %). Whether it was an only child，and different grades, gender, personality, psychological status, monitoring type, and maternal education level were statistically significant ( χ 2=39.05, 96.69, 143.00, 155.80, 461.39, 285.35, 17.10, P <0.01). The multiple correspondence analysis category graph showed higher rates of fall injuries, blunt injuries and sharp injuries among boys, extroverted personality types, and left behind pupils whose grandparents were their legal guardians. Animal bites, burns and other injury types were higher among left behind pupils with an introverted personality, pupils in peer/other guardianship situations, and those with a sub mental health status. Unharmed left behind students mainly included those with intermediate personality and mental health characteristics.@*Conclusion@#The injury detection rate among left behind primary school students is high. Gender, personality type, guardianship type, and mental health status are closely related to injury. Extroverted boys under grandparents guardians are identified as high risk groups for injury prevention and control. Under the guidance of the precision prevention model, precision intervention strategies for all round high risk groups should be carried out in order to effectively reduce the occurrence of injuries among left behind pupils.

PDBL: Improving Histopathological Tissue Classification With Plug-and-Play Pyramidal Deep-Broad Learning.

Histopathological tissue classification is a simpler way to achieve semantic segmentation for the whole slide images, which can alleviate the requirement of pixel-level dense annotations. Existing works mostly leverage the popular CNN classification backbones in computer vision to achieve histopathological tissue classification. In this paper, we propose a super lightweight plug-and-play module, named Pyramidal Deep-Broad Learning (PDBL), for any well-trained classification backbone to improve the classification performance without a re-training burden. For each patch, we construct a multi-resolution image pyramid to obtain the pyramidal contextual information. For each level in the pyramid, we extract the multi-scale deep-broad features by our proposed Deep-Broad block (DB-block). We equip PDBL in three popular classification backbones, ShuffLeNetV2, EfficientNetb0, and ResNet50 to evaluate the effectiveness and efficiency of our proposed module on two datasets (Kather Multiclass Dataset and the LC25000 Dataset). Experimental results demonstrate the proposed PDBL can steadily improve the tissue-level classification performance for any CNN backbones, especially for the lightweight models when given a small among of training samples (less than 10%). It greatly saves the computational resources and annotation efforts. The source code is available at: https://github.com/linjiatai/PDBL.

Joint optic disc and cup segmentation using semi-supervised conditional GANs.

Glaucoma is a chronic and widespread eye disease threatening humans' irreversible vision loss. The cup-to-disc ratio (CDR), one of the most important measurements used for glaucoma screening and diagnosis, requires accurate segmentation of optic disc and cup from fundus images. However, most existing techniques fail to obtain satisfactory segmentation performance because a significant number of pixel-level annotated data are often unavailable during training. To cope with this limitation, in this paper, we propose an effective joint optic disc and cup segmentation method based on semi-supervised conditional Generative Adversarial Nets (GANs). Our architecture consists of a segmentation net, a generator and a discriminator, to learn a mapping between the fundus images and the corresponding segmentation maps. Additionally, we employ both labeled and unlabeled data to improve the segmentation performance. The extensive experiments show that our method achieves state-of-the-art optic disc and cup segmentation results on both ORIGA and REFUGE datasets.

Improved de novo genome assembly and analysis of the Chinese cucurbit Siraitia grosvenorii, also known as monk fruit or luo-han-guo.

Background: Luo-han-guo (Siraitia grosvenorii), also called monk fruit, is a member of the Cucurbitaceae family. Monk fruit has become an important area for research because of the pharmacological and economic potential of its noncaloric, extremely sweet components (mogrosides). It is also commonly used in traditional Chinese medicine for the treatment of lung congestion, sore throat, and constipation. Recently, a single reference genome became available for monk fruit, assembled from 36.9x genome coverage reads via Illumina sequencing platforms. This genome assembly has a relatively short (34.2 kb) contig N50 length and lacks integrated annotations. These drawbacks make it difficult to use as a reference in assembling transcriptomes and discovering novel functional genes. Findings: Here, we offer a new high-quality draft of the S. grosvenorii genome assembled using 31 Gb (∼73.8x) long single molecule real time sequencing reads and polished with ∼50 Gb Illumina paired-end reads. The final genome assembly is approximately 469.5 Mb, with a contig N50 length of 432,384 bp, representing a 12.6-fold improvement. We further annotated 237.3 Mb of repetitive sequence and 30,565 consensus protein coding genes with combined evidence. Phylogenetic analysis showed that S. grosvenorii diverged from members of the Cucurbitaceae family approximately 40.9 million years ago. With comprehensive transcriptomic analysis and differential expression testing, we identified 4,606 up-regulated genes in the early fruit compared to the leaf, a number of which were linked to metabolic pathways regulating fruit development and ripening. Conclusions: The availability of this new monk fruit genome assembly, as well as the annotations, will facilitate the discovery of new functional genes and the genetic improvement of monk fruit.