Objective: To understand the current status of neglect among rural preschool children in Xi an under the multi child policy and the association with family characteristics, so as to provide a reference for preventing and reducing the occurrence of child neglect. Methods: A total of 7 052 parents of preschool children were selected using stratified cluster sampling across 9 suburban counties/districts in Xi an from March to April 2025. A questionnaire survey was administered using the Chinese Norm Scale for Neglect Assessment of Rural(Preschool) Children Aged 3-6. The t-test, Chi-quare test, and analysis of variance (ANOVA) were used for inter group comparisons. Results: The overall prevalence rate and mean score of neglect among rural preschool aged children in Xi an were 32.4% and 38.27±6.70, respectively. Statistically significant differences were detected in neglect rates and neglect degrees among preschool children of different genders and grade levels ( χ 2=30.41, 15.15, t/F =4.92,7.03, all P <0.05). Statistically significant differences were also detected in neglect rates and neglect degrees among preschool children from whether only one child, different family structures, numbers of children in a family and families with different annual incomes ( χ 2=29.22, 10.41 , 31.99, 186.47, t/F =-9.96, 5.50, 33.57, 68.63, all P <0.05). In multi child families, there was a statistically significant difference in neglect degree among children with different birth orders ( F =4.25, P <0.05), but there was no statistically significant difference in neglect rate ( χ 2=5.73, P >0.05). Among all subgroups, the highest neglect rates and neglect degrees were observed in children from multi child families(35.0%,39.00±6.71), other family types(50.0%，42.38±12.34) and families with three children(39.9%，39.50±7.43). Lower annual family income was associated with higher neglect rates and neglect degrees among preschool children( χ 2 trend =186.47， F =270.68，both P <0.05). Conclusions Under the multiple child policy, the neglect of preschool children in rural areas of Xi an is quite severe, particularly in families with multiple children and low income households. Targeted interventions should be implemented for high risk groups.

[摘 要] 目的：探讨异位表达血红蛋白亚基（HBA/HBB）对缺氧条件下嵌合抗原受体T细胞（CAR-T细胞）功能障碍的改善作用及其对肿瘤细胞的杀伤效应。方法：全基因合成技术合成靶向HER2的CAR序列，构建共表达HBA或HBB的CAR慢病毒载体，包装慢病毒后感染人原代T淋巴细胞，制备异位表达HBA/HBB的CAR-T细胞，命名为HBA CAR-T和HBB CAR-T。采用缺氧探针检测小鼠实体瘤缺氧状态。通过流式细胞术检测瘤内CAR-T细胞占比、异位表达血红蛋白亚基的CAR-T细胞阳性率及CAR-T细胞的活性氧、凋亡水平。WB法检测HBA CAR-T和HBB CAR-T内相关血红蛋白亚基表达情况，采用细胞计数板计数检测细胞增殖水平，通过萤光素酶报告基因法检测CAR-T细胞对肿瘤细胞的杀伤能力，qPCR检测CAR-T细胞中缺氧诱导因子-1α（HIF-1α）表达水平，利用MitoXpress Intra试剂盒检测CAR-T细胞内氧气含量。结果：不同细胞构建的实体瘤模型均存在明显缺氧情况，且CAR-T细胞浸润水平与缺氧程度呈显著负相关（P < 0.000 1）。HBA CAR-T与HBB CAR-T构建成功（阳性率 > 60%），相应血红蛋白亚基可稳定表达。缺氧环境下HBA CAR-T和HBB CAR-T的ROS水平、凋亡水平显著下降，增殖、对肿瘤细胞的体外杀伤能力显著强于传统CAR-T细胞（均P < 0.05）。HBA CAR-T与HBB CAR-T内HIF-1α表达降低（均P < 0.001），且缺氧程度显著降低（均P < 0.001）。结论：异位表达血红蛋白亚基可改善缺氧条件下CAR-T细胞功能障碍并增强其对肿瘤细胞的体外杀伤作用。

Objective: To explore the association between sleep quality and dry eye symptoms in adolescents,so as to provide the evidence for reducing the prevalence of dry eye symptoms. Methods: The study population was adolescents aged 12-24 years from the Psychology and Behavior Investigation of Chinese Residents (PBICR) survey, which was conducted from 20 June to 31 August 2022. A stratified random sampling and quota sampling method was used to select 6 456 adolescents within mainland China. The Ocular Surface Disease Index (OSDI) and Brief version of the Pittsburgh Sleep Quality Index (B-PSQI) were used to assess dry eye symptoms and sleep quality. Multiple Logistic regression model was used to explore the relationship between sleep quality and dry eye symptoms in adolescents. The influence of gender on the association was explored by using interaction terms. Results: A total of 2 815 adolescents reported having dry eye symptoms, with a prevalence of 43.6%. Logistic regression analysis results showed an increased risk of exacerbation of dry eye symptoms in adolescents with poor sleep quality. The OR (95% CI ) for mild, moderate, and severe dry eye symptoms groups were 1.39(1.16-1.67), 1.52(1.28-1.81), and 2.35(2.02-2.72), respectively, compared with the ocularly normal group ( P <0.05). There was a significant interaction between sleep quality and gender on dry eye symptoms in adolescents ( P <0.01). Conclusions Sleep quality is associated with dry eye symptoms in adolescents, and those with poor sleep quality have a higher risk of dry eye symptoms. The effect of sleep quality on dry eye symptoms is greater in boys.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Forsythia suspensa is a traditional Chinese medicine and a commonly used landscaping plant. Its dried fruit is used in medicine for its functions of clearing heat, removing toxins, reducing swelling, dissipating masses, and dispersing wind and heat. It possesses extremely high medicinal and economic value. However, the genetic differentiation and diversity of its wild populations remain unclear. In this study, chloroplast genome sequences were obtained from 15 wild individuals of F. suspensa using high-throughput sequencing technology. The sequence characteristics and intraspecific variations were analyzed. The results were as follows:(1) The full length of the F. suspensa chloroplast genome ranged from 156 184 to 156 479 bp, comprising a large single-copy region, a small single-copy region, and two inverted repeat regions. The chloroplast genome encoded a total of 132 genes, including 87 protein-coding genes, 37 tRNA genes, and 8 rRNA genes.(2) A total of 166-174 SSR loci, 792 SNV loci, and 63 InDel loci were identified in the F. suspensa chloroplast genome, indicating considerable genetic variation among individuals.(3) Population structure analysis revealed that F. suspensa could be divided into five or six groups. Both the population structure analysis and phylogenetic reconstruction results indicated significant genetic variation within the wild populations of F. suspensa, with no obvious correlation between intraspecific genetic differentiation and geographical distribution. This study provides new insights into the genetic diversity and differentiation within F. suspensa species and offers additional references for the conservation of species diversity and the utilization of germplasm resources in wild F. suspensa.
Genome, Chloroplast ; Forsythia/classification* ; Phylogeny ; Genetic Variation ; Chloroplasts/genetics* ; Microsatellite Repeats

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

This paper reviews the published articles of recent years on the application of deep learning methods in automatic localization of acupoint, and summarizes it from 3 key links, i.e. the dataset construction, the neural network model design, and the accuracy evaluation of acupoint localization. The significant progress has been obtained in the field of deep learning for acupoint localization, but the scale of acupoint detection needs to be expanded and the precision, the generalization ability, and the real-time performance of the model be advanced. The future research should focus on the support of standardized datasets, and the integration of 3D modeling and multimodal data fusion, so as to increase the accuracy and strengthen the personalization of acupoint localization.
Deep Learning ; Acupuncture Points ; Humans ; Neural Networks, Computer