Low Level of Accurate Knowledge of Undetectable Equals Untransmittable (U = U) Among Men Who Have Sex with Men (MSM) Challenges U = U Achievement in Chengdu, Southwestern China.

The introduction of the undetectable equals untransmittable (U = U) statement significantly reduces HIV stigma and provides an empirical basis for achieving "zero transmission." U = U messaging has gained increasing support and is well-developed in several countries. However, the current status of accurate knowledge of U = U and its associated factors among men who have sex with men (MSM) in China remians unclear. We conducted a cross-sectional survey among MSM recruited in Chengdu, China, from March to May 2022 to investigate the percentage of those who knew U = U accurately and to explore associations between sexual risk behaviors, HIV testing, socio-behavioral measures, and accurate knowledge of U = U. Of 497 MSM included in our study, 23.4% (116/497) had accurate knowledge of U = U. More than half of the participants (63.2%, 314/497) reported multiple sexual partnerships, 15.7% (78/497) used substance during sex, 37.4% (186/497) reported inconsistent condom use, and 76.1% (378/497) took HIV testing in the past six months. Factors associated with accurate knowledge of U = U among MSM included substance use during sex (multivariate odds ratios ORm = 1.96; 95%CI: 1.13-3.41), HIV status tested in the last six months (ORm = 2.07; 95%CI: 1.14-3.77), HIV-related literacy (ORm = 1.41; 95%CI: 1.14-1.74) and perceived higher risk of HIV infection (ORm = 1.11; 95%CI: 1.02-1.21). The findings indicate that great challenges need to be conquered for U = U achievement among the MSM population in Chengdu, China. Intervention strategies should be prioritized in health education on high-risk behaviors such as group sex and substance use, information provision and counseling of U = U in HIV testing services, and encouraging U = U knowledge sharing among MSM with their partners in China.

The associations between social support change and physical activity trajectory from late adolescence to young adulthood.

BACKGROUND: Previous research examined the associations between social support and physical activity. However, little is known about the associations between social support change and trajectories of physical activity during the transition from late adolescence to young adulthood. METHODS: The current study sought to examine these issues among 434 Chinese college students (Mage = 19.15, SDage = 0.61; 46.1% male), who completed questionnaires regarding demographics, physical activity, family support change, and peer support change across three waves (the data from one of the waves was retrospective). RESULTS: After controlling for covariates, the findings revealed that: (a) there was an increase in overall physical activity and duration, but a decrease in frequency during the transition from late adolescence (the second year of high school) to young adulthood (the third year of college); (b) family support change did not contribute to trajectories of physical activity, while peer support change significantly predicted the trajectory of overall physical activity, duration, and frequency. CONCLUSIONS: The findings extend the literature on physical activity from a developmental perspective by revealing different trends among physical activity duration and frequency, and unpacking different effects of family and peer support change on trajectories of physical activity.

Defining indicators for the scoping stage of health impact assessment to evaluate tobacco control policy in the city of Beijing.

INTRODUCTION: Beijing initiated the nation's most comprehensive tobacco control program that adheres to the WHO Framework Convention on Tobacco Control. This study aimed to identify a set of indicators for the scoping of an Health Impact Assessment (HIA) to assess this policy. METHODS: This study used a modified Delphi process. It proposed a tobacco control health impact framework based on the Driving forces- Pressure- State- Exposure- Effect- Action model and the Determinants of Health Theory. After a review of current surveillance system and literature, a working group of 13 experts with multidisciplinary background was established to formulate indicator evaluation criteria and conduct indicator scoring. Each indicator was scored by experts according to four evaluation criteria chosen. Indicators that obtained a total score above 80% and with standard error less than 5 were selected as the final set of indicators. Kendall's coefficient of concordance was calculated. RESULTS: Twenty-three out of 36 indicators were selected. Smoking prevalence, mortality rate, hospital admission rate, tobacco consumption and hospital admission fees of smoking related diseases achieved more than 90% of total scores and ranked as the top five. Kendall's concordance coefficient was 0.218 for all indicators. For all model composition, Kendall's concordance coefficients were statistically significant. CONCLUSION: This study identified a set of twenty-three indicators for scoping of HIA of a comprehensive tobacco control policy in Beijing based on a tobacco control health impact conceptual framework. The set of indicators achieved high scores and statistically significant consistency and has great potential to promote the evaluation of tobacco control policy in a global city. Further study might use the set of indicators for HIA on tobacco control policy to analyze empirical data.

Effect of Artificial Intelligence-based Health Education Accurately Linking System (AI-HEALS) for Type 2 diabetes self-management: protocol for a mixed-methods study.

BACKGROUND: Patients with type 2 diabetes (T2DM) have an increasing need for personalized and Precise management as medical technology advances. Artificial intelligence (AI) technologies on mobile devices are being developed gradually in a variety of healthcare fields. As an AI field, knowledge graph (KG) is being developed to extract and store structured knowledge from massive data sets. It has great prospects for T2DM medical information retrieval, clinical decision-making, and individual intelligent question and answering (QA), but has yet to be thoroughly researched in T2DM intervention. Therefore, we designed an artificial intelligence-based health education accurately linking system (AI-HEALS) to evaluate if the AI-HEALS-based intervention could help patients with T2DM improve their self-management abilities and blood glucose control in primary healthcare. METHODS: This is a nested mixed-method study that includes a community-based cluster-randomized control trial and personal in-depth interviews. Individuals with T2DM between the ages of 18 and 75 will be recruited from 40-45 community health centers in Beijing, China. Participants will either receive standard diabetes primary care (SDPC) (control, 3 months) or SDPC plus AI-HEALS online health education program (intervention, 3 months). The AI-HEALS runs in the WeChat service platform, which includes a KBQA, a system of physiological indicators and lifestyle recording and monitoring, medication and blood glucose monitoring reminders, and automated, personalized message sending. Data on sociodemography, medical examination, blood glucose, and self-management behavior will be collected at baseline, as well as 1,3,6,12, and 18 months later. The primary outcome is to reduce HbA1c levels. Secondary outcomes include changes in self-management behavior, social cognition, psychology, T2DM skills, and health literacy. Furthermore, the cost-effectiveness of the AI-HEALS-based intervention will be evaluated. DISCUSSION: KBQA system is an innovative and cost-effective technology for health education and promotion for T2DM patients, but it is not yet widely used in the T2DM interventions. This trial will provide evidence on the efficacy of AI and mHealth-based personalized interventions in primary care for improving T2DM outcomes and self-management behaviors. TRIAL REGISTRATION: Biomedical Ethics Committee of Peking University: IRB00001052-22,058, 2022/06/06; Clinical Trials: ChiCTR2300068952, 02/03/2023.

TraceBERT-A Feasibility Study on Reconstructing Spatial-Temporal Gaps from Incomplete Motion Trajectories via BERT Training Process on Discrete Location Sequences.

Trajectory data represent an essential source of information on travel behaviors and human mobility patterns, assuming a central role in a wide range of services related to transportation planning, personalized recommendation strategies, and resource management plans. The main issue when dealing with trajectory recordings, however, is characterized by temporary losses in the data collection, causing possible spatial-temporal gaps and missing trajectory segments. This is especially critical in those use cases based on non-repetitive individual motion traces, when the user's missing information cannot be directly reconstructed due to the absence of historical individual repetitive routes. Inserted in the context of location-based trajectory modeling, we tackle the problem by proposing a technical parallelism with the natural language processing domain. Specifically, we introduce the use of the Bidirectional Encoder Representations from Transformers (BERT), a state-of-the-art language representation model, into the trajectory processing research field. By training deep bidirectional representations from unlabeled location sequences, jointly conditioned on both left and right context, we derive an explicit predicted estimation of the missing locations along the trace. The proposed framework, named TraceBERT, was tested on a real-world large-scale trajectory dataset of short-term tourists, exploring an effective attempt of adapting advanced language modeling approaches into mobility-based applications and demonstrating a prominent potential on trajectory reconstruction over traditional statistical approaches.

Dual-specificity phosphatase 15 (DUSP15) in the nucleus accumbens is a novel negative regulator of morphine-associated contextual memory.

Drug relapse among addicts often occurs due to the learned association between drug-paired cues and the rewarding effects of these drugs, such as morphine. Contextual memory associated with morphine has a central role in maintenance and relapse. We showed that morphine-conditioned place preference (CPP) activates extracellular-regulated protein kinase (ERK) in the nucleus accumbens (NAc). The main enzymes that mediate ERK dephosphorylation are members of the dual-specificity phosphatase (DUSP) superfamily. It is unclear which members regulate the morphine CPP-induced activation of ERK. After screening, DUSP15 was found to be decreased during both morphine CPP expression and the reinstatement period. Intra-NAc infusions of AAV-DUSP15 (overexpression) not only prevented the expression of morphine-induced CPP but also facilitated extinction, inhibited reinstatement, and abolished ERK activation. However, after repeated morphine exposure and withdrawal in mice, there was no change in the expression of p-ERK and DUSP15, and the overexpression of DUSP15 in the NAc did not improve the impaired spatial memory or anxiety-like behaviour induced by morphine. Together, these findings indicate that DUSP15 not only prevents the expression of drug-paired contextual memory but also promotes the extinction of existing addiction memories, thus providing a novel therapeutic target for the treatment of drug addiction.

Hybrid Control of Digital Baker Map with Application to Pseudo-Random Number Generator.

Dynamical degradation occurs when chaotic systems are implemented on digital devices, which seriously threatens the security of chaos-based cryptosystems. The existing solutions mainly focus on the compensation of dynamical properties rather than on the elimination of the inherent biases of chaotic systems. In this paper, a unidirectional hybrid control method is proposed to improve the dynamical properties and to eliminate the biases of digital chaotic maps. A continuous chaotic system is introduced to provide external feedback control of the given digital chaotic map. Three different control modes are investigated, and the influence of control parameter on the properties of the controlled system is discussed. The experimental results show that the proposed method can not only improve the dynamical degradation of the digital chaotic map but also make the controlled digital system produce outputs with desirable performances. Finally, a pseudorandom number generator (PRNG) is proposed. Statistical analysis shows that the PRNG has good randomness and almost ideal entropy values.

An Error Compensation Method for Improving the Properties of a Digital Henon Map Based on the Generalized Mean Value Theorem of Differentiation.

Continuous chaos may collapse in the digital world. This study proposes a method of error compensation for a two-dimensional digital system based on the generalized mean value theorem of differentiation that can restore the fundamental performance of chaotic systems. Different from other methods, the compensation sequence of our method comes from the chaotic system itself and can be applied to higher-dimensional digital chaotic systems. The experimental results show that the improved system is highly consistent with the real chaotic system, and it has excellent chaotic characteristics such as high complexity, randomness, and ergodicity.

Weighted gene co-expression network analysis to explore the mechanism of heroin addiction in human nucleus accumbens.

Heroin dependence is a complex behavioral disease, and a chronic encephalopathy with the important feature of relapse. The purpose of the study was to identify the regulatory mechanism of the nucleus accumbens (NAc) in heroin dependence. We used weighted gene co-expression network analysis to analyze the GSE87823 data package, which included 27 heroin users and 22 controls of human NAc tissue. Modules were correlated with basic information of samples and enrichment analyses used to identify biological function and transcription factors and online tools were used to perform the gene ontology of significant genes. We identified one gene module from the total data (blue) and the male data (turquoise), respectively. The overlap genes of top 10 hub genes in significant modules (PRR11, SLC35E1, LPP, ZNF721, ZNF611, LRRFIP1) were selected to identify as candidate genes in the regulation mechanism of NAc in heroin dependence. Then, we accorded the results to further explore that miRNA-hsa-miR-155-5p in male and total may be a potential marker. The candidate genes may serve as novel prognostic markers and treatment targets. Hsa-miR-155-5p may be a promising regulatory point for the treatment of heroin addiction.

Evaluation of the college-based HIV/AIDS education policy in Beijing, China: a mixed method approach.

BACKGROUND: From 2010 to 2015, there was a twofold growth of new HIV/AIDS infection in Beijing among young students aged 15-24. HIV/AIDS education was found effective in promoting positive behavior change related to HIV/AIDS prevention. However, little evidence was found on the evaluation of HIV/AIDS education policy. This study aimed to evaluate the college-based HIV/AIDS education policy in Beijing. METHODS: By using a mixed method approach, the current study reviewed college-based HIV/AIDS education policy at national level and in Beijing from 1985 to 2016 and conducted policy content analysis to evaluate the policy ability to structure implementation. Cross-sectional surveys in 2006 and 2016 were used to evaluate college's implementation of relevant policies. T test, χ2 test, and logistic regression were used to analyze college students' perception of HIV/AIDS education provided in their colleges and their knowledge of HIV/AIDS and their risk factors. RESULTS: Fourteen pieces of national policy and four pieces of Beijing's policy were identified. Policy's ability to structure implementation was at moderate level. The percentage of students in Beijing who ever perceived HIV/ADIS education at colleges decreased from 71.14 to 39.80%, and the percentage of students with comprehensive knowledge of HIV/AIDS dropped from 50.00% in 2006 to 40.42% in 2016. CONCLUSIONS: HIV/AIDS education in college had drawn considerable attentions from the Chinese government, while the policy implementation needs further strengthening.

Feature Selection via Swarm Intelligence for Determining Protein Essentiality.

Protein essentiality is fundamental to comprehend the function and evolution of genes. The prediction of protein essentiality is pivotal in identifying disease genes and potential drug targets. Since the experimental methods need many investments in time and funds, it is of great value to predict protein essentiality with high accuracy using computational methods. In this study, we present a novel feature selection named Elite Search mechanism-based Flower Pollination Algorithm (ESFPA) to determine protein essentiality. Unlike other protein essentiality prediction methods, ESFPA uses an improved swarm intelligence⁻based algorithm for feature selection and selects optimal features for protein essentiality prediction. The first step is to collect numerous features with the highly predictive characteristics of essentiality. The second step is to develop a feature selection strategy based on a swarm intelligence algorithm to obtain the optimal feature subset. Furthermore, an elite search mechanism is adopted to further improve the quality of feature subset. Subsequently a hybrid classifier is applied to evaluate the essentiality for each protein. Finally, the experimental results show that our method is competitive to some well-known feature selection methods. The proposed method aims to provide a new perspective for protein essentiality determination.

Influencing factors associated with the mode of birth among childbearing women in Hunan Province: a cross-sectional study in China.

BACKGROUND: An unnecessary Caesarean section (CS) can cause increased maternal and perinatal morbidity and other adverse short- and long-term outcomes. However, countries worldwide have witnessed an increasing trend toward the use of CS. Our objectives were to explore the influencing factors associated with the mode of birth among childbearing women in Hunan Province and to provide evidence and suggestions for the improvement and further understanding of vaginal birth (VB) in China. METHODS: A total of 977 childbearing women (375 pregnant women and 602 mothers of infants) were enrolled in this study using a two-stage cluster sampling method, and a self-administered questionnaire was used to collect data relating to the mode of birth. A t-test and χ (2)-test were used to analyse the differences between groups, and logistic regression analysis was used to explore the factors that influenced the mode of birth. RESULTS: The VB ratio was 46.2 %, while the CS ratio was 53.8 % in Hunan Province. Among women whose preference was VB, only 69.4 % gave birth by VB. Among women whose preference was CS, 98.1 % gave birth by CS. The top four reasons for preferring CS were a lack of confidence in VB (37.3 %), an abnormality in the prenatal examination (36.6 %), the notion that the baby would suffer fewer risks (34.8 %) and the fear of pain from VB (32.7 %). Age, prenatal examination, and doctors' suggestion were significantly associated with women's mode of birth preference, while place of household registration, husband's preference, prenatal examination and doctors' suggestion had a significant influence on women who changed their choice from VB to CS. CONCLUSIONS: The percentage of CS in Hunan was extremely high. Medical factors, such as abnormalities in prenatal examinations, and non-medical factors, such as a lack of confidence in VB, the fear of pain during VB, the desire to select the time of birth and healthy birth systems, should be seriously considered. Targeted health promotion interventions should be implemented to improve the performance of VB.

Effect of a school-based nutrition education program on adolescents' nutrition-related knowledge, attitudes and behaviour in rural areas of China.

OBJECTIVES: This study aimed at examining the effect of the nutrition education program on adolescents' knowledge, attitudes and behaviour in relation to nutrition in rural China. METHODS: A cluster-randomised intervention trial design was employed. Two middle schools were randomly selected and assigned to the school conducting nutrition education (NE school), or to the Control school, in Mi Yun County, Beijing. From each school 65 seventh-grade students were randomly selected to participate in the study. Nutrition-related knowledge, attitudes and behaviour were measured at pre- and post-intervention surveys with the same instrument. The nutrition education lasted for 6 months. RESULTS: After the intervention, more students in NE school knew the main function of dairy products and vegetables, which micronutrient is rich in dairy products and beans, and in meat, and the symptom of food poisoning, than those in Control school. The rate of students who thought nutrition is very important to health, and foods with an expired date should be thrown away in NE school was higher than that in Control school (93.8 vs. 80.3 and 92.3 vs. 78.7%, respectively). The rate of students who ate vegetables and breakfast everyday in NE school was higher than that in Control school (96.9 vs. 80.3, and 89.2 vs. 75.4%, respectively). (p < 0.05). CONCLUSIONS: This nutrition education programme is effective in improving adolescents' knowledge, attitudes, and behaviour in relation to nutrition; therefore, the nutrition education with interactive and innovative intervention components is strongly recommended for future nutrition promotion programmes for adolescents.

The effect of prenatal education curriculum on mother's prenatal examination utilization, delivery mode and recovery status: a cross-sectional survey in China.

OBJECTIVES: To examine the participation, implementation, and effect of the prenatal education curriculum provided by hospitals in China, and to provide evidence for the improvement of prenatal education. METHODS: A cross-sectional survey was conducted in the hospitals in Hunan Province, China. Mothers aged 20-45 years who had given birth between 1 May 2011 and 1 May 2012 and not diagnosed with pregnancy-related complications were invited to participate in the study. A self-administered, structured questionnaire was used to examine the effect of prenatal education curriculum on prenatal examination utilization, delivery mode, and recovery status from delivery. RESULTS: Among the total 604 respondents, only 175 (29.1 %) surveyed mothers participated in prenatal education curriculum provided by hospitals during their latest delivery. These mothers had a higher rate of attending all the required prenatal examinations (57.9 vs. 48.3 %), and a higher rate of recovering very well and well (80 vs. 73.7 %) from the latest delivery, than those who did not participate in prenatal education curriculum (P < 0.05). However, there was no statistical difference in the delivery mode between mothers who participated and those who did not participate in the prenatal education curriculum provided by hospitals. CONCLUSIONS: Prenatal education is indispensable for the improvement of maternal and child health, and thus should be advocated. In China, a standard and convenient specification prenatal education curriculum provided by hospitals and their doctors is appropriated for providing prenatal education to pregnant women.

A feasibility study of the association of exposure to biomass smoke with vascular function, inflammation, and cellular aging.

BACKGROUND: Biomass smoke at higher concentrations is associated with respiratory symptoms and, after years of exposure, increased risk of respiratory disorders in adults, but its effects on cardiovascular diseases are not well characterized, particularly compared with other pollution sources like tobacco smoke or traffic. METHODS: We conducted a cross-sectional study and enrolled 25 women living in rural Sichuan, China. We measured integrated 24-h personal exposure to fine particulate matter (PM2.5) and black carbon, and measured PM2.5 and black carbon in their kitchens. We assessed participants' brachial and central blood pressure and arterial stiffness using pulse wave analysis, and analyzed dried blood spot and buccal cell samples for C-reactive protein and relative telomere length. We also evaluated the difference in these physiological and biomarker measures between individuals with high (≥median) versus low (

The role of meal viscosity and oat ß-glucan characteristics in human appetite control: a randomized crossover trial.

BACKGROUND: Foods that enhance satiety can help consumers to resist environmental cues to eat, and improve the nutritional quality of their diets. Viscosity generated by oat ß-glucan, influences gastrointestinal mechanisms that mediate satiety. Differences in the source, processing treatments, and interactions with other constituents in the food matrix affect the amount, solubility, molecular weight, and structure of the ß-glucan in products, which in turn influences the viscosity. This study examined the effect of two types of oatmeal and an oat-based ready-to-eat breakfast cereal (RTEC) on appetite, and assessed differences in meal viscosity and ß-glucan characteristics among the cereals. METHODS: Forty-eight individuals were enrolled in a randomized crossover trial. Subjects consumed isocaloric breakfast meals containing instant oatmeal (IO), old-fashioned oatmeal (SO) or RTEC in random order at least a week apart. Each breakfast meal contained 218 kcal (150 kcal cereal, and 68 kcal milk) Visual analogue scales measuring appetite were completed before breakfast, and over four hours, following the meal. Starch digestion kinetics, meal viscosities, and ß-glucan characteristics for each meal were determined. Appetite responses were analyzed by area under the curve. Mixed models were used to analyze response changes over time. RESULTS: IO increased fullness (p = 0.04), suppressed desire to eat (p = 0.01) and reduced prospective intake (p < 0.01) more than the RTEC over four hours, and consistently at the 60 minute time-point. SO reduced prospective intake (p = 0.04) more than the RTEC. Hunger scores were not significantly different except that IO reduced hunger more than the RTEC at the 60 minute time-point. IO and SO had higher ß-glucan content, molecular weight, gastric viscosity, and larger hydration spheres than the RTEC, and IO had greater viscosity after oral and initial gastric digestion (initial viscosity) than the RTEC. CONCLUSION: IO and SO improved appetite control over four hours compared to RTEC. Initial viscosity of oatmeal may be especially important for reducing appetite.

Relationships of health literacy, health behavior, and health status regarding infectious respiratory diseases: application of a skill-based measure.

This study aimed to explain the relationships among health literacy, health behavior, and health status, using a newly developed skills-based measure of health literacy regarding respiratory infectious diseases. This instrument was designed to measure individuals' reading, understanding, and calculating ability, as well as their oral communication and Internet-based information-seeking abilities. A pilot survey was conducted with 489 residents in Beijing, China, to test the reliability and validity of the new measure. Next, a larger study with 3,222 residents in three cities with multistage stratified cluster sampling was implemented to validate a latent variable model (goodness of fit index=0.918, root mean square residual=0.076). In this model higher educational attainment (ß=0.356) and more health knowledge (ß=0.306) were positively and directly associated with greater health literacy skill, while age was negatively associated with it (ß=-0.341). Age (ß=0.201) and health knowledge (ß=0.246) had positive and direct relationship with health behavior, which was, in turn, positively associated with health status (ß=0.209). The results illustrate the complex relationships among these constructs and should be considered when developing respiratory intervention strategies to promote health behavior and health status.

Preventing HIV transmission in Chinese internal migrants: a behavioral approach.

This study is a step towards a behavioral intervention to prevent HIV transmission among Chinese internal migrants. To explore important and changeable determinants of condom use and inspect effective and feasible methods to increase condom use for the target population, we conducted a three-round web-based Delphi study among a panel of 62 experts between October 2012 and March 2013. The panelists were purposely selected using a stepwise procedure to represent topic-related areas of expertise. The response rate per round ranges from 21% to 81%. The panelists identified 19 possible determinants of condom use and reported 16 intervention methods they considered successful. They agreed that attitude towards condom use was the most important and changeable determinant, while applying behavioral theory, increasing sexual education and condom access, performing worksite health promotion, detecting risk factors, and working closely with relevant organizations and the government were effective and feasible methods to increase condom use among internal migrants in China. In conclusion, results of this study highlight the importance of attitude in changing condom use and underscore the need to apply behavior theory and integrate multiple educational approaches for developing behavioral HIV prevention interventions targeting internal migrants in China.

Knowledge, attitudes and behaviour regarding nutrition and dietary intake of seventh-grade students in rural areas of Mi Yun County, Beijing, China.

OBJECTIVES: To investigate the knowledge, attitudes and behaviour of seventh-grade students regarding nutrition and dietary intake, to collect data that would facilitate the design and implementation of interventions aimed at promoting good nutrition in adolescents via the school and to contribute to the improvement of adolescents' health in rural regions. METHODS: This was a descriptive, cross-sectional study in which a 27-item, self-administered, structured questionnaire, designed in Chinese, was administered to seventh-grade students attending one of three rural middle schools, chosen randomly from 15 middle schools, in Mi Yun County, Beijing. RESULTS: In terms of nutrition knowledge, 84.5 % of respondents had scores of between 3 and 7. Based on our analysis of the completed questionnaires, during the 7 days preceding completion of the questionnaire 43.6 % of respondents had eaten fresh fruit at least once per day, 71.3 % had eaten vegetables at least once per day, 22.3 % had consumed at least one glass of milk (or soy milk or yogurt) each day and 68.6 % had eaten breakfast every day. In total, 95.2 % of respondents thought nutrition was very important or important to their health, and 94.2 % thought it was very important or important to develop healthy dietary habits. Of the respondents, 52.7 % thought good food quality was the most important aspect of nutrition and dietary intake and 62.2 % thought food safety was the most important aspect. CONCLUSIONS: Based on the results of our survey we suggest their should be continuous efforts to teach adolescents in rural areas of China about various aspects of good nutrition, to encourage diversity in their choice of breakfasts, to guarantee food safety and to improve food quality.

Gradient Inversion Attacks: Impact Factors Analyses and Privacy Enhancement.

Gradient inversion attacks (GIAs) have posed significant challenges to the emerging paradigm of distributed learning, which aims to reconstruct the private training data of clients (participating parties in distributed training) through the shared parameters. For counteracting GIAs, a large number of privacy-preserving methods for distributed learning scenario have emerged. However, these methods have significant limitations, either compromising the usability of global model or consuming substantial additional computational resources. Furthermore, despite the extensive efforts dedicated to defense methods, the underlying causes of data leakage in distributed learning still have not been thoroughly investigated. Therefore, this paper tries to reveal the potential reasons behind the successful implementation of existing GIAs, explore variations in the robustness of models against GIAs during the training process, and investigate the impact of different model structures on attack performance. After these explorations and analyses, this paper propose a plug-and-play GIAs defense method, which augments the training data by a designed vicinal distribution. Sufficient empirical experiments demonstrate that this easy-toimplement method can ensure the basic level of privacy without compromising the usability of global model.