Objective: To investigate the prevalence of undernutrition and its associated factors among left behind and non left behind primary and secondary school students in the Nutrition Improvement Program for Rural Compulsory Education Students (NIPRCES) areas of central and western China, so as to provide evidence for improving the nutritional status of children and adolescents. Methods: A survey was conducted among 123 782 students selected by random cluster sampling method in grades 3-9 from NIPRCES in central (Hebei, Shanxi, Heilongjiang, Jilin, Anhui, Jiangxi, Henan, Hunan, Hubei, and Hainan) and western (Gansu, Guangxi, Inner Mongolia, Ningxia, Tibet, Shaanxi, Guizhou, Sichuan, Xinjiang, the Xinjiang Production and Construction Corps, Yunnan, Qinghai, and Chongqing) China in 2023. Anthropometric measurements and questionnaires were used to assess nutritional and dietary status. The prevalence of undernutrition was compared between left behind and non left behind students by Chi square test, and associated factors were analyzed by three level Logistic mixed effects model. Results: The prevalence of undernutrition was 8.5% (4 326) in left behind students and 8.1% (5 905) in non left behind students. Three level Logistic mixed effect model analysis showed that whether left behind or non left behind, the undernutrition rates of primary and secondary students in western regions were higher than those of students in central regions [ OR (95% CI )=1.72(1.57-1.87),2.25(2.07- 2.43 )]; the undernutrition risk was lower for those whose fathers had a cultural level of high school or above [ OR (95% CI )=0.69(0.62-0.77),0.90(0.82-0.98)] or junior high school [ OR (95% CI )=0.72(0.66-0.79),0.92(0.85-0.99)] compared to those with primary school or below; picky eating or selective eating increased the risk of undernutrition [ OR (95% CI )=2.36(2.07-2.68),2.28(2.04-2.55)], and primary and secondary school students without nutritional content in health education classes had higher rates of undernutrition [ OR (95% CI )=1.12(1.03-1.23),1.09(1.01-1.17)](all P <0.05). Conclusion The prevalence of undernutrition is slightly higher in left behind primary and secondary students than in non left behind primary and secondary students in central and western NIPRCES areas, with variations across different characteristics.

Objective: To analyze the trends of the frequency of meat, egg, and milk consumption among rural primary and junior high school students in central and western China covered by the Nutrition Improvement Program for Rural Compulsory Education Students (NIPRCES) from 2015 to 2023, so as to provide basis for formulating more targeted nutrition intervention policies and health education strategies. Methods: Using data from six rounds of monitoring and evaluation (2015-2021 and 2023), the study included 323 870 students from grade 3 to 9 across 22 provinces (autonomous regions and municipalities) in central and western China. The consumption frequencies of meat, egg, and milk over the past week were collected via questionnaires. The Cochran-Armitage trend test was used to analyze temporal trends, and multivariable Logistic regression models were employed to analyze factors associated with the frequency of meat, egg and milk consumption and to test for interaction effects between the year and gender, region, and grade level. Results: From 2015 to 2023, the proportion of students consuming meat, egg, and milk ≥1 time/day increased from 23.20 %, 10.71%, and 0.74% to 35.53%, 22.09%, and 26.63%, respectively. Trend tests indicated a significant upward trend for the daily intake of all three food categories for meat, egg and milk over the years ( Z =67.18, 64.90, 93.14, all P <0.01). Multivariable Logistic regression analysis showed that the daily meat intake was lower in the central region than in the western region ( OR=0.77, 95%CI =0.76-0.78), whereas the daily intake of eggs ( OR=1.19, 95%CI =1.17-1.22) and milk ( OR= 1.27 , 95%CI =1.24-1.29) was higher in the central region (all P <0.05). Compared with grade 3-4 students, junior high school students had lower daily intake of meat, eggs, and milk≥1 time/day ( OR =0.95, 0.77, 0.77, all P <0.05), with a declining trend as grade increased. Girls also had lower daily intake of meat, eggs, and milk ≥1 time/day than boys ( OR =0.95，0.93，0.91, all P < 0.05). Significant interactions were observed between year and region, as well as between year and grade (all P <0.05). Conclusion From 2015 to 2023, the NIPRCES improved the intake level of among rural students, but the situation of relatively insufficient intake of egg and milk among females, junior high school students and those in the western region still exists.

Objective: To quantitatively assess the health risk of preservatives from beverages around primary schools in Anshun City, and to provide scientific basis for precise food safety supervision. Methods: From December 2023 to July 2024, 602 beverage samples were randomly collected from within 100 meters of 19 primary schools in Anshun City. The content of benzoic acid, sorbic acid, and dehydroacetic acid was detected according to GB 5009 series standards. Combined with children s physiological parameters (body weight 30 kg, daily intake 0.15 L), the Hazard Quotient (HQ) and Hazard Index (HI) models were used to evaluate health risks. Results: The total detection rate of preservatives from beverages around primary schools was 63.0%, and the total over limit rate was 9.0%. The detection rate of preservatives in flavored beverages was the highest (72.6%), and the highest over limit rate of preservatives in special purpose beverages was the highest (17.2%). The single preservative HQ (benzoic acid up to 0.47 ) and mixed HI (up to 0.55) of all samples were below 1(safety threshold). However, the HQ value of benzoic acid in flavored beverages (0.47) was 2.9 times that of sorbic acid (0.16), contributing significantly to health risk. Sensitivity analysis showed that if the daily consumption increased to 0.3 L, the HI value of flavored beverages would rise to 1.11, exceeding the safety threshold. Enterprise scale analysis showed that the exceedance rate of special purpose beverages in large enterprises reached 30.0%, while micro enterprises, accounting for a dominant market share (52.2%), constituted the main source of children s daily exposure to their products. Conclusions The overall health risk of perservatives in beverages sold near primary schools in Anshun City is controllable, but there is a noticeable risk of gradient. The risk of children’s exposure to preservatives through beverage consumption should not be ignored.

Objective To investigate the impact of red blood cell suspension infusion across various perioperative periods on patients with valvular heart disease. Methods The patients with valvular heart disease admitted to Tianjin Chest Hospital from 2018 to 2020 were selected. Based on the timing of perioperative red cell suspension infusion, patients were categorized into three groups: a group 1 receiving intraoperative red cell suspension infusion, a group 2 receiving red cell suspension infusion within 24 hours after entering the ICU, and a group 3 receiving red cell suspension infusion at both time points. The laboratory results, perioperative blood component infusion volume, and other relevant parameters were retrospectively analyzed. After propensity score matching, the differences in different variables among the three groups were compared. Results After propensity score matching, 102 patients were enrolled, including 52 males and 50 females, with an average age of (61.74±10.58) years. There were 34 patients in each group. The preoperative hemoglobin (Hb) value of the group 2 was significantly higher than that of the group 1 and the group 3, and the amount of red cell suspension and autoblood transfusion was the lowest (P<0.05). Group 1 had the highest postoperative Hb, as well as the highest Hb and hematocrit (HCT) levels within 24 hours post-surgery (P<0.05). The group 1 had the lowest plasma, platelet and cryoprecipitate infusion volumes, and the shortest cardiopulmonary bypass time, aortic occlusion time, postoperative ICU stay and hospital stay, and the least blood loss and total drainage volume (P<0.05). The difference between postoperative and preoperative Hb (△Hb1) was highest in group 1 (P<0.05). Conclusion For patients with valvular heart disease, intraoperative-only infusion of red blood cell suspension is associated with a better prognosis at discharge and during follow-up.

To evaluate long-term survival and clinical outcomes of patients with knee osteo-arthritis undergoing total knee arthroplasty (TKA) through long-term follow-up.

Objective:To explore the construction of an evaluation model for aortic root anatomy and calcium burden in patients with bicuspid aortic valve (BAV) stenosis before transcatheter aortic valve replacement (TAVR) based on deep learning (DL) algorithms.Methods:A retrospective collection of 362 BAV stenosis patients who underwent TAVR from September 2023 to May 2024 was performed. All patients underwent cardiac CT angiography. The patients were divided into training group ( n=104), internal validation group ( n=206), and external validation group ( n=52). A DL model was trained on the training dataset to assess aortic root anatomy and calcification burden. The evaluation included the segmentation accuracy of the algorithm, the measurement performance of key anatomical structures (i.e., valve leaflets and type-1 and type-2 fusion raphe), and calcification burden, as well as the measurement efficiency. Overall segmentation performance was assessed using the average Dice coefficient (ADC). The fine-scale segmentation quality was validated by the 95th-percentile Hausdorff distance (HD-95) and the average symmetric surface distance (ASSD). The consistency of the measurement results was assessed using the Pearson correlation coefficient and the intraclass correlation coefficient ( ICC) with a two-way mixed model for absolute agreement. In addition, the total time and total mouse movement distance required for manual assessment versus the DL model on the validation datasets were recorded and compared. Results:The algorithm demonstrated excellent segmentation performance on aortic root anatomical targets, achieving outstanding consistency within both internal and external validation datasets (0.955

Objective:To develop a deep learning model based on 3D super-resolution reconstruction technology and to analyze its feasibility and effectiveness in predicting paraglottic space invasion in hypopharyngeal cancer.Methods:A retrospective study was conducted involving 382 patients with hypopharyngeal squamous cell carcinoma treated at Qilu Hospital of Shandong University between January 2014 and December 2020. The cohort included 364 males and 18 females, with a mean age of 62±7 years. Patients were divided into a training set ( n=300) and a test set ( n=82) based on enrollment time. A generative adversarial network was used to perform 3D super-resolution reconstruction on contrast-enhanced CT images, improving spatial resolution by 16 times. A 2.5D deep learning strategy was employed to construct Resnet-NR and Resnet-SR models based on conventional and super-resolution images, respectively, to predict whether the paraglottic space was invaded. Model performance was evaluated using receiver operating characteristic (ROC) curves and area under the curve (AUC). A multi-reader multi-case study was conducted to assess the impact of the artificial intelligence (AI) model on clinicians′ diagnostic capabilities. Results:The super-resolution model Resnet-SR achieved the highest accuracy in both the training set (AUC=0.87, 95% CI: 0.84-0.90) and the test set (AUC=0.88, 95% CI: 0.81-0.96), significantly outperforming traditional clinical indicators (T stage, N stage, tumor diameter, and pathological differentiation degree) (AUC range: 0.55-0.70, all P<0.05). In comparison, the conventional-resolution model Resnet-NR achieved AUCs of 0.81 (95% CI: 0.77-0.84, P=0.005) and 0.80 (95% CI: 0.71-0.89, P=0.184) in the training and test sets, respectively. Using Resnet-SR to assist clinical decision-making improved the diagnostic accuracy of junior physicians (AUC=0.793 without AI assistance vs. AUC=0.871 with AI assistance, P=0.012) and significantly reduced diagnosis time for clinicians of all experience levels (86.5 s without AI assistance vs. 82.5 s with AI assistance, t=2.01, P=0.032). Conclusion:This study successfully develops a deep learning model based on 3D super-resolution reconstruction technology, which can assist in preoperative prediction of paraglottic space invasion in hypopharyngeal cancer. The AI-assisted tool improves diagnostic accuracy for junior physicians and enhances diagnostic efficiency for clinicians across all experience levels.

Objective:To explore the construction of an evaluation model for aortic root anatomy and calcium burden in patients with bicuspid aortic valve (BAV) stenosis before transcatheter aortic valve replacement (TAVR) based on deep learning (DL) algorithms.Methods:A retrospective collection of 362 BAV stenosis patients who underwent TAVR from September 2023 to May 2024 was performed. All patients underwent cardiac CT angiography. The patients were divided into training group ( n=104), internal validation group ( n=206), and external validation group ( n=52). A DL model was trained on the training dataset to assess aortic root anatomy and calcification burden. The evaluation included the segmentation accuracy of the algorithm, the measurement performance of key anatomical structures (i.e., valve leaflets and type-1 and type-2 fusion raphe), and calcification burden, as well as the measurement efficiency. Overall segmentation performance was assessed using the average Dice coefficient (ADC). The fine-scale segmentation quality was validated by the 95th-percentile Hausdorff distance (HD-95) and the average symmetric surface distance (ASSD). The consistency of the measurement results was assessed using the Pearson correlation coefficient and the intraclass correlation coefficient ( ICC) with a two-way mixed model for absolute agreement. In addition, the total time and total mouse movement distance required for manual assessment versus the DL model on the validation datasets were recorded and compared. Results:The algorithm demonstrated excellent segmentation performance on aortic root anatomical targets, achieving outstanding consistency within both internal and external validation datasets (0.955

Objective:To develop a deep learning model based on 3D super-resolution reconstruction technology and to analyze its feasibility and effectiveness in predicting paraglottic space invasion in hypopharyngeal cancer.Methods:A retrospective study was conducted involving 382 patients with hypopharyngeal squamous cell carcinoma treated at Qilu Hospital of Shandong University between January 2014 and December 2020. The cohort included 364 males and 18 females, with a mean age of 62±7 years. Patients were divided into a training set ( n=300) and a test set ( n=82) based on enrollment time. A generative adversarial network was used to perform 3D super-resolution reconstruction on contrast-enhanced CT images, improving spatial resolution by 16 times. A 2.5D deep learning strategy was employed to construct Resnet-NR and Resnet-SR models based on conventional and super-resolution images, respectively, to predict whether the paraglottic space was invaded. Model performance was evaluated using receiver operating characteristic (ROC) curves and area under the curve (AUC). A multi-reader multi-case study was conducted to assess the impact of the artificial intelligence (AI) model on clinicians′ diagnostic capabilities. Results:The super-resolution model Resnet-SR achieved the highest accuracy in both the training set (AUC=0.87, 95% CI: 0.84-0.90) and the test set (AUC=0.88, 95% CI: 0.81-0.96), significantly outperforming traditional clinical indicators (T stage, N stage, tumor diameter, and pathological differentiation degree) (AUC range: 0.55-0.70, all P<0.05). In comparison, the conventional-resolution model Resnet-NR achieved AUCs of 0.81 (95% CI: 0.77-0.84, P=0.005) and 0.80 (95% CI: 0.71-0.89, P=0.184) in the training and test sets, respectively. Using Resnet-SR to assist clinical decision-making improved the diagnostic accuracy of junior physicians (AUC=0.793 without AI assistance vs. AUC=0.871 with AI assistance, P=0.012) and significantly reduced diagnosis time for clinicians of all experience levels (86.5 s without AI assistance vs. 82.5 s with AI assistance, t=2.01, P=0.032). Conclusion:This study successfully develops a deep learning model based on 3D super-resolution reconstruction technology, which can assist in preoperative prediction of paraglottic space invasion in hypopharyngeal cancer. The AI-assisted tool improves diagnostic accuracy for junior physicians and enhances diagnostic efficiency for clinicians across all experience levels.

OBJECTIVE To explore the effect of failure mode and effect analysis(FMEA)on management of hospi-tal-associated infections(HAIs)in hemodialysis center.METHODS In Nov.2023,the risk priority number(RPN)integrated with action priority(AP)was adopted to identify,analyze and evaluate the risk factors in man-agement of HAIs in hemodialysis center of Tongji Hospital Affiliated to Tongji Medical College,Huazhong Uni-versity of Science and Technology by FMEA method.The high risk points that needed to be taken interventions were screened out,and the targeted measures were formulated to control the risks.At the end of the intervention period,a second round of risk assessment was carried out for improvement status of the high-risk points in Nov.2024,and the effect on the management of HAIs was evaluated.RESULTS The risk assessment was carried out for 48 risk points covering eight aspects,including organizational structure,self-inspection and supervision,staff management,environmental layout,cleaning and disinfection,surveillance,operation procedures and i-tem management.There were 9 risk points with the RPN values greater than 125,3 of which were with the AP value of"H".There were 8 risk points with the RPN value less than 125 and 6 risk points with the AP value drop-ping down to L after the targeted intervention measures were taken,indicating that the risk management has a-chieved favorable effect.CONCLUSIONS The RPN and AP integrated with FMEA can accurately identify the high-risk points in the quality management of the hemodialysis center.It is necessary to take targeted interven-tion measures so as to boost the effect on prevention and control of HAIs in the hemodialysis center and reduce the risk of HAIs in the hemodialysis patients.