Objective To develop a deep learning model based on fundus retinal images to improve the detection rate of mild cognitive impairment(MCI)in patients with coronary heart disease,achieve early intervention and improve prognosis.Methods The study was a single-center cross-sectional study that retrospectively included patients diagnosed with coronary heart disease(CHD)by coronary angiography(≥50％ stenosis of at least one coronary vessel)from Beijing Anzhen Hospital between November 2021 and December 2022.The whole data set was randomly divided into the training set and the testing set according to the ratio of 8∶2 for model development.After that,the patient data of the same center from January 2023 to April 2023 were included in the time verification method to verify the model.The diagnostic criteria for MCI were MMSE＜27 or MoCA＜26.Four kinds of convolutional neural network(CNN)architectures were used to train fundus images,and a comprehensive vision model of MCI detection was established through model integration.The area under the curve(AUC),sensitivity and specificity of the receiver operating curve(ROC)were used to evaluate the performance of the AI model.Results We collected 5 880 eligible fundus images from 3 368 CHD patients.Based on the results of the MMSE scale,the algorithm was labeled,including 2 898 males and 527 MCI patients.The AUC of the deep learning model in the test group is 0.733(95％CI 0.688-0.778),and the sensitivity of the algorithm in the test group is 0.577(95％CI 0.528-0.625)by using the operating point with the maximum sum of sensitivity and specificity.With a specificity of 0.758(95％CI 0.714-0.802),corresponding to a validated AUC of 0.710(95％CI 0.601-0.818).Based on the results of the MoCA scale,the algorithm labels 2 437 males and 1 626 MCI patients.The AUC of the deep learning model in the test group was 0.702(95％CI 0.671-0.733).The operating point with the maximum sum of sensitivity and specificity was selected,and the sensitivity of the algorithm was 0.749(95％CI 0.719-0.778)and the specificity was 0.561(95％CI 0.527-0.595),corresponding to the AUC value of the verification group was 0.674(95％CI 0.622-0.726).Conclusions The deep learning algorithm model based on fundus images has good diagnostic performance,and may be used as a new non-invasive,convenient and rapid screening method for MCI in CHD population.

Objective To evaluate the influence of the wearing position of dosimeters outside lead aprons on effective dose estimation for interventional radiology workers, analyze the differences between single and double dosimeter methods in effective dose estimation, and provide a reference for the personal dose monitoring of interventional radiology workers. Methods This study employed a combined approach of on-site monitoring and Monte Carlo simulation to evaluate the impact of the wearing position of dosimeters outside lead aprons on effective dose estimation, as well as the differences between effective doses measured using single and double dosimeters. Interventional radiology workers wore dosimeters at three positions: the neck outside the lead collar, the left chest outside the lead apron, and inside the lead apron. Effective doses were estimated using the single and double dosimeter methods specified in GBZ 128-2019 Specifications for individual monitoring of occupational external exposure, and the impact of different wearing positions on the estimation results was compared. Geant4 Monte Carlo simulations were used to model dose distributions at the neck outside the lead collar and at the left chest outside the lead apron for operators performing cardiovascular interventions under tube voltages of 70, 80, 90, and 100 kVp and exposure angles of posteroanterior (PA), anteroposterior (AP), and left anterior oblique 45° (LAO45°) positions. The study assessed the impact of dosimeter wearing position on effective dose estimation. Results Monte Carlo simulations demonstrated that neck doses consistently exceeded left chest doses across different tube voltages and exposure angles, with neck-to-chest dose ratios of 0.80-0.90. Under identical tube voltage conditions, AP showed the highest doses, followed by LAO45°, and PA demonstrated the lowest doses. The single and double dosimeter methods exhibited consistent patterns in effective dose estimation. Single dosimeter method generally yielded higher effective doses with relative deviations of 9.9% to 83%, though these deviations decreased under high tube voltages. Field monitoring data indicated that most interventional radiology workers maintained relative deviations between single and double dosimeter calculations below 6%, with neck-to-chest dose ratios of 0.95-1.1. The estimation patterns remained consistent across both methods, though single dosimeter method showed slightly higher results. Conclusion Under PA, AP, or LAO45°, the doses at the neck consistently exceeded those at the left chest. Therefore, when wearing lead protective equipment, the dosimeter should be properly positioned at the neck outside the lead collar to accurately reflect the radiation doses of surgeons. Some interventional radiology workers improperly positioned the dosimeter (intended at the neck outside the lead collar) at the left chest outside the lead apron, and this may result in an underestimation of the effective dose.

With the rapid development of generative artificial intelligence(GAI)technologies,their widespread application in academic research and writing is continuously expanding the boundaries of sci-entific inquiry.However,this trend has also raised a series of ethical and regulatory challenges,inclu-ding issues related to authorship,content authenticity,citation accuracy,and accountability.In light of the growing involvement of AI in generating academic content,establishing an open,controllable,and trustworthy ethical governance framework has become a key task for safeguarding research integrity and maintaining trust within the academic community.This expert consensus outlines ethical requirements across key stages of AI-assisted academic writing-including topic selection,data management,citation practices,and authorship attribution.It aims to clarify the boundaries and ethical obligations surrounding AI use in academic writing,ensuring that technological tools enhance efficiency without compromising in-tegrity.The goal is to provide guidance and institutional support for building a responsible and sustainable research ecosystem.

With the rapid development of generative artificial intelligence(GAI)technologies,their widespread application in academic research and writing is continuously expanding the boundaries of sci-entific inquiry.However,this trend has also raised a series of ethical and regulatory challenges,inclu-ding issues related to authorship,content authenticity,citation accuracy,and accountability.In light of the growing involvement of AI in generating academic content,establishing an open,controllable,and trustworthy ethical governance framework has become a key task for safeguarding research integrity and maintaining trust within the academic community.This expert consensus outlines ethical requirements across key stages of AI-assisted academic writing-including topic selection,data management,citation practices,and authorship attribution.It aims to clarify the boundaries and ethical obligations surrounding AI use in academic writing,ensuring that technological tools enhance efficiency without compromising in-tegrity.The goal is to provide guidance and institutional support for building a responsible and sustainable research ecosystem.

Objective To develop a deep learning model based on fundus retinal images to improve the detection rate of mild cognitive impairment(MCI)in patients with coronary heart disease,achieve early intervention and improve prognosis.Methods The study was a single-center cross-sectional study that retrospectively included patients diagnosed with coronary heart disease(CHD)by coronary angiography(≥50％ stenosis of at least one coronary vessel)from Beijing Anzhen Hospital between November 2021 and December 2022.The whole data set was randomly divided into the training set and the testing set according to the ratio of 8∶2 for model development.After that,the patient data of the same center from January 2023 to April 2023 were included in the time verification method to verify the model.The diagnostic criteria for MCI were MMSE＜27 or MoCA＜26.Four kinds of convolutional neural network(CNN)architectures were used to train fundus images,and a comprehensive vision model of MCI detection was established through model integration.The area under the curve(AUC),sensitivity and specificity of the receiver operating curve(ROC)were used to evaluate the performance of the AI model.Results We collected 5 880 eligible fundus images from 3 368 CHD patients.Based on the results of the MMSE scale,the algorithm was labeled,including 2 898 males and 527 MCI patients.The AUC of the deep learning model in the test group is 0.733(95％CI 0.688-0.778),and the sensitivity of the algorithm in the test group is 0.577(95％CI 0.528-0.625)by using the operating point with the maximum sum of sensitivity and specificity.With a specificity of 0.758(95％CI 0.714-0.802),corresponding to a validated AUC of 0.710(95％CI 0.601-0.818).Based on the results of the MoCA scale,the algorithm labels 2 437 males and 1 626 MCI patients.The AUC of the deep learning model in the test group was 0.702(95％CI 0.671-0.733).The operating point with the maximum sum of sensitivity and specificity was selected,and the sensitivity of the algorithm was 0.749(95％CI 0.719-0.778)and the specificity was 0.561(95％CI 0.527-0.595),corresponding to the AUC value of the verification group was 0.674(95％CI 0.622-0.726).Conclusions The deep learning algorithm model based on fundus images has good diagnostic performance,and may be used as a new non-invasive,convenient and rapid screening method for MCI in CHD population.

Objective To analyze the cognition level of health care workers(HCWs)and the management status of various levels of medical institutions towards allergy reactions to commonly used antimicrobial agents.Methods HCWs and clinical pharmacists who were related to the diagnosis and treatment of antimicrobial agents in 14 medical institutions of city-level and autonomous prefectures in Gansu Province were randomly selected for a questionnaire survey.The survey contents included respondents'basic information,criteria for judging antimicrobial allergy,awareness on procedures related to antimicrobial allergy,and antimicrobial management level of different levels of medical institutions.Results A total of 8 670 valid questionnaires from HCWs were collected,including 3 300 phy-sicians,5 024 nurses and 328 pharmacists.160,775,2 123 and 5 612 HCWs were with senior,associate,interme-diate and junior professional titles,respectively.87.66％of the HCWs received relevant training on antimicrobial management in the past two years,the proportion of HCWs from different levels of medical institutions who have received training on antimicrobial management in the past two years was statistically significant different(x2=42.668,P＜0.001).HCWs with senior professional titles had the highest proportion of receiving relevant training(93.75％),there was a statistically significant difference in the proportion of receiving antimicrobial management training among HCWs with different professional titles in the past two years(x2=69.782,P＜0.001).50.98％of HCWs were not clear about penicillin allergy,and most of whom were with junior professional titles,accounting for 68.52％.25.19％of HCWs expressed uncertainty about whether patients with penicillin allergy could use cephalosporins,225 of whom were with associate professional titles,accounting for 29.03％of the total number of HCWs with associate profe-ssional titles.6.11％of HCWs had no experience in skin test procedure;46.94％of HCWs expressed that their medical institutions had no or unclear about whether their medical institutions had an antimicrobial allergy assess-ment team.Conclusion HCWs'judgment on allergy reactions to commonly used antimicrobial agents and aware-ness on antimicrobial application is not high enough,and the overall management level of antimicrobial allergy in all levels of medical institutions is poor.The popularity of antimicrobial allergy assessment teams is not high,and there is an urgent need to strengthen supervision,management,training,et al.

Objective To retrieve and extract the best evidence for preventing intracranial infections related to ex-ternal cerebrospinal fluid(CSF)drainage,and provide evidence-based support for reducing the incidence of intracra-nial infection caused by external CSF drainage.Methods Evidence-based care issues were determined according to PIPOST,and the best evidence on intracranial infection related to external CSF drainage tube was retrieved from top to bottom.The literature retrieval period was 2013-2023.Quality control of the literatures,as well as extraction and summary of the evidence were carried out by 2 trained graduate students.Results A total of 17 literatures were included in the analysis,including 3 guidelines,5 expert consensus,8 systematic reviews,and 1 randomized con-trolled trial.Management strategies from 3 dimensions(pre-catheterization,in-catheterization and post-catheteriza-tion)were obtained,including 20 pieces of evidence for preventing intracranial infection,such as preparation for ex-ternal CSF drainage tube,precautions during catheterization,and post-catheterization disposal.Conclusion There are differences in the management of external CSF drainage tube in clinical practice.It is necessary to develop uni-fied,standardized,and rational bundle strategies to prevent intracranial infection,so as to reduce the incidence of catheter-related intracranial infection.

AIM To investigate the renoprotective effects and mechanism of Caragana sinica roots,Astragali Radix and their combination use on the rat model of diabetic kidney disease(DKD).METHODS Sixty SD rats were randomly divided into the normal group,the model group,the Empagliflozin group(10 mg/kg),the C.sinica roots group(3.1 g/kg),the Astragali Radix group(3.1 g/kg),and the C.sinica roots plus Astragali Radix group(6.2 g/kg).In contrast to the intact rats of the normal group,rats of the other groups underwent left nephrectomy and intraperitoneal injection of streptozotocin(STZ)followed by 8-week intragastric gavage of the corresponding agent,during which their levels of FBG and 24 h urinary microprotein(24 h U-mAlb)were detected regularly.The rats killed at the end of the trial had their levels of Scr,BUN and Cystatin C detected;their renal pathological changes observed by HE,PAS and Masson stainings;their expressions of macrophage marker proteins CD68 and iNOS detected by immunohistochemistry;their expressions of renal JNK/SAPK pathway proteins such as JNK,p-JNK,TNF-α,IL-1β and ICAM-1 detected by Western blot;and their serum levels of TNF-α,IL-1β and ICAM-1 detected by ELISA as well.RESULTS Compared with the normal group,the model group displayed increased levels of FBG,24 h U-mAlb,BUN,Scr and Cystatin C(P＜0.01);more renal pathological damage,and increased levels of TNF-α,IL-1β and ICAM-1 in the renal tissue and serum(P＜0.01);and increased renal protein expressions of JNK and p-JNK(P＜0.01).Compared with the model group,all of the groups intervened with an agent shared decreased levels of FBG,24 h U-mAlb,BUN,Scr and Cystatin C(P＜0.05,P＜0.01);alleviated renal pathological damage,and decreased levels of TNF-α,IL-1β and ICAM-1 in renal tissue and serum(P＜0.01).There existed no group difference between the Astragali Radix group and the C.sinica roots group in terms of all indices levels(P＞0.05).The C.sinica roots plus Astragali Radix group demonstrated its superiority over either C.sinica roots group or Astragali Radix group in terms of all the indices levels(P＜0.05,P＜0.01).CONCLUSION C.sinica roots,Astragali Radix or their combination use can alleviate the renal pathological damage and improve the renal function of DKD rats through inhibiting the M1 macrophages,reducing the secretion of inflammatory factors,whose mechanism may lie in the inhibition of JNK/SAPK signal pathway activation.A better effect can be anticipated by the combination use of C.sinica roots and Astragali Radix.

AIM To investigate the impact of the combination use of Caragana sinica Radix and Astragali Radix on a rat model of diabetic kidney disease(DKD).METHODS The SD rats were randomly divided into the normal group,the model group,the Engelgin group,the Caragana sinica Radix group,the Astragali Radix group and the Caragana sinica Radix-Astragali Radix compatibility group,with 10 rats in each group.Following the successful establishment of a DKD model by unilateral amputate renal combined with intraperitoneal injection of streptozotocin(STZ),the corresponding gastric gavage of drugs were administered for 8 weeks.The rats had their 24 h urinary microalbumin(24 h U-mALB)detected at 0,4 and 8 weeks;their levels of Scr,BUN,CysC,MDA and SOD activity detected by ELISA;their renal ROS expression detected by fluorescence probe method;their renal pathological changes observed by HE,PAS,Masson and PASM-Masson staining;their renal expressions of NOX4,Drp1,MFN2 and P62 detected by immunohistochemistry;and their renal expressions of PINK1,MFN2,Parkin,LC3-Ⅱ/Ⅰ,P62 and p-Drp1 proteins detected by Western blot.RESULTS Compared with the model group,each treatment group displayed lower contents of 24 h U-mALB,BUN,Scr and CysC in the serum of rats(P＜0.01);reduced pathological structure damage of the renal tissue;decreased MDA level in serum and kidney(P＜0.01);increased SOD activity(P＜0.01);increased renal protein expressions of PINK1,MFN2,Parkin and LC3-Ⅱ/Ⅰ(P＜0.05,P＜0.01);and decreased protein expressions of p-Drp1 and P62(P＜0.01).And the Astragali Radix group and the Caragana sinica Radix-Astragali Radix compatibility group took the lead(P＜0.05,P＜0.01).CONCLUSION Upon the rat model of DKD,the compatibility of Caragana sinica Radix and Astragali Radix may alleviate their renal pathological damage and improve their renal function by activating the mitochondrial autophagy to improve mitochondrial dynamics and inhibit their oxidative stress via PINK1/MFN2/Parkin pathway.

Objective To investigate the value of multi-phase MRI-based radiomics machine learning models in differentiating small renal cell carcinoma(sRCC)from fat-poor renal angiomyolipoma(fp-AML).Methods 79 cases of sRCCs and 35 cases of fp-AMLs(diameter≤4 cm)which were confirmed by pathology were retrospectively analyzed.The volume of interest(VOI)of the total tumor was manually delineated on the images of T2WI(T2),unenhanced phase(UP),corticomedullary phase(CMP)and nephrographic phase(NP)and then the radiomics of the VOIs were extracted respectively.The training set and the test set were set according to the ratio of 7∶3.The t-test,maximal relevance and minimal redundancy(mRMR)and the least absolute shrinkage and selection operator(LASSO)were used to select the radiomics features.The selected features were used to build classification models with logistic regression(LR)and support vector machine(SVM).The receiver operating characteristic(ROC)curve was used to evaluate the classification performances of the models.Results There were 4,12,3,11 and 15 optimal features obtained from T2、UP、CMP、NP and the combined four phases,respectively.The radiomics features based on NP or the combined four phases with LR model performed best,AUCs were respectively 0.956,0.986 in the training set and both were 0.881 in the test set.Conclusion The multi-phase MRI-based radiomics machine learning model has favorable diagnostic performance in differentiating sRCC from fp-AML.