Objective:To describe the hotspots and application trends of artificial intelligence (AI) in epidemiology in the past decade and analyze its advantages and challenges.Methods:The literatures with AI and epidemiology related keywords were systematically retrieved from Web of Science and China National Knowledge Infrastructure from 2014 to 2024. CiteSpace was used for bibliometric analysis of publication volume, keyword co-occurrence, clustering, emergence and cited literature co-occurrence analysis.Results:A total of 5 389 English papers and 1 659 Chinese papers were included, showing an increasing publication trend. High-frequency Chinese keywords included prediction, influencing factor, and machine learning, while English keywords frequently used were machine learning, prediction, and artificial intelligence. The Chinese keywords formed 14 clusters such as epidemiological characteristic, dietary pattern, and elderly individual, and the English keywords formed 21 clusters including prediction model, risk factor, and adult. In international studies, health policy, COVID-19, and digital health were the emerging frontier keywords. Eleven core papers were selected, covering key areas like traffic accident risk assessment, public health big data application, and deep learning in medical diagnosis.Conclusions:This study systematically summarized the research hotspots and development trends of AI applications in epidemiology over the past decade by using bibliometric methods, which indicated that current AI-based epidemiological studies are still in the exploratory phase, with the coexisting of both advantages and challenges. Continued attention should be paid to the future development of this field.

Objective:To describe the hotspots and application trends of artificial intelligence (AI) in epidemiology in the past decade and analyze its advantages and challenges.Methods:The literatures with AI and epidemiology related keywords were systematically retrieved from Web of Science and China National Knowledge Infrastructure from 2014 to 2024. CiteSpace was used for bibliometric analysis of publication volume, keyword co-occurrence, clustering, emergence and cited literature co-occurrence analysis.Results:A total of 5 389 English papers and 1 659 Chinese papers were included, showing an increasing publication trend. High-frequency Chinese keywords included prediction, influencing factor, and machine learning, while English keywords frequently used were machine learning, prediction, and artificial intelligence. The Chinese keywords formed 14 clusters such as epidemiological characteristic, dietary pattern, and elderly individual, and the English keywords formed 21 clusters including prediction model, risk factor, and adult. In international studies, health policy, COVID-19, and digital health were the emerging frontier keywords. Eleven core papers were selected, covering key areas like traffic accident risk assessment, public health big data application, and deep learning in medical diagnosis.Conclusions:This study systematically summarized the research hotspots and development trends of AI applications in epidemiology over the past decade by using bibliometric methods, which indicated that current AI-based epidemiological studies are still in the exploratory phase, with the coexisting of both advantages and challenges. Continued attention should be paid to the future development of this field.

Objective To explore the influence factors of timed up and go test(TUG)in patients with maintenance hemodialysis(MHD).Methods Cross-sectional investigation method was used to select patients who underwent regular hemodialysis in Beijing Luhe Hospital,Capital Medical University from June 1,2023 to January 31,2024,with complete data and cooperation.The patients were respectively treated with a simple mental state checklist and 5 times sit to stand test(STS-5)and TUG scale were used to assess cognitive function,lower limb muscle strength,balance and gait function.Grouping was carried out according to the results of TUG.If TUG 12s,it is judged as the TUG normal group;if TUG＞12s,it is judged as the TUG prolonged group.Multi-factor Logistic regression was used to analyze the influencing factors in the TUG prolonged group,and the predictive value of influencing factors to TUG prolonged patients was analyzed by receiver operating characteristic(ROC)curve.Results A total of 146 patients with MHD were included in this study,including 86 patients in TUG prolonged group and 60 patients in TUG normal group.Gender,age,combined hypertension,combined diabetes,blood glucose,education level,Minimum Mental State Examination(MMSE)score,C-reactive protein were compared between groups.Serum albumin,serum creatinine,serum sodium,whole parathyroid hormone,low density lipoprotein-cholesterol,serum uric acid,STS-5 results,and single urea clearance index(Kt/V)had statistical differences(P＜0.05).Multivariate Logistic regression showed that old age,female,low MMSE score,low Kt/V,and STS-5 duration were independent risk factors for TUG prolonged(P＜0.05).ROC curve showed that the area under the curve(AUC)of age,gender,total MMSE score and STS-5 time in predicting TUG prolongation in MHD patients were 0.825,0.678,0.777 and 0.836,respectively(P＜0.01),which had high predictive value.The AUC of Kt/V in predicting TUG prolongation was 0.602(P=0.037),has a certain predictive value;AUC of TUG prolongation in MHD patients predicted by the above 5 indicators as a global model was 0.923(P＜0.01).Conclusions The age,gender,MMSE score,STS-5 prolongation and Kt/V level of MHD patients are independent risk factors for TUG prolongation,and may be biological indicators for predicting TUG prolongation.

Objective To explore the influence factors of timed up and go test(TUG)in patients with maintenance hemodialysis(MHD).Methods Cross-sectional investigation method was used to select patients who underwent regular hemodialysis in Beijing Luhe Hospital,Capital Medical University from June 1,2023 to January 31,2024,with complete data and cooperation.The patients were respectively treated with a simple mental state checklist and 5 times sit to stand test(STS-5)and TUG scale were used to assess cognitive function,lower limb muscle strength,balance and gait function.Grouping was carried out according to the results of TUG.If TUG 12s,it is judged as the TUG normal group;if TUG＞12s,it is judged as the TUG prolonged group.Multi-factor Logistic regression was used to analyze the influencing factors in the TUG prolonged group,and the predictive value of influencing factors to TUG prolonged patients was analyzed by receiver operating characteristic(ROC)curve.Results A total of 146 patients with MHD were included in this study,including 86 patients in TUG prolonged group and 60 patients in TUG normal group.Gender,age,combined hypertension,combined diabetes,blood glucose,education level,Minimum Mental State Examination(MMSE)score,C-reactive protein were compared between groups.Serum albumin,serum creatinine,serum sodium,whole parathyroid hormone,low density lipoprotein-cholesterol,serum uric acid,STS-5 results,and single urea clearance index(Kt/V)had statistical differences(P＜0.05).Multivariate Logistic regression showed that old age,female,low MMSE score,low Kt/V,and STS-5 duration were independent risk factors for TUG prolonged(P＜0.05).ROC curve showed that the area under the curve(AUC)of age,gender,total MMSE score and STS-5 time in predicting TUG prolongation in MHD patients were 0.825,0.678,0.777 and 0.836,respectively(P＜0.01),which had high predictive value.The AUC of Kt/V in predicting TUG prolongation was 0.602(P=0.037),has a certain predictive value;AUC of TUG prolongation in MHD patients predicted by the above 5 indicators as a global model was 0.923(P＜0.01).Conclusions The age,gender,MMSE score,STS-5 prolongation and Kt/V level of MHD patients are independent risk factors for TUG prolongation,and may be biological indicators for predicting TUG prolongation.

ObjectiveTo sort out the historical evolution， prescription evolution and modern clinical application of Huagaisan. MethodHuagaisan and its synonym Huagaitang are used as keywords to search the databases of Traditional Chinese Medicine Think Tank， Chinese Medical Dictionary， Airusheng Chinese Medical Database and China National Knowledge Infrastructure（CNKI）. According to the inclusion and exclusion criteria， we obtained the information of ancient books and modern clinical research literature related to Huagaisan， and systematically reviewed and analyzed the historical origin， prescription composition， preparation method， dosage， efficacy， medicinal material origin， processing method and modern clinical application of Huagaisan. ResultA total of 198 pieces of ancient book information were included， involving 93 ancient Chinese medicine books. Huagaisan was composed of fried Perillae Fructus， red Poria， fried Mori Cortex， Citri Eoxcarpium Rubrum， stir-fried Armeniacae Semen Amarum， Ephedrae Herba and fried Glycyrrhizae Radix et Rhizoma， which had the efficacy of promoting the lungs and relieving epidemiological symptoms， expelling phlegm and relieving cough， and treating cough with wind-cold bundled epidemiological symptoms and stagnation of phlegm and Qi. The preparation method was suggested as boiling powder， crushing the seven herbs into coarse particles， the dosage of each drug was fried Perillae Fructus of 1.27 g， red Poria of 1.27 g， fried Mori Cortex of 1.27 g， Citri Eoxcarpium Rubrum of 1.27 g， stir-fried Armeniacae Semen Amarum of 1.27 g， Ephedrae Herba of 1.27 g and fried Glycyrrhizae Radix et Rhizoma of 0.64 g， taking 8.26 g when decocting， adding 300 mL of water， decocting to 210 mL， removing the dregs， and taking it warmly after meals. Twenty-one clinical research papers were included to analyze the modern clinical application of Huagaisan， which was mainly used in the treatment of respiratory diseases such as pneumonia， asthma， bronchitis and so on. ConclusionThis paper has verified and summarized the key information of the famous classical formula Huagaisan， which can provide a detailed reference basis for the development and clinical application of its compound preparation.

Long non-coding RNAs(lncRNAs)play an indispensable role in the occurrence and development of ovarian cancer(OC).However,the potential involvement of lncRNAs in the progression of OC is largely unknown.To investigate the detailed roles and mechanisms of RAD51 homolog B-antisense 1(RAD51B-AS1),a novel lncRNA in OC,reverse transcription-quantitative polymerase chain reaction(RT-qPCR)was performed to verify the expression of RAD51B-AS1.Cellular proliferation,metastasis,and apoptosis were detected using the cell counting kit-8(CCK-8),colony-formation,transwell,and flow cytometry assays.Mouse xenograft models were established for the detection of tumorigenesis.The results revealed that RAD51B-AS1 was significantly upregulated in a highly metastatic human OC cell line and OC tissues.RAD51B-AS1 significantly increased the proliferation and metastasis of OC cells and enhanced their resistance to anoikis.Biogenetics prediction analysis revealed that the only target gene of RAD51B-AS1 was RAD51B.Subsequent gene function experiments revealed that RAD51B exerts the same biological effects as RAD51B-AS1.Rescue experiments demonstrated that the malignant biological behaviors promoted by RAD51B-AS1 overexpression were partially or completely reversed by RAD51B silencing in vitro and in vivo.Thus,RAD51B-AS1 promotes the malignant biological behaviors of OC and activates the protein kinase B(Akt)/B cell lymphoma protein-2(Bcl-2)signaling pathway,and these effects may be associated with the positive regulation of RAD51B expression.RAD51B-AS1 is expected to serve as a novel molecular biomarker for the diagnosis and prediction of poor prognosis in OC,and as a potential therapeutic target for disease management.

Objective To conduct bibliometric visualization analyses of the literature domestic and overseas on active surveillance of post-marketing safety of drugs,aiming to display the current status and trend of hotspots in this field and to provide references for future research and the improvement of the Chinese management system of active surveillance.Methods The English and Chinese literature on active surveillance of post-marketing safety of drugs were searched in Web of Science and CNKI respectively and imported into CiteSpace 6.3.R2 software for the analysis of the number of publications,authors,institutions,and national cooperative networks,and the analysis of keyword co-occurrence,clustering and emergence.Results 415 Chinese and 676 English literature were included,with an overall increasing trend in annual publication volume.The author collaboration network of Chinese literature was smaller than that of English literature,and the partnership network was sparse,with no strong centralized institution.Domestic drug regulatory agencies played an important role in the field,while drug companies'monitoring research on their own products was still relatively scarce.The research topic covered active surveillance systems,technical method research,and drug safety active surveillance practice research for specific drugs and diseases.Conclusion Countries worldwide have widely considered active surveillance of post-marketing drug safety.The heat of research activities in China has shown a significant growth trend.However,there is still a significant gap compared with the international frontiers.Further cooperation needs to be strengthened to promote the improvement of the active surveillance management system in China.

Single-cell sequencing is a new technology for high-throughput sequencing analysis of the genome, transcriptome, and epigenome at the single-cell level. Currently, single-cell sequencing technology has been widely used in bladder cancer research, opening up new ways to understand the biology of subtle bladder tumors by identifying different cell subpopulations, immune microenvironment, and single-cell technology is expected to make important changes in bladder diagnosis and treatment by identifying new biomarkers and using targeted therapies. This paper discussed the application of single-cell sequencing in bladder cancer research and clinical practice by summarizing recent advances in heterogeneous cell subpopulations, developmental patterns, and immune microenvironment and drug resistance in bladder cancer.

Objective To investigate the changes of functional brain connectivity over multiple frequency bands in resting-state electroencephalography (EEG) for lower limb amputation patients. Methods Resting-state EEG was collected from 18 lower limb amputees and 22 healthy controls from November, 2020 to June, 2021. Functional connectivity matrix was constructed with phase-locked values (PLV), and compared between groups. Results The functional connectivity was weaker in the amputees than in the controls on α band (t = 3.433, P = 0.001) and β band (t = 3.806, P = 0.001), and there was no significant difference on δ band (t = 1.429, P = 0.161) and θ band (t = 1.211, P = 0.233). Conclusion EEG functional connectivity weakens in lower limb amputees in on the α and β band, which results in neuroplasticity of multiple brain regions, not only for limb-respond cortex, but also for frontal, temporal and occipital cortices.

Objective:To investigate the predictive value of diffusion tensor imaging (DTI) parameters in upper extremity motor function recovery after surgery in patients with acute cervical spinal cord injury (CSCI).Methods:Twenty-three patients with acute CSCI who received postoperative systemic rehabilitation therapy in Department of Neurosurgery, 900 th Hospital of Joint Logistics Team of People's Liberation Army from May 2019 to July 2021 were selected as an experimental group, and 22 healthy subjects (healthy control group) matched with age and gender were selected from Physical Examination Center of the same hospital at the same period. Routine MRI sequence and DTI scan of the cervical spinal cord, scale of American Association for Spinal Cord Injury (ASIA) and modified Barthe index (mBI) were performed in patients of the experimental group 1 d and 3 months after surgery. Routine MRI sequence and DTI scan of the cervical spinal cord were performed in healthy subjects after enrollment. The DTI parameters in different regions between the two groups were compared, and the differences in DTI parameters, ASIA scores and mBI in patients of the experimental group before and after surgery were compared. Correlations of preoperative DTI parameters with preoperative upper extremity motor ASIA scores and upper extremity motor recovery rate 3 months after surgery were analyzed by Pearson correlation analysis. Receiver operating characteristic (ROC) curve was used to analyze the predictive efficacy of preoperative fractional anisotropy (FA) in upper extremity motor function recovery in CSCI patients 3 months after surgery. Results:As compared with the healthy control group, the experimental group had significantly lower preoperative FA in the injury area and distal injury area, and statistically higher preoperative apparent diffusion coefficient (ADC, P<0.05). In patients of the experimental group, preoperative FA in the injury area was significantly lower and ADC in the injury area was significantly higher as compared with those in the distal injury area ( P<0.05); patients of the experimental group had significantly higher FA in these two regions, upper extremity motor ASIA scores and mBI, and significantly lower ADC 3 months after surgery as compared with those 1 d before surgery ( P<0.05). The preoperative FA in the injury area and distal injury area in CSCI patients were positively correlated with preoperative upper extremity motor ASIA scores and upper extremity motor recovery rate 3 months after surgery ( P<0.05). ROC curve results showed that the area under the curve (AUC) of preoperative FA in injury area in predicting upper extremity motor function recovery 3 months after surgery was 0.912 ( 95%CI: 0.783-1.000, P<0.001); that of preoperative FA in the distal injury area was 0.842 ( 95%CI: 0.682-1.000, P<0.001). Conclusion:DTI parameters FA and ADC are sensitive indicators for detecting CSCI; preoperative FA in the injury area and distal injury area can be used to predict the upper extremity motor function recovery, but the efficacy of the former is superior to that of the later.