Objective:To investigate the current awareness of large language models (LLM) among Chinese clinical physicians and analyze the application needs of cardiovascular specialists.Methods:This is a cross-sectional study utilized convenience sampling. In December 2023, a self-designed questionnaire was distributed to 7 980 clinical physicians, including 930 cardiologists. The survey collected demographic information, including work city (categorized as first-tier, new first-tier, second-tier, third-tier, and fourth-tier and below), hospital level, professional title, and department. And the awareness of LLM, and their application demands in clinical decision-making support, information filtering, and scientific research work were also collected. Differences in awareness and application requirements across geographic regions, hospital tiers, professional ranks, and medical departments were analyzed. Besides, specific demands of cardiovascular specialists were further examined.Results:Among the 7 980 clinical physicians, the awareness rate of LLM was 76.3% (6 088/7 980), and the utilization rate was 11.8% (942/7 980). For the 930 cardiologists, the awareness rate was 78.5% (730/930) and the utilization rate was 11.4% (106/930). Significant differences in awareness and utilization rates were observed across city tiers, hospital grades, and departments (all P<0.05). No significant difference was found among professional titles ( P=0.053). Among the 6 088 physicians aware of LLM, demand rates for clinical information filtering, clinical decision support, and research assistance were 87.3% (5 312/6 088), 78.4% (4 774/6 088), and 75.8% (4 616/6 088), respectively. For the 730 cardiologists aware of LLM, these rates were 91.0% (664/730), 79.2% (578/730), and 75.9% (554/730), respectively. Significant differences in demands for clinical information filtering and research assistance were observed across city tiers, hospital grades, professional titles, and departments (all P<0.05), while no significant difference was noted for decision support demands across hospital grades ( P=0.085). In clinical information screening and acquisition, cardiologists from different city tiers exhibited statistically significant differences in the demand for literature interpretation. Similarly, variations in the demand for conference summaries, expert biographies, healthcare policies, and social news were noted among cardiologists with different professional titles, while disparities in patient education and science popularization needs were identified across city tiers and hospital grades (all P<0.05). In clinical decision-making support, cardiologists from diverse city tiers and professional titles demonstrated distinct differences in guideline and consensus inquiries, and those from various city tiers showed varied demands for pharmaceutical and medical device-related content (all P<0.05). For research support, cardiologists across city tiers and professional titles exhibited statistically significant differences in trial protocol design requirements, while those from varying city tiers differed in literature search/analysis and research application procedures. Additionally, physicians from different hospital grades displayed divergent needs for data collection (all P<0.05). Conclusions:The adoption of LLM is significantly influenced by regional disparities, institutional resources, and professional backgrounds. Implementing targeted interventions, such as enhancing technical training, optimizing LLM functionalities, and improving accessibility across diverse healthcare settings, could encourage widespread integration of LLM into clinical practice. Such measures could ultimately enhance the quality and efficiency of medical services in China and foster innovations in healthcare delivery.

Objective:To investigate the association between photodynamic diagnosis(PDD)and treatment efficacy of 5-aminolevu-linic acid(ALA)photodynamic therapy(PDT)in patients with cutaneous squamous cell carcinoma in situ(cSCCis).Methods:A pro-spective cohort study was conducted among the patients with cSCCis who underwent ALA-PDT in Department of Dermatology,Huashan Hospital,Fudan University,from January 2020 to November 2024.All patients were diagnosed with cSCCis based on biopsy,and invasive squamous cell carcinoma was excluded.Clinical information was collected and PDD was performed before treatment,and the patients were divided into moderate fluorescence group and strong fluorescence groups based on fluorescence intensity.All patients received six sessions of PDT treatment at an interval of 1-2 weeks.The primary endpoint was the initial complete clearance rate at 3 months after the last treatment session,and secondary endpoints were the sustained complete clearance rate at 12 months after the last treatment session and treatment failure rate.The multivariate regression analysis,the survival curves,and the Cox regression analysis were used to investigate the association between PDD fluorescence intensity and treatment efficacy,as well as other influencing factors for treatment efficacy.Results:Compared with the strong fluorescence group,the moderate fluorescence group had significantly lower initial complete clearance rate[57.14%(8/14)vs.93.33%(28/30),odds ratio OR=0.100,P=0.010]and sustained complete clearance rate[42.86%(6/14)vs.76.67%(23/30),OR=0.230,P=0.030)].The multivariate regression analysis showed that moderate fluores-cence was an independent risk factor for initial complete clearance(OR=0.030,P=0.030).The moderate fluorescence group had a significantly higher treatment failure rate than the strong fluorescence group[57.14%(8/14)vs.23.33%(7/30),P=0.030].The survival analysis and the Cox regression analysis showed that moderate fluo-rescence was an independent risk factor for PDT treatment failure(hazard ratio=3.040,P=0.048).There were no significant differences in adverse reactions between the two groups.Conclusion:PDD fluorescence intensity can predict the efficacy of ALA-PDT in patients with cSCCis.Moderate fluorescence indicates a higher risk of treatment failure,which can help to guide clinicians in develop-ing individualized treatment strategies.

Objective:To analyze the current status and factors influencing clinician participation in remote patient management in China.Methods:In December 2023, a structured electronic questionnaire was administered to 7 980 clinicians, including 930 cardiologists. The survey assessed clinicians′ participation in online doctor-patient interactions; differences in hospital support across city tiers, hospital grades, professional titles, and departments; and factors influencing the willingness of clinicians to invest time in remote patient management.Results:Among the 7 980 surveyed clinicians, online consultations had the highest participation rate (72.2%). Among cardiologists, participation rates for online consultations, health education, and post-consultation management were 73.3%, 66.9%, and 38.5%, respectively, which were relatively higher than those of other specialties. Hospital-based support for physicians in remote patient management showed significant variations across specialties and regions. Among cardiologists, 68.4% received "encouraging" policies, with the majority falling under "encouragement without incentive policies" (42.6%). In tier 3 cities, the proportion of physicians receiving "encouragement without incentive policies" was the highest (47.9%), while the proportion in the "cautious, requiring reporting" category was the lowest (3.9%). During remote patient management, the proportions of clinicians receiving support from professional teams were highest among those in tier 3 cities (29.6%) and cardiologists (30.5%). A significant interaction effect was observed between hospital policy and specialty (cardiologists vs. all clinicians) regarding physicians′ willingness to invest time in remote patient management ( F=5.95, P<0.001). Among cardiologists, those working in institutions with "encouraging, with incentives" policies reported a significantly longer median weekly investment time (10.0 h) compared to those under "neutral, unrestricted" policies (7.0 h, P<0.001). Cardiologists with team support reported a significant increase in the time they were willing to invest (10 h/week) than those without team support (7.0 h/week, P<0.001), although no significant interaction effect was found when compared with all clinicians ( P=0.186). Cardiologists with a high online income (>5 000 Yuan/month) reported a significantly longer weekly investment time in remote management (25.0 h) compared to those with lower income (<200 yuan/month; 8.0 h, P<0.001). However, whether the income met their personal expectations had no significant effect on their time commitment ( P=0.638). Conclusions:Clinicians from tertiary hospitals and tier 3 cities demonstrated a higher level of engagement in remote patient management. Strengthening hospital policy support, enhancing team-based collaborations, and increasing online income levels may help promote the broader adoption of telemedicine.

Objective:This study aims to construct a machine learning model to predict depression symptoms in the elderly and analyze the key influencing factors of depression in the elderly using the shapley additive interpretation (SHAP) method.Methods:Based on entries from the 2018 China Health and Retirement Longitudinal Study database, a sample of 5 954 elderly individuals was selected. Feature selection using Support Vector Machine Recursive Feature Elimination, Extreme Gradient Boosting (XGBoost) - Recursive Feature Elimination (RFE), and the Lasso algorithm, which was combined with five classifiers-logistic regression, decision trees, random forests, support vector machines, and XGBoost-to explore the classification effectiveness for depressive symptoms in the elderly. Finally, the SHAP method was used to interpret the analysis of the model with the highest receiver operating characteristic curve areas under the curve (AUC).Results:The accuracy of 15 prediction models ranged from 0.702 to 0.743, with AUC between 0.730 and 0.795. Sensitivity was reported at 0.546 to 0.588, while specificity ranges from 0.783 to 0.865. The model XGBoost-RFE-XGBoost presented the highest AUC. Based on SHAP values, the top four factors influencing depressive symptoms in older adults were life satisfaction, duration of nighttime sleep, disability status, and self-rated health.Conclusion:This study developed a highly efficient and interpretable risk prediction model for depressive symptoms in older adults, which could help identify high-risk older adults and give personalized interventions.

Objective To explore the mediating role of activities of daily living (ADL) in pain and depressive symptoms in the elderly in China. Methods Utilizing the data from 2020 China Health and Retirement Longitudinal Study, 4403 Chinese elderly individuals aged ≥ 60 years old were selected as the research subjects. Depression Scale (CES-D 10) of the Center for Epidemiological Survey and ADL scale were used in the study. The PROCESS4.1 macro was used to test the mediating effect of daily living activities between pain and depressive symptoms, and the Bootstrap method was applied for verification of the mediating variables. Results A total of 2368 cases of depressive symptoms were detected in the elderly in China, with a detection rate of 53.78%. Pain was positively correlated with depressive symptoms (r=0.27, P<0.01), and activities of daily living were negatively correlated with pain and depressive symptoms (r=-0.27, -0.337, P<0.01). The results showed that the total effect value of pain on depressive symptoms was 0.33, the direct effect value was 0.24, and the mediating effect value of daily living activities was 0.09, accounting for 27.27%. Conclusion Pain and activities of daily living are important factors influencing depressive symptoms in the elderly, and activities of daily living play a partial mediating role in the relationship between pain and depressive symptoms in the elderly.

Objective: To investigate the spatio-temporal clustering characteristics of influenza in Ningxia Hui Autonomous Region from 2014 to 2023, so as to provide the basis for strengthening influenza prevention and control. Methods: Data pertaining to influenza cases reported in Ningxia Hui Autonomous Region from 2014 to 2023 were retrieved from the Infectious Disease Surveillance System of the Chinese Disease Prevention and Control Information System, including age, sex, current residence, onset date, and reporting date. The seasonal incidence of influenza was analyzed using seasonal index. The spatio-temporal clustering characteristics of influenza were identified using spatial autocorrelation analysis and spatio-temporal scan analysis. Results: A total of 20 377 influenza cases were reported in Ningxia Hui Autonomous Region from 2014 to 2023, with a male-to-female ratio of 1.15∶1. The majority were children under 15 years, with 10 950 cases accounting for 53.74%. Influenza was highly prevalent in January, February, March, and December, with seasonal indices of 219.06%, 111.00%, 246.65%, and 366.24%, respectively. The average annual reported incidence was 29.55/100 000, among which Pengyang County, Jinfeng District, Dawukou District, Xiji County, and Litong District had higher average annual reported incidence, at 63.99/100 000, 55.71/100 000, 55.70/100 000, 49.49/100 000, and 49.04/100 000, respectively. Spatial autocorrelation analysis showed that in 2023, there was spatial clustering of influenza cases in Ningxia Hui Autonomous Region (Moran's I=0.333, P<0.05), with a high-high cluster in Jingyuan County, while in other years, the distribution of influenza cases was random (all P>0.05). Spatio-temporal scan analysis showed that from 2014 to 2023, there were four space-time clusters in Ningxia Hui Autonomous Region, including one type Ⅰ cluster in Hongsibao District of Wuzhong City, with the clustering period from January 20 to 26, 2014; and three type Ⅱ clusters, mainly in January, February, March and December, covering one area in Shizuishan City, five areas in Guyuan City, one area in Zhongwei City, three areas in Wuzhong City, and four areas in Yinchuan City. Conclusions From 2014 to 2023, children under 15 years were the primary population affected by influenza in Ningxia Hui Autonomous Region, with distinct spatio-temporal distribution characteristics. The peak incidence occurred during the winter and spring seasons, and the main clustering areas were in the southern regions.

Objective:To explore the medication law of national TCM master Lu Fang in the treatment of primary trigeminal neuralgia (PTN) based on data mining.Methods:With the prescription of the outpatient patients of Harbin Traditional Chinese Medicine Hospital of Professor Lu Fang from September 2014 to September 2022 as the data source, the frequency, property and taste, and meridian tropism of the prescribed drugs were analyzed using Excel 2022 software. R 4.2.1 was used for mining analysis on Chinese materia medica, including correlation, relevance, and clustering,and the medication law in the treatment of PTN was discussed.Results:A total of 300 prescriptions were analyzed, involving 177 kinds of Chinese materia medica, with a frequency of 3 120 times, and 34 kinds of of high-frequency Chinese materia medica. The high frequently Chinese materia medica included Chuanxiong Rhizoma, Angelicae Dahuricae Radix, Puerariae Lobatae Radix, Ligustici Rhizoma et Radix, and Viticis Fructus. The main properties were warm, slightly cold, and neutral, while the main tastes were pungent, bitter, and sweet. The meridian tropism analysis ranked the liver, lung, spleen, and stomach meridians in descending order. Analysis yielded 21 strong association rules, and the association analysis formed a core prescription group based on Chuanxiong Rhizoma, Angelicae Dahuricae Radix, and Ligustici Rhizoma et Radix. The analysis obtained 5 types of clustering combinations.Conclusion:Professor Lu Fang's the medication law to treat primary trigeminal neuralgia is mainly dispelling wind and alleviating pain, which is often combined with the methods, such as searching and dredging collaterals, clearing and dispelling the stagnated heat, calming the liver and subduing yang, soothing the liver and invigorating the spleen.

Objective To investigate the structural changes and influencing factors of the average inpatient cost per admis-sion in public hospitals in Guangdong Province.Methods Grey relational analysis and structural variation degree analysis were used to analyze the correlation and changes between the average inpatient cost per admission and various cost components in public hospitals of Guangdong Province from 2017 to 2023.Results The average inpatient cost per admission in public hospitals of Guangdong Province showed an overall upward trend from 2017 to 2023,with an average annual growth rate of 3.84％.Among the components,laboratory test fees and examination fees grew at average annual rates of 6.17％and 6.68％,respectively.The top four cost components with the highest grey relational degree with the average inpatient cost were laboratory test fees(0.867),exam-ination fees(0.835),nursing fees(0.784),and treatment fees(0.728).The top four components with the largest structural vari-ation values were surgery fees(2.57％),medical material fees(1.77％),laboratory test fees(1.56％),and examination fees(1.45％).Conclusion The growth of the average inpatient cost per admission has slowed,and the cost structure has been opti-mized to some extent.However,the relatively rapid increase in laboratory test and examination fees has a significant impact on the cost structure.It is necessary to deepen the coordinated governance of healthcare,medical insurance,and medicine,strengthen the leveraging role of medical insurance payment,improve the external governance system and scientific compensation mechanism,and combine these with refined hospital management to promote reasonable cost control and high-quality development in public hospitals.

Objective:This study aims to construct a machine learning model to predict depression symptoms in the elderly and analyze the key influencing factors of depression in the elderly using the shapley additive interpretation (SHAP) method.Methods:Based on entries from the 2018 China Health and Retirement Longitudinal Study database, a sample of 5 954 elderly individuals was selected. Feature selection using Support Vector Machine Recursive Feature Elimination, Extreme Gradient Boosting (XGBoost) - Recursive Feature Elimination (RFE), and the Lasso algorithm, which was combined with five classifiers-logistic regression, decision trees, random forests, support vector machines, and XGBoost-to explore the classification effectiveness for depressive symptoms in the elderly. Finally, the SHAP method was used to interpret the analysis of the model with the highest receiver operating characteristic curve areas under the curve (AUC).Results:The accuracy of 15 prediction models ranged from 0.702 to 0.743, with AUC between 0.730 and 0.795. Sensitivity was reported at 0.546 to 0.588, while specificity ranges from 0.783 to 0.865. The model XGBoost-RFE-XGBoost presented the highest AUC. Based on SHAP values, the top four factors influencing depressive symptoms in older adults were life satisfaction, duration of nighttime sleep, disability status, and self-rated health.Conclusion:This study developed a highly efficient and interpretable risk prediction model for depressive symptoms in older adults, which could help identify high-risk older adults and give personalized interventions.

Objective To investigate the structural changes and influencing factors of the average inpatient cost per admis-sion in public hospitals in Guangdong Province.Methods Grey relational analysis and structural variation degree analysis were used to analyze the correlation and changes between the average inpatient cost per admission and various cost components in public hospitals of Guangdong Province from 2017 to 2023.Results The average inpatient cost per admission in public hospitals of Guangdong Province showed an overall upward trend from 2017 to 2023,with an average annual growth rate of 3.84％.Among the components,laboratory test fees and examination fees grew at average annual rates of 6.17％and 6.68％,respectively.The top four cost components with the highest grey relational degree with the average inpatient cost were laboratory test fees(0.867),exam-ination fees(0.835),nursing fees(0.784),and treatment fees(0.728).The top four components with the largest structural vari-ation values were surgery fees(2.57％),medical material fees(1.77％),laboratory test fees(1.56％),and examination fees(1.45％).Conclusion The growth of the average inpatient cost per admission has slowed,and the cost structure has been opti-mized to some extent.However,the relatively rapid increase in laboratory test and examination fees has a significant impact on the cost structure.It is necessary to deepen the coordinated governance of healthcare,medical insurance,and medicine,strengthen the leveraging role of medical insurance payment,improve the external governance system and scientific compensation mechanism,and combine these with refined hospital management to promote reasonable cost control and high-quality development in public hospitals.