Objective: To address the lack of fine-grained clinical recognition for specific Yang deficiency syndrome subtypes and the limitations of conventional object detection models in extracting irregular, low-contrast tongue phenotypes. This study aims to develop an objective subtype recognition framework based on an improved You Only Look Once nano (YOLO11n) architecture, using a standardized visual phenotype matrix to translate macroscopic traditional Chinese medicine (TCM) descriptions into quantifiable clinical targets. Methods: This cross-sectional diagnostic study consecutively enrolled adult inpatients admitted to the Department of Thyroid and Breast Surgery, The First Affiliated Hospital of Wannan Medical University (Yijishan Hospital), between September 1, 2024 and June 1, 2025, who were suspected of having Yang deficiency constitution based on initial TCM consultation. Clinical tongue image data were collected for analysis. Based on an Expert Visual Phenotype Annotation Matrix, a five-category recognition system was established, including the following TCM syndrome subtypes: spleen-dampness exuberance syndrome, mild kidney Yang deficiency syndrome, upper heat and lower cold syndrome, simultaneous Yin-Yang deficiency syndrome, and Yin deficiency and fluid depletion syndrome (negative control). The proposed Yang deficiency YOLO (YD-YOLO) model, built upon the YOLO11n baseline, integrates the Cross Stage Partial with kernel size 2 (C3k2)-GhostBottleneck-Dynamic Convolution (GBDC) module into the backbone to adaptively extract low-contrast features, and embeds the multipath aggregation coordinate attention (MACA) mechanism into the neck to suppress background interference through multi-scale spatial coordination. Gradient-weighted class activation mapping (Grad-CAM) was used to visualize feature attribution and evaluate the biological plausibility of the model’s focus. Model performance was evaluated through ablation and comparative experiments using mean average precision (mAP), precision, recall, F1 score, inference speed (frames per second, FPS), overall accuracy, Cohen’s kappa, and the area under the receiver operating characteristic (ROC) curve (AUC). Results: Based on the final inclusion of 1 186 clinical cases, the YD-YOLO model had an overall accuracy of 91.5%, a Cohen’s kappa of 0.912, and an mAP@50 of 0.731 [higher than the YOLO11n baseline (0.681)], with AUC ranging from 0.91 to 0.97 across all TCM syndrome subtypes. Among the TCM syndrome subtypes, the mild kidney Yang deficiency syndrome had the highest mAP@50 (0.900), and the inference speed reached 89.00 FPS. Grad-CAM analysis showed that the model localized activation to key TCM pathological features, such as marginal tooth marks and focal root coatings, while suppressing non-diagnostic oral background noise. Conclusion The YD-YOLO model demonstrates the feasibility of deep learning for the fine-grained classification of TCM Yang deficiency subtypes. By integrating visual phenotype quantification with model interpretability, the proposed framework provides an objective basis for syndrome differentiation, supporting the development of standardized digital diagnostic systems and the provision of clinical decision support in TCM practice.

In order to establish a method for the detection of serum antibodies to donkey-derived e-quine coronavirus(ECoV),recombinant ECoV N protein was expressed in E.coli system,purified by nickel column affinity chromatography and identified by Western blot.After optimizing the re-action conditions,the indirect ELISA(iELISA)detection method was established using the puri-fied recombinant protein as coating antigen and used to detect 143 clinical serum samples.The re-sults showed that the recombinant N protein,which has good reaction activity with serum antibod-y,was successfully expressed.The optimum conditions of the established iELISA method were as follows:the amount of antigen coated was 0.2 μg/well and overnight at 4 ℃,10％skimmed milk powder solution was sealed at 37℃ for 1.5 h,the dilution concentration of serum was 1∶200,and the enzyme-labeled secondary antibody diluted at 1∶10 000.The sensitivity test results showed that the positive serum could be diluted to 1∶6 400.The specificity test results showed that all an-tibodies to several donkey pathogens were negative.The repetitive test results showed that the in-tra-and inter-batch coefficients of variation were 2.90％-6.12％and 2.29％-7.88％respectively.The positive rate of clinical donkey serum was 57.3％.The iELISA established in this study pro-vides a technical support for epidemiological investigation and antibody surveillance.

In order to establish a method for the detection of serum antibodies to donkey-derived e-quine coronavirus(ECoV),recombinant ECoV N protein was expressed in E.coli system,purified by nickel column affinity chromatography and identified by Western blot.After optimizing the re-action conditions,the indirect ELISA(iELISA)detection method was established using the puri-fied recombinant protein as coating antigen and used to detect 143 clinical serum samples.The re-sults showed that the recombinant N protein,which has good reaction activity with serum antibod-y,was successfully expressed.The optimum conditions of the established iELISA method were as follows:the amount of antigen coated was 0.2 μg/well and overnight at 4 ℃,10％skimmed milk powder solution was sealed at 37℃ for 1.5 h,the dilution concentration of serum was 1∶200,and the enzyme-labeled secondary antibody diluted at 1∶10 000.The sensitivity test results showed that the positive serum could be diluted to 1∶6 400.The specificity test results showed that all an-tibodies to several donkey pathogens were negative.The repetitive test results showed that the in-tra-and inter-batch coefficients of variation were 2.90％-6.12％and 2.29％-7.88％respectively.The positive rate of clinical donkey serum was 57.3％.The iELISA established in this study pro-vides a technical support for epidemiological investigation and antibody surveillance.

Objective:Exploring the association rules of factors influencing high hospitalization costs for cancer patients, providing references for hospitals to optimize medical cost management measures.Methods:In the inpatient case information system of a tertiary general hospital, the medical record homepages of inpatients in the DRG groups of the oncology department in 2022 were obtained. The upper four scores of hospitalization costs was used as the threshold for patient grouping. Patients with hospitalization costs≥this threshold were the high-cost group, while other patients were control group; 12 factors, including age, gender, and admission condition, etc, were considered as potential influencing factors of high hospitalization costs. FP-Growth and Apriori algorithms were used to excavate the potential association rules between the influencing factors of high hospitalization costs. Logistic regression was used to analyze the independent influencing factors of high hospitalization costs.Results:A total of 5 512 hospitalized patients were included, including 1 378 patients in the high-cost group. Thirteen validated strong association rules for factors influencing high hospitalization costs were obtained, of which the rule antecedents included age (≥70 years), number of days in hospital (≥7 days), other diagnoses (≥5), surgery, planned readmission, use of antibiotics, admission (general/critical), living admission score (61~99), level of care (level 1/level 2), non-day ward, criticality during hospitalisation. Logistic regression results showed that all nine influencing factors except gender, use of antibiotics, and readmission plans were independent influences on high hospitalization costs ( P<0.05). Conclusions:The joint application of FP-Growth and Apriori algorithm could effectively explore the association rules of high hospitalization costs for oncology patients. The early warning information mainly included the number of hospitalization days, the number of other diagnoses, surgeries, and so on. It was suggested that medical institutions can reasonably control the high hospitalization costs through clinical pathway management, diagnosis and treatment process reengineering, admission risk assessment, and multidisciplinary collaborative diagnosis and treatment strategies.

Objective:To analyze the influencing factors of China′s DRG payment system reform(DRG reform) and its hierarchical relationship, for references for the in-depth promotion of China′s medical insurance payment reform.Methods:Relevant literature on DRG reform in China from databases such as CNKI, Wanfang Database, Pubmed, etc, were obtained. Content analysis method was used to extract the influencing factors of DRG reform. The correlation between each influencing factor was determined through expert discussion. An interpretive structural model(ISM) was constructed to analyze the hierarchical relationship of factors influencing DRG reform.Results:After analysis, the influencing factors(12) of DRG reform in China were included such as medical level, hospital management, and medical staff′s cognition and behavior. Among them, the local situation was the deep-level factor affecting DRG reform, 9 factors such as data quality assurance and policy design/implementation were the middle-level factors, and patients′ interests/needs and disease grouping were the surface-level factors.Conclusions:There were many influencing factors on the reform of China′s DRG payment system. It was suggested that relevant management departments in various regions should focus on the actual situation of the locality, take data quality and policy design and implementation as the key points of reform, formulate a scientific and reasonable DRG grouping scheme, safeguard the interests of patients, so as to promote the deepening of DRG reform.

Objective:Exploring the association rules of factors influencing high hospitalization costs for cancer patients, providing references for hospitals to optimize medical cost management measures.Methods:In the inpatient case information system of a tertiary general hospital, the medical record homepages of inpatients in the DRG groups of the oncology department in 2022 were obtained. The upper four scores of hospitalization costs was used as the threshold for patient grouping. Patients with hospitalization costs≥this threshold were the high-cost group, while other patients were control group; 12 factors, including age, gender, and admission condition, etc, were considered as potential influencing factors of high hospitalization costs. FP-Growth and Apriori algorithms were used to excavate the potential association rules between the influencing factors of high hospitalization costs. Logistic regression was used to analyze the independent influencing factors of high hospitalization costs.Results:A total of 5 512 hospitalized patients were included, including 1 378 patients in the high-cost group. Thirteen validated strong association rules for factors influencing high hospitalization costs were obtained, of which the rule antecedents included age (≥70 years), number of days in hospital (≥7 days), other diagnoses (≥5), surgery, planned readmission, use of antibiotics, admission (general/critical), living admission score (61~99), level of care (level 1/level 2), non-day ward, criticality during hospitalisation. Logistic regression results showed that all nine influencing factors except gender, use of antibiotics, and readmission plans were independent influences on high hospitalization costs ( P<0.05). Conclusions:The joint application of FP-Growth and Apriori algorithm could effectively explore the association rules of high hospitalization costs for oncology patients. The early warning information mainly included the number of hospitalization days, the number of other diagnoses, surgeries, and so on. It was suggested that medical institutions can reasonably control the high hospitalization costs through clinical pathway management, diagnosis and treatment process reengineering, admission risk assessment, and multidisciplinary collaborative diagnosis and treatment strategies.

Objective:To analyze the influencing factors of China′s DRG payment system reform(DRG reform) and its hierarchical relationship, for references for the in-depth promotion of China′s medical insurance payment reform.Methods:Relevant literature on DRG reform in China from databases such as CNKI, Wanfang Database, Pubmed, etc, were obtained. Content analysis method was used to extract the influencing factors of DRG reform. The correlation between each influencing factor was determined through expert discussion. An interpretive structural model(ISM) was constructed to analyze the hierarchical relationship of factors influencing DRG reform.Results:After analysis, the influencing factors(12) of DRG reform in China were included such as medical level, hospital management, and medical staff′s cognition and behavior. Among them, the local situation was the deep-level factor affecting DRG reform, 9 factors such as data quality assurance and policy design/implementation were the middle-level factors, and patients′ interests/needs and disease grouping were the surface-level factors.Conclusions:There were many influencing factors on the reform of China′s DRG payment system. It was suggested that relevant management departments in various regions should focus on the actual situation of the locality, take data quality and policy design and implementation as the key points of reform, formulate a scientific and reasonable DRG grouping scheme, safeguard the interests of patients, so as to promote the deepening of DRG reform.

DRG payment reform is an important means to control the unreasonable growth of medical expenses, improve the quality of medical services and achieve a win-win situation among three sides of hospitals, medical insurance and patients. This study adopted the Smith policy implementation process model to analyze the difficulties in the DRG policy implementation process from four aspects(idealized policies, policy implementation institutions, target groups, and policy environment), including the deviation between policy connotations and actual needs; the interest objectives of all parties were not completely aligned, the target group lacked a sense of identity, and the social impact and technological support needed to be improved. It was suggested that optimization should be carried out from four dimensions: policy supply coordination and precision, performance evaluation and personnel literacy, target group cognitive level and participation willingness, and policy implementation environment and atmosphere, in order to synergistically promote the effective implementation of DRG policies.

Objective:To master the current status of iodine nutrition of residents in Tongchuan City and provide basis for policy adjustment of relevant department.Methods:In 2019, the cross-sectional survey method was used to divide 5 sampling areas in the 4 districts (counties) of Tongchuan City according to the east, west, south, north, and middle, and each area sampled 1 township (street, including at least 1 street), each township (street) selected one elementary school, each elementary school selected 42 non-boarding students aged 8 - 10, and each township (street) selected 21 pregnant women as the survey subjects. Home salt samples and urine samples were collected, salt iodine and urinary iodine were monitored, and thyroid examinations were performed on students.Results:A total of 1 260 salt samples were collected, including 2 non-iodized salt and 1 240 qualified iodized salt; the median salt iodine was 24.32 mg/kg; the coverage rate of iodized salt, the qualified rate of iodized salt, and the edible rate of qualified iodized salt were 99.84% (1 258/1 260), 98.57% (1 240/1 258) and 98.41% (1 240/1 260), respectively. A total of 840 urine samples were collected from students, the median urinary iodine was 196.19 μg/L. The medians urinary iodine of students aged 8, 9, and 10 were 182.59, 222.16, and 190.36 μg/L, respectively. The median urinary iodine of male and female students were 211.27 and 186.76 μg/L. A total of 840 students were tested for thyroid, thyroid rate was 1.79% (15/840) by B-ultrasound. A total of 420 urine samples were tested of pregnant women. The median urinary iodine was 155.05 μg/L. The medians urinary iodine of pregnant women in the first, second and third trimester of pregnancy were 166.79, 176.11 and 129.82 μg/L, respectively. There was a statistically significant difference in urine iodine content of pregnant women in different pregnancy periods ( H = 9.317, P < 0.05). Conclusions:The iodine nutrition level of residents in Tongchuan City is generally suitable. Pregnant women's urinary iodine shows iodine deficiency in the third trimester. Monitoring of iodine nutrition of pregnant women should be strengthened, classified guidance and scientific iodine supplementation should be adhered, to ensure the appropriate level of iodine nutrition for pregnant women.

Objective To analyze the iodine nutritional status and the trend of children aged 8-10 years in Tongchuan City after the implementation of the new iodized salt standard,and to provide scientific basis for prevention and treatment of iodine deficiency disorders (IDD) in the region.Methods In 2013-2015,one town was selected respectively from 5 areas (east,west,south,north,center) in 4 counties of the city.A central primary school was selected in each sampled town,42 children aged 8-10 years in every school were selected for detection of the thyroid volume by palpation,and for collection of 15 urine samples for determination of urinary iodine (42 urine samples were collected in 2015).Four villages were selected from each town,15 residents were selected to determine salt iodide content by quantitative detection.Urinary iodine was tested using arsenic cerium catalytic spectrophotometry (WS/T 107-2006).Salt iodine was tested using direct titration method (the arbitration method was adopted for quantitative determination in the case of Sichuan salt or other special salts,GB/T 13025.7-2012).Results The thyroid palpate welling rate was 3.40％ (29/854),4.52％ (38/840) and 2.98％ (25/840) in children aged 8-10,respectively,and there was no statistical significant difference between different years (x2 =3.078,P ＞0.05).Totally,1 320 urine samples were collected from 8-10 years old children,the median of urinary iodine (MUI) was 185.14 μg/L;in the 3 years,the MUI in each year was 229.43,183.34 and 173.80 μg/L,respectively.The proportion of urinary iodine under 50 μg/L was less than 20％,under 100 μg/L was also far below 50％;urinary iodine proportion in 100 ～ ＜ 200 μg/L rose year by year.There were significant differences in the MUI among the 8 and 9 years age groups (H =12.736,10.128,P ＜ 0.05),there was no significant difference in the 10 years age group (H =3.849,P ＞ 0.05).In gender groups,there was significant difference in the MUI among male children (H =9.261,P＜ 0.05),there was no significant difference in the MUI among female children (H =4.759,P ＞ 0.05).The median of salt iodine was 24.10,24.75 and 24.10 mg/kg,respectively.The coverage rates of iodized salt were all higher than 95％,the qualified rates of iodized salt were all higher than 90％.Conclusions After implementation of the new standard iodized salt,the iodine level of children aged 8-10 years is at the appropriate level.The IDD surveillance indicators all meet the national standards for elimination of the disease.