The digitization of traditional Chinese medicine (TCM) has generated vast amounts of data. However, these data are characterized by significant heterogeneity and complex semantic structures, posing substantial challenges for systematic integration and intelligent analysis, and limiting its potential for modern clinical and computational research. To address the challenges posed by the high heterogeneity and complex structure in TCM data, we designed and developed the TCM Data Hub platform, which is powered by the YiYuan large language models (LLMs). This platform aims to enhance intelligent data processing capabilities and unlock the potential for clinical application of TCM data through systematic integration and efficient utilization, thereby bridging the gap between traditional knowledge and modern computational research. This study first analyzed the heterogeneity and complexity of TCM information with respect to data types, structures, and semantics. A standardized data framework was constructed to enhance data integration and interoperability. Based on the TCM Intelligent Computing Platform of the China Academy of Chinese Medical Sciences, we trained the YiYuan LLMs to acquire domain-specific semantic understanding of TCM, thereby improving the platform’s comprehension of specialized terminology and knowledge systems. Leveraging the natural language processing capabilities of the LLMs, we developed a human-in-the-loop data processing system to enable efficient extraction, cleansing, and structured organization of TCM data. In addition, utilizing Vue and Java technologies, we developed multiple LLM-powered intelligent agents and systems, including a human-in-the-loop data processing system, as well as automated prescription mining and network pharmacology analysis agents. Task-specific agents tailored to TCM data processing were developed to enhance the model’s effectiveness in clinical knowledge discovery. System functionality and platform infrastructure were implemented using Java and Vue technologies.The TCM Data Hub platform has completed system construction and core functionality implementation. It supported integrated management and efficient access to 8 key types of TCM data: prescriptions, materia medica, ingredients, targets, diseases (Western medicine), diseases (TCM), syndromes, and therapeutic methods. The human-in-the-loop data processing system achieved an accuracy of 95.34% in structuring TCM data and supported annotation for data requiring manual labeling. The intelligent agent-driven big-data analytics module enabled 1-click, end-to-end workflows for TCM prescription mining, herb-syndrome association analysis, network pharmacology, and molecular biology research, completing a full data mining task in approximately 30 minutes. Users can interact with and manipulate data through a visual front-end interface. The system demonstrated stable performance, strong scalability, and a user-friendly experience. Empowered by the YiYuan LLMs, the TCM Data Hub platform significantly improves the accessibility, usability, and intelligence of TCM data. It effectively bridges traditional TCM knowledge with modern intelligent technologies, providing robust data support and intelligent tools for TCM research and clinical applications.

Objective To address the need for rapid review of automated organ-at-risk (OAR) contouring for head-and-neck cancer in online adaptive radiotherapy, this study used dosimetric indices as the reference standard to evaluate the suitability of geometric similarity metrics and to develop a rapid screening model that balances the risk of missed errors with review efficiency. Methods This retrospective study included 29 patients with head-and-neck radiotherapy, yielding 243 pairs of OAR contours (auto-generated vs. manual). Geometric similarity metrics, including the Dice similarity coefficient (Dice), the 95th percentile Hausdorff distance (HD95), and the maximum Hausdorff distance (HD100), were computed between the two contour sets. Using the dose distributions from clinical treatment plans, dosimetric differences between the two contour sets were calculated for key indices, including mean dose (D_mean) and maximum dose (D_max). Correlations between geometric metrics and dosimetric differences were analyzed. Dosimetric discrepancy events were defined as Y_3Gy using a threshold of 3 Gy. Within a univariable logistic regression triage framework, each geometric metric was used as an independent variable to estimate the probability of Y_3Gy. Discriminative performance was assessed using receiver operating characteristic (ROC) curves and precision-recall (PR) curves, with the area under the curve (AUC) including ROC-AUC and PR-AUC. Practical operating thresholds were determined via threshold sweeping. Results Geometric similarity metrics showed weak-to-moderate correlations with dosimetric differences; compared with maximum dose, correlations with mean dose differences were more consistent. Among the evaluated metrics, HD95 achieved the best classification performance for Y_3Gy. Threshold sweeping suggested that an HD95 threshold in the range of 4-9 mm can balance the risk of missed discrepancies with the efficiency of automated contour quality assurance (QA). Conclusion Several geometric similarity metrics demonstrated only weak-to-moderate associations with dosimetric differences. For head-and-neck OAR contour QA in this study, HD95 provided the best discriminative performance and can support rapid triage of automated contours, with a practical operating threshold range of 4-9 mm, potentially improving the efficiency of adaptive radiotherapy workflows.

Objective To address the need for rapid review of automated organ-at-risk (OAR) contouring for head-and-neck cancer in online adaptive radiotherapy, this study used dosimetric indices as the reference standard to evaluate the suitability of geometric similarity metrics and to develop a rapid screening model that balances the risk of missed errors with review efficiency. Methods This retrospective study included 29 patients with head-and-neck radiotherapy, yielding 243 pairs of OAR contours (auto-generated vs. manual). Geometric similarity metrics, including the Dice similarity coefficient (Dice), the 95th percentile Hausdorff distance (HD95), and the maximum Hausdorff distance (HD100), were computed between the two contour sets. Using the dose distributions from clinical treatment plans, dosimetric differences between the two contour sets were calculated for key indices, including mean dose (D_mean) and maximum dose (D_max). Correlations between geometric metrics and dosimetric differences were analyzed. Dosimetric discrepancy events were defined as Y_3Gy using a threshold of 3 Gy. Within a univariable logistic regression triage framework, each geometric metric was used as an independent variable to estimate the probability of Y_3Gy. Discriminative performance was assessed using receiver operating characteristic (ROC) curves and precision-recall (PR) curves, with the area under the curve (AUC) including ROC-AUC and PR-AUC. Practical operating thresholds were determined via threshold sweeping. Results Geometric similarity metrics showed weak-to-moderate correlations with dosimetric differences; compared with maximum dose, correlations with mean dose differences were more consistent. Among the evaluated metrics, HD95 achieved the best classification performance for Y_3Gy. Threshold sweeping suggested that an HD95 threshold in the range of 4-9 mm can balance the risk of missed discrepancies with the efficiency of automated contour quality assurance (QA). Conclusion Several geometric similarity metrics demonstrated only weak-to-moderate associations with dosimetric differences. For head-and-neck OAR contour QA in this study, HD95 provided the best discriminative performance and can support rapid triage of automated contours, with a practical operating threshold range of 4-9 mm, potentially improving the efficiency of adaptive radiotherapy workflows.

Objective To address the need for rapid review of automated organ-at-risk (OAR) contouring for head-and-neck cancer in online adaptive radiotherapy, this study used dosimetric indices as the reference standard to evaluate the suitability of geometric similarity metrics and to develop a rapid screening model that balances the risk of missed errors with review efficiency. Methods This retrospective study included 29 patients with head-and-neck radiotherapy, yielding 243 pairs of OAR contours (auto-generated vs. manual). Geometric similarity metrics, including the Dice similarity coefficient (Dice), the 95th percentile Hausdorff distance (HD95), and the maximum Hausdorff distance (HD100), were computed between the two contour sets. Using the dose distributions from clinical treatment plans, dosimetric differences between the two contour sets were calculated for key indices, including mean dose (D_mean) and maximum dose (D_max). Correlations between geometric metrics and dosimetric differences were analyzed. Dosimetric discrepancy events were defined as Y_3Gy using a threshold of 3 Gy. Within a univariable logistic regression triage framework, each geometric metric was used as an independent variable to estimate the probability of Y_3Gy. Discriminative performance was assessed using receiver operating characteristic (ROC) curves and precision-recall (PR) curves, with the area under the curve (AUC) including ROC-AUC and PR-AUC. Practical operating thresholds were determined via threshold sweeping. Results Geometric similarity metrics showed weak-to-moderate correlations with dosimetric differences; compared with maximum dose, correlations with mean dose differences were more consistent. Among the evaluated metrics, HD95 achieved the best classification performance for Y_3Gy. Threshold sweeping suggested that an HD95 threshold in the range of 4-9 mm can balance the risk of missed discrepancies with the efficiency of automated contour quality assurance (QA). Conclusion Several geometric similarity metrics demonstrated only weak-to-moderate associations with dosimetric differences. For head-and-neck OAR contour QA in this study, HD95 provided the best discriminative performance and can support rapid triage of automated contours, with a practical operating threshold range of 4-9 mm, potentially improving the efficiency of adaptive radiotherapy workflows.

ObjectiveTo explore the construction of a machine learning model for the diagnosis of traditional Chinese medicine （TCM） syndromes in chronic cough and the optimization of this model using the Voting ensemble algorithm. MethodsA retrospective analysis was conducted using clinical data from 921 patients with chronic cough treated at the Respiratory Department of Dongfang Hospital， Beijing University of Chinese Medicine. After standardized processing， 84 clinical features were extracted to determine TCM syndrome types. A specialized dataset for TCM syndrome diagnosis in chronic cough was formed by selecting syndrome types with more than 50 cases. The synthetic minority over-sampling technique （SMOTE） was employed to balance the dataset. Four base models， logistic regression （LR）， decision tree （dt）， multilayer perceptron （MLP）， and Bagging， were constructed and integrated using a hard voting strategy to form a Voting ensemble model. Model performance was evaluated using accuracy， recall， precision， F1-score， receiver operating characteristic （ROC） curve， area under the curve （AUC）， and confusion matrix. ResultsAmong the 921 cases， six syndrome types had over 50 cases each， phlegm-heat obstructing the lung （294 cases）， wind pathogen latent in the lung （103 cases）， cold-phlegm obstructing the lung （102 cases）， damp-heat stagnating in the lung （64 cases）， lung yang deficiency （54 cases）， and phlegm-damp obstructing the lung （53 cases）， yielding a total of 670 cases in the specialized dataset. High-frequency symptoms among these patients included cough， expectoration， odor-induced cough， throat itchiness， itch-induced cough， and cough triggered by cold wind. Among the four base models， the MLP model showed the best diagnostic performance （test accuracy： 0.9104； AUC： 0.9828）. Compared with the base models， the Voting ensemble model achieved superior performance with an accuracy of 0.9289 on the training set and 0.9253 on the test set， showing a minimal overfitting gap of 0.0036. It also achieved the highest AUC （0.9836） in the test set， outperforming all base models. The model exhi-bited especially strong diagnostic performance for damp-heat stagnating in the lung （AUC： 0.9984） and wind pathogen latent in the lung （AUC： 0.9970）. ConclusionThe Voting ensemble algorithm effectively integrates the strengths of multiple machine learning models， resulting in an optimized diagnostic model for TCM syndromes in chronic cough with high accuracy and enhanced generalization ability.

Objective To analyze the hotspots and trends in the research field of inflammation in polycystic ovary syndrome(PCOS).Methods This study retrieved data from the Web of Science Core Collection database,with the retrieval period spanning from January 2000 to December 2023.The litera-ture types included were papers and review papers in English.The search was conducted using"Poly-cystic Ovarian Syndrome"and its synonyms,along with"inflammation"as subject terms.Visualization analysis was performed using software such as CiteSpace,VOSviewer,and SCImago Graphica.Results A total of 1,803 articles were included in this study.The number of publications in this field had rapidly increased after 2018.China had the highest number of publications,and Tehran University of Medical Sciences was the institution with the most publications.Key scholars included Asemi Z,Gonzalez AM,Escobar-Morreale HF,etc.,and a core group of authors had been formed.The main research hotspots centered around insulin resistance,obesity,oxidative stress,etc.,and five clusters were formed.Highly cited and burst-cited literature were mainly review papers.Conclusion The re-search field of inflammation in PCOS is currently in a stage of rapid development,with research content covering PCOS inflammation markers,the impact of inflammation on PCOS pathophysiology,inflamma-tion-based PCOS treatment and prognosis,etc.The debate on whether the chronic low-grade inflammatory state in PC OS patients is related to obesity or not remains a focal point of contention.

Objective:To investigate the clinical effect of microplasma radiofrequency combined with oral traditional Chinese medicine (TCM) Babao Dan capsule to improve damp-heat TCM constitution in the treatment of hypertrophic scar.Methods:A prospective randomized controlled trial method was used to recruit patients diagnosed with damp-heat TCM constitution and suffering from hypertrophic scar who admitted to the Fourth Medical Center of Chinese PLA General Hospital from March to July 2024. All patients were randomly divided into microplasma radiofrequency combined with TCM treatment group (combination group), microplasma radiofrequency (RF) treatment group (RF group), and oral TCM treatment group (TCM group). Both the combination group and the RF group were treated with a microplasma radiofrequency instrument. Transverse, longitudinal and oblique scar grinding were performed three times with 90 W power, and regular scabs appeared in the treatment area as the treatment end point. After RF treatment in the combination group and the TCM group, Babao Dan capsule was taken with regular warm water, 0.6 g each time, 3 times/day, for 3 months. The Vancouver scar scale (VSS) score (higher score indicating more severe scar), visual analogue scale (VAS) of scar pain and itch (higher score indicating more severe symptom), and TCM constitution score (higher score indicating more severe moist heat constitution) were compared before treatment and 3 months after the first treatment. SPSS 26.0 software was used for statistical analysis. Measurement data conforming to normal distribution were expressed as Mean±SD, ANOVA was used for comparison among groups, and LSD- t method was used for pair comparison between groups. Paired sample t test was used before and after treatment in the same group. Counting data were expressed as cases (%), and the χ2 test was used for inter-group comparison. P<0.05 was considered statistically significant. Results:A total of 60 patients were included, 20 in each group. There were no significant differences in gender distribution, disease course, scar area and distribution among the three groups ( P>0.05), indicating comparability. Before treatment, there were no statistically significant differences in VSS, VAS and TCM physique scores among the 3 groups ( P>0.05). After treatment, VAS scores in the 3 groups, VSS scores in combination group and RF group, TCM physique scores in combination group and TCM group were improved compared with those before treatment, and the differences were statistically significant ( P<0.05). The VSS score of the combination group was significantly lower than that of the RF group and the TCM group ( P<0.05), but there was no statistical significance between the RF group and the TCM group ( P>0.05). The pruritus VAS score of the combination group was significantly lower than that of TCM group and RF group ( P<0.05), but there was no statistical significance between RF group and TCM group ( P>0.05). There was no significant difference in pain VAS scores among the three groups after treatment ( P>0.05). The TCM physique scores of the combination group and TCM group were significantly lower than those of the RF group ( P<0.01), but there was no statistical significance between the combination group and TCM group ( P>0.05). Conclusion:Microplasma radiofrequency combined with oral Babao Dan capsule can effectively improve the VSS score, pain and pruritus of hypertrophic scars in patients with damp-heat constitution, and the effect is better than that of radiofrequency or TCM alone.

Objective:To identify potential therapeutic targets for the treatment of hair color and hair shaft abnormalities, providing novel insights and approaches for managing related conditions.Methods:Using the pQTL data from large-scale proteomics studies, a two-sample Mendelian randomization (TwoSampleMR) was conducted to preliminarily identify potential drug therapeutic targets. Subsequently, sensitivity analysis was performed to evaluate potential confounding factors such as heterogeneity and horizontal pleiotropy, and a leave-one-out sensitivity analysis was conducted by eliminating each single nucleotide polymorphism (SNP) one by one. Finally, co-localization analysis was carried out to explore whether there are shared genetic variants between the identified plasma proteins and traits.Results:The proteomic data used in this study included 4, 907 pQTLs, while the genetic data related to hair pigmentation and shaft abnormalities comprised 124 cases and 432, 686 controls. After Mendelian randomization screening, six candidate protein genes were identified: HLA-DQA2, CTSB, KIR2DS2, SVEP1, HOMER2, and HOMER1. Sensitivity analyses revealed no evidence of heterogeneity or horizontal pleiotropy among these proteins. Leave-one-out sensitivity analysis indicated that no single SNP significantly influenced the overall result. Notably, HOMER2 was supported by colocalization analysis with strong evidence, suggesting its potential role in regulating genetic mechanisms underlying hair pigmentation and shaft health.Conclusions:This study identified six potential therapeutic targets for hair pigmentation and shaft abnormalities, with HOMER2 showing stronger evidence. These findings provide novel directions for the treatment of hair color and hair shaft abnormalities.

Objective To investigate the predictive value of triglyceride glucose-body mass index(TyG-BMI),serum triglyceride-to-high-density lipoprotein cholesterol ratio(TG/HDL-C),and metabolic score for insulin resistance(METS-IR)for fatty pancreatic disease(FPD).Methods A total of 240 patients with FPD treated in The First Affiliated Hospital of Zhengzhou University from January 2020 to November 2023 were included as the case group,while 480 healthy subjects who underwent healthy checks in the same period were randomly selected as the control group.General clinical data and laboratory indicators were collected.The Mann-Whitney U test and chi-square test were used to compare non-normally distributed continuous variables,and categorical variables between groups,respectively.A binary logistic regression model was used to assess the relationship between TyG-BMI,TG/HDL-C,and METS-IR and FPD.The receiver operating characteristic(ROC)curve was plotted,and the area under the curve(AUC)was calculated to evaluate the predictive diagnostic value of those simplified insulin resistance indicators for FPD in the general population and different sex populations.Results Age,BMI,systolic blood pressure,diastolic blood pressure,fasting plasma glucose,uric acid,alanine aminotransferase,aspartate aminotransferase,gamma-glutamyl transferase,total cholesterol,triglyceride,low-density lipoprotein cholesterol,TyG-BMI,TG/HDL-C,and METS-IR in the case group were significantly higher than those in the control group(all P<0.05).The case group had significantly higher proportions of individuals with hypertension,diabetes,and fatty liver disease than the control group(all P<0.05).The high-density lipoprotein cholesterol level was significantly lower in the case group than in the control group(P<0.05).The multivariable Logistic regression analysis showed that after adjusting for various influencing factors,TyG-BMI,TG/HDL-C,and METS-IR remained as independent risk factors for the development of FPD,with the odds ratios(95%confidence intervals)being 1.027(1.018-1.037),6.964(2.022-23.989),and 1.184(1.123-1.248),respectively.In the ROC curve analysis,the AUCs of METS-IR and TyG-BMI were 0.823 and 0.803,respectively,with their sensitivities being 76.3%and 75.8%,specificities being 74.6%and 71.7%,and optimal cut-off values being 34.86 and 196.70,respectively;the next were BMI(AUC=0.758)and TG/HDL-C(AUC=0.734);in the sex-stratified analysis,the AUC values of METS-IR were highest in both the male and female subgroups,which were 0.834 and 0.810,respectively.Conclusion TyG-BMI,TG/HDL-C,and METS-IR show good predictive value for the development of FPD,in which METS-IR is more excellent.

Objective To investigate the proportion of Th17 cells,Th1-like Th17 cells and FoxP3+Th17 cells in bone marrow of patients with myelodysplastic syndrome(MDS),the expression of interleukin-17A(IL-17A)in bone marrow supernatant and its clinical significance.Methods Forty MDS patients(MDS group)and 18 patients with nutritional anemia(control group)were selected.MDS patients were classified into the low blast(MDS-LB)group(19 cases)and the increased blast(MDS-IB)group(21 cases,including 11 cases of type IB1 and 10 cases of type IB2)based on morphological definition.The MDS patients were scored according to the revised International Prognostic Scoring System(IPSS-R),with 18 cases in the low-risk group(≤4.5)and 22 cases in the high-risk group(>4.5).Flow cytometry was used to detect the proportion of Th17 cells,Th1-like Th17 cells and FoxP3+Th17 cells in bone marrow of the MDS group and the control group.Enzyme-linked immunosorbent assay(ELISA)was used to detect the level of IL-17A in bone marrow supernatant of the above samples.Results The proportion of Th17 cells and the level of IL-17A were higher in patients of the MDS group than those in the control group(P＜0.05).According to the median expression level of IL-17A,the MDS group was divided into the low-expression group(＜13.71 ng/L,20 cases)and the high-expression group(≥13.71 ng/L,20 cases).Compared with the low-expression group,there were higher proportion of patients with blast cells＜5%and low-risk patients(P＜0.05)in the high-expression group.Compared with the IL-17A low-expression group,the IL-17A high-expression group had a higher proportion of patients with blast cells＜5%and relatively low-risk patients(P＜0.05).Compared with the low-risk patients,high-risk patients had a lower proportion of Th17 cells,IL-17A levels and Th1-like Th17 cells,and a higher proportion of FoxP3+Th17 cells(P＜0.05).Compared with the MDS-LB group,the MDS-IB group had a lower proportion of Th17 cells,IL-17A levels and Th1-like Th17 cells,and a higher proportion of FoxP3+Th17 cells(P＜0.05).Conclusion The proportion of Th17 cells and the level of IL-17A are significantly increased in MDS patients.The decreased proportion of Th1-like Th17 cells and the increased proportion of FoxP3+Th17 cells may be related to the increased proportion of blast cells and higher risk stratification in patients.