ObjectiveThis study aims to develop and validate a novel multimodal predictive model， termed criss-cross network（CCNet）-directed graph neural network（DGNN）（CGN）， for accurate assessment of pulmonary nodule progression in high-risk individuals for lung cancer， by integrating longitudinal chest computed tomography（CT） imaging with both traditional Chinese and western clinical evaluation data. MethodsA cohort of 4 432 patients with pulmonary nodules was retrospectively analyzed. A twin CCNet was employed to extract spatiotemporal representations from paired sequential CT scans. Structured clinical assessment and imaging-derived features were encoded via a multilayer perceptron， and a similarity-based alignment strategy was adopted to harmonize multimodal imaging features across temporal dimensions. Subsequently， a DGNN was constructed to integrate heterogeneous features， where nodes represented modality-specific embeddings and edges denoted inter-modal information flow. Finally， model optimization was performed using a joint loss function combining cross-entropy and cosine similarity loss， facilitating robust classification of nodule progression status. ResultsThe proposed CGN model demonstrated superior predictive performance on the held-out test set， achieving an area under the receiver operating characteristic curve（AUC） of 0.830， accuracy of 0.843， sensitivity of 0.657， specificity of 0.712， Cohen's Kappa of 0.417， and F₁ score of 0.544. Compared with unimodal baselines， the CGN model yielded a 36%-48% relative improvement in AUC. Ablation studies revealed a 2%-22% increase in AUC when compared to simplified architectures lacking key components， substantiating the efficacy of the proposed multimodal fusion strategy and modular design. Incorporation of traditional Chinese medicine （TCM）-specific symptomatology led to an additional 5% improvement in AUC， underscoring the complementary value of integrating TCM and western clinical data. Through gradient-weighted activation mapping visualization analysis， it was found that the model's attention predominantly focused on nodule regions and effectively captured dynamic associations between clinical data and imaging-derived features. ConclusionThe CGN model， by synergistically combining cross-attention encoding with directed graph-based feature integration， enables effective alignment and fusion of heterogeneous multimodal data. The incorporation of both TCM and western clinical information facilitates complementary feature enrichment， thereby enhancing predictive accuracy for pulmonary nodule progression. This approach holds significant potential for supporting intelligent risk stratification and personalized surveillance strategies in lung cancer prevention.

Non-alcoholic fatty liver disease is a common chronic liver disease in clinical practice. In recent years, with increasing social attention to the health of women and the elderly, the prevention and treatment of non-alcoholic fatty liver disease after menopause has increasingly become a research hotspot in metabolic diseases. This study explores the pathogenesis and treatment method of non-alcoholic fatty liver disease in postmenopausal women based on the theory of "deficient qi and stagnation" and combined with the physiological and pathological characteristics of postmenopausal women and the Western medicine understanding of non-alcoholic fatty liver disease. We believe that the pathogenesis of non-alcoholic fatty liver disease in postmenopausal women is rooted in the "deficient qi" caused by depletion of liver and kidney essence and blood. The imbalance between the physical and functional aspects of the liver due to this "deficient qi" is the primary factor, while the "stagnation" of phlegm and blood stasis is the manifestation. Furthermore, the "deficient qi" and "stagnation" reinforce each other, with the deficiency leading to stagnation and stagnation exacerbating the deficiency, thus accelerating the progression of the disease. The treatment approach should be one that combines nourishing deficiency and resolving stagnation, addressing both root cause and maifestations. Given the female characteristic of "the liver as the innate organ" and the post-menopausal physiological state of "gradual decline of kidney essence", it is important to focus on nourishing the liver and kidneys, nurturing the liver′s physical body while maintaining its function, and also promoting the circulation of qi, resolving phlegm, and invigorating blood circulation to remove blood stasis. This approach aims to reduce the accumulation of lipids in the liver, offering a new perspective and approach for the treatment of non-alcoholic fatty liver disease in post-menopausal women with traditional Chinese medicine.

ObjectiveIn order to improve the recognition accuracy of named entities in medical record texts and realize the effective mining and utilization of medical record knowledge， a Bert-Radical-Lexicon（BRL） neural network model is constructed to recognize medical record entities with respect to the characteristics of medical record texts. MethodWe selected 408 medical records related to hypertension from the the Complete Library of Famous Medical Records of Chinese Dynasties and constructed a dataset consisting of 1 672 medical records by manually labeling. Then， we randomly divided the dataset into three subsets， including the training set（1 004 cases）， the testing set （334 cases） and the validation set（334 cases）. Based on this dataset， we built a BRL model that fused various text features of medical records， as well as its variants BRL-B， BRL-L and BRL-R， and a baseline model Base for experiments. During the model training phase， we trained the above models using the training set to reduce the risk of overfitting. We continuously monitored the performance of each model on the validation set during training and saved the model with the best performance. Finally， we evaluated the performance of these models on the testing set. ResultCompared with other models， the BRL model had the best performance in the medical records named entity recognition task， with an overall recognition precision of 90.09%， a recall of 90.61%， and the harmonic mean of the precision and recall（F1） of 90.35% for eight types of entities， including disease， symptom， tongue manifestation， pulse condition， syndrome， method of treatment， prescription and traditional Chinese medicine（TCM）. Compared with the Base model， the BRL model improved the overall F1 value of entity recognition by 5.22%， and the F1 value of pulse condition entity increased by 6.92%， which was the largest increase. ConclusionBy incorporating a variety of medical record text features in the embedding layer， the BRL neural network model has stronger named entity recognition ability， and thus extracts more accurate and reliable TCM clinical information.

ObjectiveTo investigate the trends and significant determinants of delayed HIV diagnosis （DHD） among newly reported HIV/AIDS cases in Xuhui District， Shanghai between 2018 and 2022. MethodsIn the newly reported HIV/AIDS cases， patients died within one year without accident， HIV/AIDS cases with CD₄ cell count <200 cells·μL^-1， and AIDS cases with a CD₄ cell count between 200 to 499 cells·μL^-1 were defined as delayed diagnosis. Univariate and multivariate logistic analysis were employed to explore the influencing factors of DHD. ResultsAmong the 862 newly reported HIV/AIDS cases， The DHD rate was 39.79% without statistically significant difference by year（χ²=4.508， P=0.342）. Patients with CD₄ cell count <200 cells·μL^-1 contributed the largest proportion of DHD. During 2018‒2022， the DHD rate declined among HIV/AIDS patients who were younger than 35 years old or 45‒65 years old， never married， original diagnosis from tertiary specialized hospitals. Patients who were 65 years or older， married or divorced， with heterosexual transmitted HIV/AIDS， and original diagnosis from other types of testing and tertiary metropolitan hospital， had sustainably higher DHD rates. The number of HIV screening and diagnosed from voluntary counseling and testing （VCT） decreased during the COVID-19 epidemic， while the DHD rate increased sharply. Multivariate logistic regression analysis suggested the DHD rates were higher among older age， other types of testing（OR=3.805， 95%CI： 2.260‒6.406）and pre-operative testing（OR=2.411， 95%CI： 1.424‒4.081）. Patients who received CD₄ test in 15 days had a higher DHD rate compared to the cases received CD₄ test exceeding 90 days （OR=0.336， 95%CI： 0.216‒0.522）. ConclusionThere is no significant decrease of delayed HIV diagnosis rate in Xuhui District in recent years， and the number of HIV tests has decreased in 2022. Monitoring of newly reported HIV/AIDS should be conducted continuously. Expansion of HIV antibody screening should be conducted in non-infectious departments and inpatient departments in healthcare institutions， particularly metropolitan hospitals. Assistance should be provided to clinicians and elderly patients for improving their ability to recognize and perceive the risk of HIV/AIDS， in order to enhance early diagnosis and subsequent treatment.

Objective To compare the outcomes of transvaginal natural orifice transluminal endoscopic surgery(vNOTES)and laparoendoscopic single-site surgery(LESS)for total uterine excision in obese patients with benign uterine lesions,and to investigate the utility of vNOTES in this patient population.Methods A total of 100 obese patients(BMI>28.0 kg/m2)diagnosed with benign uterine lesions requiring total uterine and bilateral salpingectomy between January 2022 and January 2023 were included in this study.They were randomly assigned to two groups:the LESS group(n=51)and the vNOTES group(n=49).Patient demographics,surgical duration,intraoperative blood loss,changes in hemoglobin levels,pain scores,time to first flatus postoperatively,length of hospital stay,pelvic floor function,sexual quality of life,and postoperative complications were compared between the two groups.Results The two groups did not show any statistically significant differences in terms of blood loss,pre-and postoperative hemoglobin changes,pelvic floor function,sexual quality of life,or postoperative complications(P>0.05).However,the vNOTES group exhibited shorter surgical durations,time to first flatus postoperatively,and length of hospital stay compared to the LESS group(P<0.05).Additionally,the vNOTES group demonstrated lower intraoperative pain scores than the LESS group.(P<0.05).Conclusions In obese patients with benign uterine lesions,vNOTES total uterine excision surgery demonstrated shorter surgical durations and postoperative hospital stays,lower postoperative pain scores,and better adherence to the principles of en-hanced recovery after surgery(ERAS),indicating its potential for broader application.

Objective This study aimed to investigate the effects of silencing Ras homolog family member C(RhoC)on the proliferation,apoptosis,invasion,migration,and epithelial-mesenchymal transition(EMT)of salivary adenoid cystic carcinoma(SACC)and its molecular mechanisms.Methods A total of 27 SACC lesions and normal salivary gland tissues that were surgically resected at Qingdao Municipal Hospital from January 1,2019 to March 1,2024 were selected,and the expression levels of RhoC were detected by Western blot and immunohistochemistry.Three small interfering RNA(siRNAs)were designed to target the RhoC gene sequence,transfected into SACC-LM and SACC-83 cell lines,and evaluated for transfection efficiency.The protein expression levels of RhoC,Rho-asso-ciated protein kinase-1(ROCK1),p38 mitogen-activated protein kinase(p38MAPK),phosphorylated-p38MAPK(p-p38MAPK),twist family bHLH transcription factor 1(TWIST1),E-cadherin,N-cadherin,and Vimentin were com-pared using Western blot.CCK-8 assay,flow cytometry,transwell invasion assay,and wound healing assay were conducted to assess the differences in cell proliferation,apoptosis,invasion,and migration abilities among the groups.Bioinformatics methods were also used to predict possible upstream micro RNAs(miRNAs)of RhoC and their expression levels in SACC.Moreover,dual-luciferase reporter gene experiments were performed to verify the binding sites of miR-138-5p and RhoC.Results RhoC was highly expressed in SACC(P＜0.05).After silencing RhoC,the test group showed a significant decrease in the expression level of ROCK1,p-p38MAPK,TWIST1,N-cadherin,and Vimentin,as well as a significant increase in the expression level of E-cadherin(P＜0.05).No signifi-cant difference in the expression level of p38MAPK was observed(P＞0.05).The cell proliferation,invasion,and mi-gration ability decreased in the test group,whereas the apoptosis rates significantly increased(P＜0.05).miR-138-5p was lowly expressed in SACC,and miR-138-5p mimic can significantly downregulated the luciferase activity of 293T cells after transfection with a RhoC wild-type plasmid(P＜0.05).Conclusion RhoC is highly expressed in SACC,and RhoC silencing may target the downstream ROCK1/p38MAPK/TWISTl signaling pathway,thereby in-hibiting the proliferation,invasion,migration,and EMT of SACC while promoting its apoptosis.On the contrary,miR-138-5p is lowly expressed in SACC and is a potential upstream gene of RhoC,and there may be binding sites between the two genes.

Objective To construct a nomogram risk prediction model for symptomatic cerebrovas-cular ischaemia based on intra-plaque neovascularisation in carotid arteries in patients with ische-mic cerebrovascular disease(ICVD).Methods A retrospective study was conducted on 320 ICVD patients who were consecutively admitted to Wuhan Fourth Hospital from April 2020 to April 2024.In a ratio of 3∶1,240 cases were assigned into a training set and 80 cases into a validation set.The patients in the training set were further divided into 147 cases in the symptomatic sub-group and 93 cases in the asymptomatic subgroup according to the presence or absence of relevant symptoms or signs.Multifactorial logistic regression analysis was used to identify the risk factors for symptomatic ischemia in ICVD patients,and a nomogram risk prediction model was construc-ted and the prediction efficacy of the model was evaluated.Results The proportions of plaque multiplicity,ulcerated plaque,stenosis ≥70％,and intra-plaque neovascularization were signifi-cantly higher in the symptomatic subgroup than the asymptomatic subgroup(P＜0.05,P＜0.01).Multifactorial logistic regression analysis showed that plaque multiplicity(OR=1.261,95％CI:1.088-1.539,P=0.003),ulcerated plaque(OR=1.458,95％CI:1.132-1.661,P=0.001),and stenosis ≥70％(OR=2.023,95％CI:1.458-2.561,P=0.001),and intra-plaque neovasculariza-tion(OR=1.206,95％CI:1.057-1.489,P=0.002)were independent risk factors for the occur-rence of symptomatic cerebral ischemia in ICVD patients.H-L deviation test showed that the con-structed nomogram risk prediction model had a good fit(x2=9.362,P=0.295).Internal and ex-ternal validation showed that the calibration curves for both the training and validation sets were consistent with the original curves,and the AUC value was were 0.871 and 0.864,respectively.De-cision curve analysis showed that the model had a significant standardized clinical net benefit when the predicted risk threshold exceeded 0.01.Conclusion Ultrasonography is instructive in as-sessment of the presence or absence of intra-plaque neovascularization in carotid atherosclerotic plaques.And our constructed nomogram risk prediction model has good predictive value for the development of symptomatic cerebral ischemia.

ObjectiveTo investigate the identification of kidney Yang deficiency syndrome of patients with osteoporosis（OP）， and to form the clinical syndrome identification rules of traditional Chinese medicine（TCM）. MethodBasic information， etiology， clinical symptoms and other characteristics of 982 OP patients were included， and statistical tests were used to screen the variables associated with kidney Yang deficiency syndrome. Taking the decision tree as the base model， bootstrap aggregation algorithm（Bagging algorithm） was utilized to establish the classification model of kidney Yang deficiency syndrome in OP， generating numerous rules and removing redundancy. Combining least absolute shrinkage and selection operator（LASSO） regression to screen key rules and integrate them to construct an identification model， achieving the identification of kidney Yang deficiency syndrome in OP patients. ResultEighteen key identification rules were screened out， and of these， where 11 rules with regression coefficients>0 correlated positively with the kidney Yang deficiency syndrome， the rule with the highest coefficient was chilliness（present）&feverish sensation over the palm and sole（absent）. The other 7 rules with regression coefficients<0 correlated negatively with the syndrome， the rule with the lowest coefficient was reddish tongue（present）&diarrhea（absent）&deficiency of endowment（absent）. According to the regression coefficients of each key rule， variables with importance>0.2 were ranked as chilliness， reddish tongue， feverish sensation over the palm and sole， cold limbs， clear urine， diarrhea， deficiency of endowment， prolonged illness. The results of the partial dependence analysis of the identification model showed that compared to OP patients without chilliness， those with chilliness（present） had a 0.266 8 higher probability of being identified as having kidney Yang deficiency syndrome， indicating that this variable had the highest impact on identification of the syndrome. Similarly， compared to OP patients without reddish tongue， those with reddish tongue had a 0.141 9 lower probability of being identified as having kidney Yang deficiency syndrome， indicating that this variable had the highest impact on identifying non-kidney Yang deficiency syndrome. The accuracy， sensitivity， specificity and area under receiver operating characteristic curve（AUC） of the established kidney Yang deficiency syndrome identification model in the test set were 0.865 9， 0.853 7， 0.872 0 and 0.931 5， respectively. ConclusionA precise identification model of OP kidney Yang deficiency syndrome is conducted basing on the rule ensemble method of Bagging combining LASSO regression， and the screened key rules can explain the identification process of kidney Yang deficiency syndrome. In this research， according to the regression coefficients of rules， the importance and partial dependence of variables， combined with the thinking of TCM， the influence of patient characteristics on the identification of syndromes is described， so as to reveal the primary and secondary syndromes of identification and assist the clinical identification of kidney Yang deficiency syndrome.

ObjectiveTo construct a quantitative differentiation model of traditional Chinese medicine （TCM） syndromes by taking primary osteoporosis （POP） with kidney yang deficiency syndrome as an example， and to provide methodological reference for the standardization of syndrome differentiation. MethodsHigh-frequency clinical features of POP were screened by descriptive statistical analysis， and strong association features of POP were obtained by association rule algorithm. On this basis， a latent structure （latent tree） model was established through latent structure analysis， and the implicit and explicit variables （features） related to POP with kidney yang deficiency syndrome were comprehensively clustered， and the clustering results were interpreted by the indexes of mutual information and cumulative information coverage， to explore the primary and secondary symptoms， and to deduce the categories of POP with kidney yang deficiency syndrome based on the probability of the features appearing in the various latent categories. Based on the categories， the clinical feature scores and identification thresholds were calculated， and the syndrome differentiation model of POP with kidney yang deficiency was initially constructed by combining the comprehensive judgment rules. Finally， the results of TCM professionals' judgment were used as the gold standard to further evaluate the effectiveness of the model in assisting the syndrome differentiation. ResultsThe 32 features strongly associated with POP were obtained， and the Bayes information critedon score of the further constructed latent tree model was -15291.93. Based on the mutual information and the cumulative information coverage， the main symptoms of POP with kidney yang deficiency syndrome were bone weakness， fatigue， pale tongue， clear urine， frequent nocturnal urination， cold limbs， thin pulse， white coating， and secondary symptoms were weakness， loss of libido， loose stool， frequent urination， lumbar and knee weakness， and fear of cold. From the probability of the occurrence of each clinical feature in different latent categories of POP with kidney yang deficiency syndrome， the state was introduced as S0 category （none/mild kidney yang syndrome）/ S1 category （moderate kidney yang syndrome）/ S2 category （severe kidney yang syndrome）. Optimizing the preliminary rules of state identification and refining the state of S1 category， the results showed that among 970 patients with POP， there were 520 patients having no/mild kidney yang deficiency syndrome， 224 patients with moderate to mild kidney yang deficiency syndrome， 81 patients with moderate to severe kidney yang deficiency syndrome， and 145 patients with severe kidney yang deficiency syndrome. During the evaluation and validation process， the correct rate of the model assessment index was 0.8835， while the sensitivity was 0.7181， and the specificity was 0.9437. ConclusionCombined with the latent structure analysis of the association rule， the syndrome differentiation model for POP with kidney yang deficiency could be constructed， and the model shows a good quantitative identification effect， which can provide methodological supports for clinicians to improve the efficiency and accuracy of TCM diagnosis.

OBJECTIVE: Pseudostellaria heterophylla has been paid more attention in recent years, mainly as a medicine food homology plant. The content determination of P. heterophylla is not specified in the Chinese Pharmacopoeia (version 2020). The environmental conditions in different production areas could exert an influence on the quality of P. heterophylla. The purpose of this study is to discriminate P. heterophylla collected from different geographical origins of China. METHODS: In this study, the content of polysaccharide in 28 batches of P. heterophylla was determined using phenol-sulfuric acid. HPLC fingerprints were established under optimised HPLC-PDA methods. Subsequently, the similarity analysis (SA) and the quantification of heterophyllin B were analyzed. The metabolites of P. heterophylla were identified and evaluated using UHPLC-Q Exactive HF orbitrap MS system. Principal component analysis (PCA), partial least squares discriminant analysis (PLS-DA), hierarchical cluster analysis (HCA) and orthogonal PLS-DA (OPLS-DA) were performed based on all peak areas. RESULTS: The polysaccharide content in Guizhou and Jiangsu was higher than that of other production areas, which varied significant from different origins. While the content of heterophyllin B in Anhui and Jiangsu was high. The correlation coefficients of HPLC fingerprints for 28 batches samples ranged from 0.877 to 0.990, and the characteristic map can be used to identify and evaluate the quality of P. heterophylla. The samples from Fujian, Guizhou, Jiangsu provinces can be relatively separated using multivariate statistical analysis including PCA, PLS-DA, HCA, OPLS-DA, indicating that their metabolic compositions were significantly different. Ultimately, a total of 15 metabolites which were filtrated by a VIP-value > 1 and a P-value < 0.05 associated with the separation of different origins were identified. CONCLUSION HPLC fingerprint was established to evaluate the quality and authenticity of P. heterophylla. The present work showed that the difference of geographic distributions had an influence on the internal chemical compositions. A sensitive and rapid untargeted metabolomics approach by UHPLC-Q Exactive HF orbitrap MS was utilized to evaluate P. heterophylla from different origins in China for the first time. Overall, this study provides insights to metabolomics of P. heterophylla and supplies important reference values for the development of functional foods.