Objective:To explore the association between sublingual microcirculatory perfusion vessel proportion (PPV), high mobility group protein B1 (HMGB1), and sepsis-induced coagulopathy (SIC), and to identify early predictive markers for clinical intervention.Methods:A total of 66 septic patients admitted to the Respiratory Intensive Care Unit (RICU) between November 2021 and May 2024 were enrolled. Based on SIC diagnosis within 24 hours of admission, patients were categorized into SIC ( n=36) and non-SIC ( n=30) groups. Clinical parameters, including prothrombin time (PT), activated partial thromboplastin time (APTT), sublingual PPV, and serum HMGB1 levels (measured at 0 h and 6 h), were analyzed. Logistic regression and receiver operating characteristic (ROC) curve analyses were performed to identify independent predictors. Results:Compared with the non-SIC group, the SIC group demonstrated significantly prolonged PT ( P<0.05) and APTT ( P<0.05), elevated 0 h lactate levels, and increased 6 h HMGB1. Spearman analysis revealed a positive correlation between 6 h PPV and platelet count ( P<0.05), while an inverse correlation was observed with 6 h PT ( P<0.05). Multivariate regression identified APTT, 6 h PPV, and 6 h HMGB1 as independent predictors of SIC. ROC analysis yielded AUC values of 0.732 for APTT, 0.802 for 6 h PPV, and 0.765 for 6 h HMGB1. The combination of 6 h PPV and HMGB1 further improved predictive accuracy (AUC=0.873). Conclusions:Prolonged APTT, decreased 6 h PPV, and elevated 6h HMGB1 are key indicators of SIC. Integrated assessment of these markers enhances early risk stratification in septic patients, facilitating timely clinical intervention.

Objective:To develop an artificial intelligence（AI）-based sequential ultrasound-magnetic resonance imaging（US-MRI）diagnostic strategy to optimize the imaging workflow for ovarian masses.Methods:A total of 1 120 patients with pathologically confirmed ovarian masses who underwent both preoperative pelvic ultrasound and MRI between January 2021 and December 2023 at Women's Hospital，Zhejiang University School of Medicine were retrospectively included. Patients were randomly divided into the training（ n=672）and internal test set（ n=448）at a ratio of 6∶4. An external test set（ n=128）was established at the Forth Affiliated Hospital of School of Medicine. Deep learning was used for automated segmentation of MRI lesions，followed by radiomic feature extraction and machine learning classification to construct both a US-MRI multimodal model and sequential US-MRI strategy. Diagnostic performance and potential healthcare cost-saving effects were evaluated across strategies. Results:In the internal test set（ n=448），the AI-based sequential US-MRI strategy achieved a F1 score of 0.863 and a diagnostic accuracy of 82.14%，with no significant difference compared to the US-MRI multi-modal model（ P>0.05）. The sequential strategy identified 82 cases（18.30%，82/448）of patients as low-risk true negatives during initial ultrasound screening，suggesting a potential to reduce the need for MRI examinations in future clinical practice. In the external test set（ n=128），the strategy achieved an F1 score of 0.800 and a confirmed diagnosis rate of 85.94%，with a theoretical reduction of 26.56%（34 cases）in MRI utilization while maintaining a diagnostic accuracy rate higher than that of the multi-modal model（82.18%）. Conclusions:The AI-based US-MRI sequential diagnostic strategy demonstrates favorable diagnostic accuracy while offering the potential to optimize MRI utilization. This approach may enhance the efficiency of imaging resource allocation and reduce healthcare burden in the management of ovarian masses.

Objective:To develop an artificial intelligence（AI）-based sequential ultrasound-magnetic resonance imaging（US-MRI）diagnostic strategy to optimize the imaging workflow for ovarian masses.Methods:A total of 1 120 patients with pathologically confirmed ovarian masses who underwent both preoperative pelvic ultrasound and MRI between January 2021 and December 2023 at Women's Hospital，Zhejiang University School of Medicine were retrospectively included. Patients were randomly divided into the training（ n=672）and internal test set（ n=448）at a ratio of 6∶4. An external test set（ n=128）was established at the Forth Affiliated Hospital of School of Medicine. Deep learning was used for automated segmentation of MRI lesions，followed by radiomic feature extraction and machine learning classification to construct both a US-MRI multimodal model and sequential US-MRI strategy. Diagnostic performance and potential healthcare cost-saving effects were evaluated across strategies. Results:In the internal test set（ n=448），the AI-based sequential US-MRI strategy achieved a F1 score of 0.863 and a diagnostic accuracy of 82.14%，with no significant difference compared to the US-MRI multi-modal model（ P>0.05）. The sequential strategy identified 82 cases（18.30%，82/448）of patients as low-risk true negatives during initial ultrasound screening，suggesting a potential to reduce the need for MRI examinations in future clinical practice. In the external test set（ n=128），the strategy achieved an F1 score of 0.800 and a confirmed diagnosis rate of 85.94%，with a theoretical reduction of 26.56%（34 cases）in MRI utilization while maintaining a diagnostic accuracy rate higher than that of the multi-modal model（82.18%）. Conclusions:The AI-based US-MRI sequential diagnostic strategy demonstrates favorable diagnostic accuracy while offering the potential to optimize MRI utilization. This approach may enhance the efficiency of imaging resource allocation and reduce healthcare burden in the management of ovarian masses.

Objective To investigate the correlation between serum nickel-like protein(Metrnl)level and glycolipid metabolism and inflammatory state in patients with T2DM complicated with abdominal obesity.Methods One hundred and twenty-four T2DM patients and 140 non-diabetic controls who were hospitalized in Gansu Provincial People's Hospital from May to September 2022 were selected and divided into T2DM combined abdominal obesity group(T2DM+AO,n=81),T2DM group(n=43),abdominal obesity group(AO,n=69)and normal control group(NC,n=71)according to whether they suffered from T2DM and abdominal obesity.ELISA method was used to determine the levels of serum Metrnl,IL-4 and IL-13.The correlation between serum Metrnl and the indicators of glucose and lipid metabolism,WC,BMI,VFA,FPG,HbA1c,TC,TG,LDL-C,HDL-C and the indicators of inflammatory state,IL-4 and IL-13 were analyzed.Results Compared with NC group,serum Metrnl levels in T2DM group,AO and T2DM+AO group were decreased(P＜0.05).Serum Metrnl was negatively correlated with WC,BMI,VFA,FPG,HbA1c,TG,IL-4,IL-13(P＜0.01),and positively correlated with HDL-C(P＜0.05).Age,WC,BMI and IL-4 were the influencing factors of Metrnl.WC,VFA,FPG and Metrnl were the influencing factors of T2DM combined with abdominal obesity.The ROC curve showed that serum Metrnl had a sensitivity of 82.5%and a specificity of 93.8%for the diagnosis of T2DM with abdominal obesity.Conclusions Serum Metrnl in T2DM patients with abdominal obesity is significantly reduced,and is closely related to glucose and lipid metabolism and inflammatory status.Serum Metrnl may be a novel biomarker factor for T2DM complicated with abdominal obesity.

OBJECTIVE To evaluate the comprehensive value of four first-line combination immunotherapy for unresectable hepatocellular carcinoma， and provide a reference for determining the optimal clinical treatment decision for unresectable hepatocellular carcinoma. METHODS R4.2 software was used for network meta-analysis to obtain the effect values of the efficacy and safety indicators of four combination therapies [atezolizumab combined with bevacizumab （AB）， sintilimab combined with bevacizumab biosimilars （SB）， camrelizumab combined with apatinib （CA）， durvalumab combined with tremelimumab （DT）]. Combined with the efficacy， safety and economic indicators， the categorical based evaluation technique （M-MACBETH） was used to establish the value tree. At the same time， the comprehensive value scores of four therapies were calculated， and sensitivity analysis was performed to evaluate the robustness. RESULTS In terms of prolonging median overall survival， the advantage order of the four therapies was ranked as SB， CA， AB and DT. In terms of extending median progression-free survival， the advantage order of the four therapies was CA， SB， AB and DT. In terms of safety， the order of advantages was DT， AB， SB and CA. In terms of economy， the order of advantages was CA， SB， AB and DT. The comprehensive scores of SB， CA， AB and DT were 67.11， 57.77， 52.53 and 42.59 points， respectively. The results of the sensitivity analysis showed that the ranking results of comprehensive value for four regimens were robust. CONCLUSIONS Among the four first-line immune combination therapies for unresectable hepatocellular carcinoma， SB is the optimal treatment regimen， followed by CA， AB and DT.

ObjectiveThis study aims to explore risk factors for the development of major adverse cardiovascular and cerebrovascular events （MACCEs） in middle-aged and elderly patients with type 2 diabetes mellitus complicated with stable angina pectoris （T2DM-SAP） based on real-world clinical data in traditional Chinese medicine （TCM）， so as to develop a COX proportional risk prediction model and visualize the predicted results using a nomogram. MethodBased on the clinical scientific research information sharing system， the medical records of 586 T2DM-SAP patients （45-94 years old） were collected from January 2012 to December 2019， including age， gender， course of disease， major medical history， laboratory examination， tongue image， pulse image， TCM syndrome， and major treatment drugs. MACCE outcome indicators of patients were obtained by telephone follow-up and re-hospitalization records. The data was divided into a training set and a validation set according to 7∶3. In the training set， COX univariate analysis was used to determine the risk factors for MACCE in T2DM-SAP patients， and then variables were screened by forward-backward stepwise regression method， so as to establish a MACCE risk prediction model and construct a nomogram. The predictive efficacy of the model was reflected by the C-index， receiver operating characteristic （ROC） curve， calibration map， and clinical decision curve. ResultThe history of cerebrovascular disease ［Hazard ratio （HR）=1.983， 95% confidence interval （CI，1.314-2.993）］， low-density lipoprotein （LDL-C/mmol·L^-1）≥4.1［HR=2.683， 95%CI（1.461-4.925）］， dull red tongue ［HR=1.955， 95%CI（1.273-3.002）］， dull purple tongue ［HR=4.214， 95%CI（2.017-8.803）］， white thick coating ［HR=3.030， 95%CI（1.634-9.293）］， thin and weak pulse ［HR=2.233， 95%CI（1.283-3.888）］， and syndrome of wind-phlegm blocking collaterals ［HR=2.007， 95%CI（1.179-3.418）］ were found to be risk factors in middle-aged and elderly T2DM-SAP patients. Insulin ［HR=0.604， 95%CI（0.399-0.914）］， glycosidase inhibitor ［HR=0.627， 95%CI（0.409-0.962）］， and TCM treatment ［HR=0.328， 95%CI（0.214-0.503）］ were protective factors in middle-aged and elderly T2DM-SAP patients. The prediction model was constructed based on the above risk factors. The C-index of the model was 0.818 （95% CI 0.777 -0.859） in the training set and 0.814 （95% CI 0.773-0.855） in the validation set， and the change of C-index over time was plotted. The AUC of patients for 5， 10， 15 years in the training set was 0.71， 0.67， and 0.61. The AUC of patients for 5， 10， and 15 years in the validation set was 0.60， 0.68， and 0.63， respectively. The calibration map and clinical decision curves of 5， 10， 15 years were drawn in the training set and the validation set， respectively. The model was well calibrated and clinically effective. ConclusionThe history of cerebrovascular disease， LDL， dull red tongue， dull purple tongue， white thick coating， thin and weak pulse， and syndrome of wind-phlegm blocking collaterals are risk factors for MACCE in middle-aged and elderly T2DM-SAP patients， and insulin， glycosidase inhibitors， TCM treatment are protective factors for MACCE in middle-aged and elderly T2DM-SAP patients. A clinical prediction model is established accordingly. This model has good discrimination， calibration degree， and clinical effectiveness and provides a scientific basis for the prevention and treatment of MACCE in middle-aged and elderly T2DM-SAP patients.

Objective:To investigate the distribution of allergens in patients with allergic rhinitis （AR） in Ningxia, and provide theoretical data for the prevention and treatment of AR in this region. Methods:A total of 1664 patients diagnosed with AR in the Otorhinolaryngology Head and Neck Surgery Department of Yinchuan First People's Hospital Outpatient Clinic from January 2018 to December 2021 were retrospectively collected. Use the allergen sIgE antibody detection kit （immunoblotting method） to detect inhalation and ingestion allergens in patients.Results: ①Among all AR patients, 1 158 cases were detected positive, resulting in the detection rate was 69.59%; ②The detection rate of inhalation allergen was 65.87%, and the detection rate of ingestion allergen was 19.83%; ③Mugwort was the most sensitive allergen, and 76.32% of the patients having a positive grade ≥3; ④Out of the patients, 294 cases （25.39%） were allergic to only one allergen, 244 cases （21.07%） were allergic to two allergens, and 620 cases （53.54%） were allergic to three or more allergens; ⑤During different seasons, the highest number of positive allergens detected was in the summer, with 968 cases （83.59%）. Mugwort was the main allergen during this season （69.01%）. After the COVID-19 epidemic, the total positive rate of sIgE tests in AR patients decreased compared to before, and the difference was statistically significant （P<0.001）; ⑥Mugwort, dog epithelium, mold combination, egg, peanut, soybean, Marine fish combination and fruit combination all showed statistically significant differences between different gender groups （P<0.05）; ⑦Common ragweed, mugwort, dust mite combination, cockroach, egg, milk, Marine fish combination, shrimp, fruit combination and nut combination all showed statistically significant differences among different age groups （P<0.05）; ⑧There were statistically significant differences in hay dust among different ethnic groups （P<0.05）. Conclusion:Artemisia argyi is the main allergen in Ningxia, and the distribution characteristics of different allergens are influenced by treatment season, the COVID-19 epidemic, gender, age, ethnicity, and other factors, showing certain distribution patterns and rules.
Allergens ; Artemisia ; COVID-19 ; Retrospective Studies ; Rhinitis, Allergic ; Skin Tests ; Humans ; Male ; Female

ObjectiveTo evaluate the efficacy and influencing factors of Biejiajianwan in the treatment of primary liver cancer based on real-world data of traditional Chinese medicine （TCM）. MethodClinical diagnosis and treatment data of patients with primary liver cancer admitted to five Grade-A tertiary hospitals in Henan Province from January 2015 to December 2020 were collected from the medical electronic database. The patients treated with Biejiajianwan for ≥30 days were assigned to the exposure group and those without treatment with Biejiajianwan or treated with Biejiajianwan for <30 days to the non-exposure group. The propensity score matching model was used to balance confounding factors between the two groups according to the 1∶1 genetic matching method. Kaplan-Meier method was used for survival analysis and survival curve plotting. Log-rank was used to test the difference in survival rate between the two groups. Univariate analysis of Biejiajianwan in the treatment of primary liver cancer was performed by Log-rank test combined with the Kaplan-Meier method. The factors with statistical significance （P<0.05） were combined with unbalanced factors by the propensity score matching model， and at the same time， clinical common sense and relevant prognostic factors by literature search were considered， which were subjected to multivariate analysis by Cox proportional hazards regression model. ResultA total of 2 207 electronic cases were collected，including 174 cases in the exposure group （Biejiajianwan group） and 2 033 cases in the non-exposure group. After propensity score matching， there were 174 cases in the exposure group and 174 cases in the non-exposure group. The Kaplan-Meier method was used for survival analysis on the matched data， and the Log-rank test results showed that the survival rate of patients with primary liver cancer in the Biejiajianwan group was higher than that in the control group （χ²=12.193， P<0.01）. Cox proportional hazards regression model analysis showed that the regression coefficient of Biejiajianwan was -0.916 4 with the hazard ratio （HR） ［95% confidence interval （CI）］=0.4 （0.239 5-0.668 0）， P<0.01， and the regression coefficient of radiofrequency ablation treatment was -0.976 5 with HR （95% CI）=0.376 6 （0.172 8-0.821 1， P<0.05）. Fibrinogen （FIB） abnormal regression coefficient was 0.481 4 with HR （95% CI）=1.618 4（1.022 0-2.562 9），P<0.05. ConclusionBiejiajianwan can prolong the survival period of patients with primary liver cancer. Radiofrequency ablation is an independent protective factor for Biejiajianwan in the treatment of primary liver cancer，while abnormal FIB are independent risk factors for Biejiajianwan in the treatment of primary liver cancer.

Humans ; Diabetic Neuropathies/diagnostic imaging* ; Diabetes Mellitus, Type 2 ; Ultrasonography ; Diabetic Foot/diagnostic imaging*

Objective To explore the classification and characteristics of common syndromes of primary liver cancer and provide reference for clinical treatment.Methods Collect the four diagnostic information of patients with primary liver cancer from two top three TCM hospitals in Henan Province,and build a database.Using Lantern 5.0 software,based on two-step hidden tree analysis,a hidden structure model was constructed,and common syndromes of primary liver cancer were extracted through comprehensive clustering.SPSS 23.0 software was used for factor analysis and systematic cluster analysis to infer the potential syndromes.Combined with the results of different methods and professional knowledge,the syndrome classification of primary liver cancer was determined.Results A total of 1353 patients with 105 symptoms of primary liver cancer were included.59 symptoms with an analysis frequency≥40 were included to construct a hidden structure model,24 hidden variables were obtained,and 5 common syndromes were obtained by comprehensive clustering,namely,qi deficiency syndrome,liver depression and qi stagnation syndrome,blood stasis syndrome,water dampness stagnation syndrome,liver and gallbladder damp heat syndrome.20 common factors were obtained by factor analysis for symptoms with frequency>3%,and 8 common syndromes were inferred by cluster analysis with common factors.7 common syndromes and characteristics were finally determined by combining different methods and expertise.Conclusion The common syndromes of primary liver cancer are qi deficiency syndrome,liver depression and qi stagnation syndrome,blood stasis syndrome,water dampness stagnation syndrome,yin deficiency syndrome,liver and gallbladder damp heat syndrome,spleen deficiency and dampness stagnation syndrome.The results objectively reflect the actual situation of patients with primary liver cancer,and can provide reference for the treatment of primary liver cancer based on syndrome differentiation.