Objective To explore the prediction efficiency of Ki-67 expression status in pituitary adenoma(PA)based on a joint model of siamese network and transfer learning.Methods The preoperative MR T1WI enhanced sequence images of 370 patients with PA diagnosed by surgery and pathology were retrospectively collected.According to the results of immunohistochemical,all patients were divided into high proliferation index group(Ki-67≥3,n=97)and low proliferation index group(Ki-67＜3,n=273),and all the sample data were randomly divided into training set and test set according to the ratio of 7∶3.Two single predictive models,ResNet-50 and VGGNet-16 and combined them with the siamese network and transfer learning were built.The prediction efficiency of different models were evaluated via accuracy rate,precision rate,recall rate,F1 score and receiver operating characteristic(ROC)curve as the main criteria.Results Compared with a single predictive model,the model combined with siamese network and transfer learning showed a better performance for predicting the Ki-67 of PA.Additionally,the ResNet-50 joint model exhibited the superior predictive performance.The accuracy rate was 0.872 7,the precision rate was 0.812 5,the recall rate was 0.764 7,the F1 score was 0.787 9,and the area under the curve(AUC)was 0.841 6.Conclusion The joint model based on siamese network and transfer learning exhibits a higher efficiency for predicting the Ki-67 expression status in PA,which can help the clinicians to formulate more personalized treatment for the patients.

Objective To explore the prediction efficiency of Ki-67 expression status in pituitary adenoma(PA)based on a joint model of siamese network and transfer learning.Methods The preoperative MR T1WI enhanced sequence images of 370 patients with PA diagnosed by surgery and pathology were retrospectively collected.According to the results of immunohistochemical,all patients were divided into high proliferation index group(Ki-67≥3,n=97)and low proliferation index group(Ki-67＜3,n=273),and all the sample data were randomly divided into training set and test set according to the ratio of 7∶3.Two single predictive models,ResNet-50 and VGGNet-16 and combined them with the siamese network and transfer learning were built.The prediction efficiency of different models were evaluated via accuracy rate,precision rate,recall rate,F1 score and receiver operating characteristic(ROC)curve as the main criteria.Results Compared with a single predictive model,the model combined with siamese network and transfer learning showed a better performance for predicting the Ki-67 of PA.Additionally,the ResNet-50 joint model exhibited the superior predictive performance.The accuracy rate was 0.872 7,the precision rate was 0.812 5,the recall rate was 0.764 7,the F1 score was 0.787 9,and the area under the curve(AUC)was 0.841 6.Conclusion The joint model based on siamese network and transfer learning exhibits a higher efficiency for predicting the Ki-67 expression status in PA,which can help the clinicians to formulate more personalized treatment for the patients.

Objective:To investigate the clinical features, pathogen composition, and prognosis of septic shock in pediatric intensive care units (PICU) in China.Methods:A multicenter retrospective cohort study. A retrospective analysis was conducted on the clinical data of children with septic shock from 10 hospitals in China between January 2018 and December 2021. The clinical features, pathogen composition, and outcomes were collected. Patients were categorized into malignant tumor and non-malignant tumor groups, as well as survival and mortality groups. T test, Mann Whitney U test or Chi square test were used respectively for comparing clinical characteristics and prognosis between 2 groups. Multiple Logistic regression was used to identify risk factors for mortality. Results:A total of 1 247 children with septic shock were included, with 748 males (59.9%) and the age of 3.1 (0.9, 8.8) years. The in-patient mortality rate was 23.2% (289 cases). The overall pathogen positive rate was 68.2% (851 cases), with 1 229 pathogens identified. Bacterial accounted for 61.4% (754 strains) and virus for 24.8% (305 strains). Among all bacterium, Gram negative bacteria constituted 64.2% (484 strains), with Pseudomonas aeruginosa and Enterobacter being the most common; Gram positive bacteria comprised 35.8% (270 strains), primarily Streptococcus and Staphylococcus species. Influenza virus (86 strains (28.2%)), Epstein-Barr virus (53 strains (17.4%)), and respiratory syncytial virus (46 strains (17.1%)) were the top three viruses. Children with malignant tumors were older and had higher pediatric risk of mortality (PRISM) Ⅲ score, paediatric sequential organ failure assessment (pSOFA) score (7.9 (4.3, 11.8) vs. 2.3 (0.8, 7.5) years old, 22 (16, 26) vs. 16 (10, 24) points, 10 (5, 14) vs. 8 (4, 12) points, Z=11.32, 0.87, 4.00, all P<0.05), and higher pathogen positive rate, and in-hospital mortality (77.7% (240/309) vs. 65.1% (611/938), 29.7% (92/309) vs. 21.0% (197/938), χ2=16.84, 10.04, both P<0.05) compared to the non-tumor group. In the death group, the score of PRISM Ⅲ, pSOFA (16 (22, 29) vs. 14 (10, 20) points, 8 (12, 15) vs. 6 (3, 9) points, Z=4.92, 11.88, both P<0.05) were all higher, and presence of neoplastic disease, positive rate of pathogen and proportion of invasive mechanical ventilation in death group were also all higher than those in survival group (29.7% (87/289) vs. 23.2% (222/958), 77.8% (225/289) vs. 65.4% (626/958), 73.7% (213/289) vs. 50.6% (485/958), χ2=5.72, 16.03, 49.98, all P<0.05). Multiple Logistic regression showed that PRISM Ⅲ, pSOFA, and malignant tumor were the independent risk factors for mortality ( OR=1.04, 1.09, 0.67, 95% CI 1.01-1.05, 1.04-1.12, 0.47-0.94, all P<0.05). Conclusions:Bacterial infection are predominant in pediatric septic shock, but viral infection are also significant. Children with malignancies are more severe and resource consumptive. The overall mortality rate for pediatric septic shock remains high, and mortality are associated with malignant tumor, PRISM Ⅲ and pSOFA scores.

BACKGROUND:Studies have found that after stimulation, chemerin can further amplify the inflammatory signal, resulting in the release of inflammatory factors and matrix metaloproteinases. It indicates that chemerin may serve as an important regulator of cartilage metabolism, and plays an important role in the pathogenesis of osteoarthritis. OBJECTIVE: To explore the correlation between level of chemerin in synovia and serum and severity of knee osteoarthritis. METHODS: Eighty patients with knee osteoarthritis were enroled in the study. Knee osteoarthritis grading was evaluated according to the Kelgren-Lawrence classification. Chemerin levels in synovia and serum were examined with enzyme-linked immunosorbent assay. RESULTS AND CONCLUSION:The level of chemerin in synovia and serum in patients with knee osteoarthritis were positively correlated with the Kelgren-Lawrence classification (r=0.981,P< 0.001;r=0.901,P < 0.001). And there was a positive correlation between the level of chemerin in synovia and in serum (r=0.596,P < 0.01). Experimental findings suggest that the levels of chemerin in synovial fluid and serum that are positively correlated with the Kelgren-Lawrence classification can be used as a reference biomarker of severity of knee osteoarthritis.

A thermal desorption ( TD) device was developed and coupled to gas chromatography ( GC) or gas chromatography-mass spectrometry ( GC-MS ) for the qualitative and quantitative analysis of semi-volatile organic compounds on atmospheric particulate matters ( PM ) . The TD was operated by direct heating and placed on the GC injector, leading to high heating rate and easy transfer of analytes to GC without focusing of analytes by cold trap. For establishing the TD-GC method, the materials used for supporting PM samples, temperature and time of thermal desorption, and types of sample injection were investigated for detection of sixteen polycyclic aromatic hydrocarbons ( PAHs) and nine n-alkanes. The limits of detection of the proposed TD-GC method were in the range of 0. 014-0. 093 ng for PAHs, and 0. 016-0. 026 ng for n-alkanes, respectively, with the correlation coefficients of correlation above 0. 9975. The TD-GC method was applied to the determination of trace PAHs and n-alkanes on PM10 samples from three cities. The recoveries were in the range of 95%-135% ( PAHs) and 95%-115% ( n-alkanes) , respectively. Finally, the TD was coupled to GC-MS for comparison of the contents of PAHs and n-alkanes on PMx with different particulate size ( x=10 , 5, 2, 1, 0. 5, 0. 25, 0. 1).