Objective: To investigate the expression levels and clinical significance of serum suppressor of cytokine signaling-3 (SOCS3) and associated cytokines in HIV/TB co-infected patients. Methods: The serum levels of SOCS3, IFN-γ, IL-2, IL-4, IL-6, IL-10, IL-17 and IL-22 were quantified by using enzyme-linked immunosorbent assays (ELISA) in 50 HIV-infected patients, 48 HIV/TB co-infected patients and 50 healthy donors. Pearson correlation analysis was used to analyze the correlation between SOCS3 and other seven cytokines. Results: Serum levels of SOCS3 expression in HIV/TB co-infection group were significantly higher than those in HIV-infection alone and the control group. There was also significant correlation between SOCS3 and IFN-γ, IL-4, IL-6, IL-10, IL-2 in HIV/TB co-infection group. Conclusions These findings indicated that SOCS3 may play an important role in the immune response of patients with HIV/TB co-infection and it may be helpful in the diagnosis of HIV/TB co-infection.

Objective: To analyze the causes of preoperative miscarriage of pancreatic serous cystadenoma (SCN) and find the ways to improve it. Methods: Clinical data of 425 pancreatic cystic neoplasm patients who underwent surgical resection from January 2006 to December 2016 in Department of Pancreatic Surgery in Huashan Hospital were retrospectively analyzed.Excel database was created which covered 128 fields of 7 fields: general information of patients, preoperative blood biochemical indexes, tumor markers, surgical related data, postoperative complications, imaging findings and pathology.One hundred and sixty-one cases of SCN were analyzed in depth, mainly in three aspects: surgical benefit, preoperative imaging diagnostic value and interference factors in preoperative judgement.The classification data were analyzed by χ² test and the quantitative data were analyzed by t test.The Logistic regression model was used for multiple factor analysis. Results: Of the 425 PCN cases surgically removed, 161 cases (37.9%) were SCN, the incidence of operative complications was 40.4%(65/161), the hospitalization days was (20.7±12.1)days and the medical cost was (75 267±37 866) yuan.Only 3 of 161 cases of SCN were accurately diagnosed by preoperative imaging methods, 61 cases were diagnosed as "cystic lesions of pancreas" (37.9%) and 52 cases were diagnosed as "pancreatic cystadenoma" (32.3%). SCN was misdiagnosed as MCN(32.3%) and IPMN(28%) before operation.25.5% of them were diagnosed as SCN before operation, but still underwent radical operation.The rate of preoperative imaging diagnosis for identifying SCN was 62.8%.The lack of preoperative endoscopy and the lack of understanding of the image characteristics and biological behavior of SCN were the most important factors affecting the accuracy of preoperative judgment.Statistics found that gender, age, CA125 and tumor location can be used as independent factors contribute to the clinical identification(χ²=8.995, P=0.003; χ²=10.019, P=0.007; t=3.157, P=0.002; χ²=6.790, P=0.009). Logistic analysis showed that women, older than 60 years old, the tumors located in the pancreatic body and tail were the independent factors of SCN classification and diagnosis (OR=0.481, 0.376, 0.577, 0.666, 95% CI: 0.305-0.759, 0.199-0.710, 0.361-0.924, 0.433-1.024, P=0.002, 0.003, 0.022, 0.064). Conclusions SCN has more benign biological behavior.Although surgical excision is acceptable for clinical safety, the corresponding benefit is very limited.It is possible to improve the rationality of SCN clinical operation decisions to some extent by performing endoscopic examination, imaging doctors to improve the SCN feature recognition and surgeons to enhance the awareness of SCN.

This study aims to predict expression of Ki67 molecular marker in pancreatic cystic neoplasm using radiomics. We firstly manually segmented tumor area in multi-detector computed tomography (MDCT) images. Then 409 high-throughput features were automatically extracted and the least absolute shrinkage selection operator (LASSO) regression model was used for feature selection. After 200 bootstrapping repetitions of LASSO, 20 most frequently selected features made up the optimal feature set. Then 200 bootstrapping repetitions of support vector machine (SVM) classifier with 10-fold cross-validation were used to avoid overfitting and accurately predict the Ki67 expression. The highest prediction accuracy could achieve 85.29% and the highest area under the receiver operating characteristic curve (AUC) was 91.54% with a sensitivity (SENS) of 81.88% and a specificity (SPEC) of 86.75%. According to the results of experiment, the feasibility of predicting expression of Ki67 in pancreatic cystic neoplasm based on radiomics was verified.