Objective:To construct classification prediction models for gamma passing rate using radiomics-based machine learning approaches and data from multiple radiotherapy institutions and evaluate the models′ performance.Methods:The data from 572 volumetric-modulated arc therapy (VMAT) patients across three radiotherapy institutions (514 for training and 58 for testing)were retrospectively collected. Additionally, 45 VMAT plans were collected from a single institution as an independent external validation set. For all the data, a three-dimensional dose validation approach based on actual measurements of phantoms was utilized, and gamma analysis was performed at the 3%/2 mm criterion using a dose threshold of 10%, absolute doses, and global normalization. After radiomic features were extracted from dose files, feature selection was performed using the random forest (RF) method and RF combined with Shapley Additive exPlanation (SHAP). Then, feature subsets of varying sizes (10, 20, 30, 40, and 50) were selected based on feature rankings. Using these subsets as inputs, data training was conducted using the Extreme Gradient Boosting (XGBoost) algorithm. Finally, the models′ classification performance was assessed using the area under the curve (AUC) values and F1-score.Results:Under the 3%/2 mm criterion, all models performed the best in the case of 20 feature subsets. The optimal prediction model established based on the feature selection using RF exhibited AUC and F1-score of 0.88 and 0.89, respectively on the testing set and 0.82 and 0.90, respectively, on the validation set. The optimal prediction model built based on the feature selection using RF combined with SHAP yielded AUC and F1-score of 0.86 and 0.92 on the testing set and 0.87 and 0.89, respectively, on the validation set, along with superior robustness. Therefore, the second model possessed certain advantages over the first model.Conclusions:For multicenter dose verification result, it is feasible to construct a machine learning prediction model with high classification performance using radiomic features derived from dose files, combined with feature selection based on SHAP. This approach can assist in advancing the clinical applications and implementation of gamma passing rate prediction models.

Objective:To construct classification prediction models for gamma passing rate using radiomics-based machine learning approaches and data from multiple radiotherapy institutions and evaluate the models′ performance.Methods:The data from 572 volumetric-modulated arc therapy (VMAT) patients across three radiotherapy institutions (514 for training and 58 for testing)were retrospectively collected. Additionally, 45 VMAT plans were collected from a single institution as an independent external validation set. For all the data, a three-dimensional dose validation approach based on actual measurements of phantoms was utilized, and gamma analysis was performed at the 3%/2 mm criterion using a dose threshold of 10%, absolute doses, and global normalization. After radiomic features were extracted from dose files, feature selection was performed using the random forest (RF) method and RF combined with Shapley Additive exPlanation (SHAP). Then, feature subsets of varying sizes (10, 20, 30, 40, and 50) were selected based on feature rankings. Using these subsets as inputs, data training was conducted using the Extreme Gradient Boosting (XGBoost) algorithm. Finally, the models′ classification performance was assessed using the area under the curve (AUC) values and F1-score.Results:Under the 3%/2 mm criterion, all models performed the best in the case of 20 feature subsets. The optimal prediction model established based on the feature selection using RF exhibited AUC and F1-score of 0.88 and 0.89, respectively on the testing set and 0.82 and 0.90, respectively, on the validation set. The optimal prediction model built based on the feature selection using RF combined with SHAP yielded AUC and F1-score of 0.86 and 0.92 on the testing set and 0.87 and 0.89, respectively, on the validation set, along with superior robustness. Therefore, the second model possessed certain advantages over the first model.Conclusions:For multicenter dose verification result, it is feasible to construct a machine learning prediction model with high classification performance using radiomic features derived from dose files, combined with feature selection based on SHAP. This approach can assist in advancing the clinical applications and implementation of gamma passing rate prediction models.

Objective:To explore the feasibility of a classification prediction model for gamma pass rates (GPRs) under different intensity-modulated radiation therapy techniques for pelvic tumors using a radiomics-based machine learning approach, and compare the classification performance of four integrated tree models.Methods:With a retrospective collection of 409 plans using different IMRT techniques, the three-dimensional dose validation results were adopted based on modality measurements, with a GPR criterion of 3%/2 mm and 10% dose threshold. Then prediction were built models by extracting radiomics features based on dose documentation. Four machine learning algorithms were used, namely random forest (RF), adaptive boosting (AdaBoost), extreme gradient boosting (XGBoost), and light gradient boosting machine (LightGBM). Their classification performance was evaluated by calculating sensitivity, specificity, F1 score, and AUC value. Results:The RF, AdaBoost, XGBoost, and LightGBM models had sensitivities of 0.96, 0.82, 0.93, and 0.89, specificities of 0.38, 0.54, 0.62, and 0.62, F1 scores of 0.86, 0.81, 0.88, and 0.86, and AUC values of 0.81, 0.77, 0.85, and 0.83, respectively. XGBoost model showed the highest sensitivity, specificity, F1 score, and AUC value, outperforming the other three models. Conclusions:To build a GPR classification prediction model using a radiomics-based machine learning approach is feasible for plans using different intensity-modulated radiotherapy techniques for pelvic tumors, providing a basis for future multi-institutional collaborative research on GPR prediction.

Objective To investigate the influence of WeChat + family centered health intervention on the self-efficacy of elderly hypertensive patients in General clinic. Methods A total of 102 elderly hypertensive patients aged≥60 years in General clinic from February 2017 to July 2017 were enrolled.They were divided into experimental group and control group with the table of random number, with 51 cases in each group. The experimental group used the WeChat+ family centered health intervention model,while the control group adopted the traditional health education model.The changes of blood pressure and self-efficacy were evaluated before intervention, 3 months and 6 months after intervention respectively. Results There were no statistic significant difference in blood pressure and self-efficacy total score between the two groups before intervention(P>0.05). At 3 months after intervention,the blood pressure(systolic pressure/diastolic pressure)of the patients were(144.20±4.60), (80.00±5.00)mmHg(1 mmHg=0.133 kPa)in the experimental group,those of which were(154.20±7.16), (87.00±3.81)mmHg in the control group,the differences were statistically significant (t=-2.628,-2.490,P<0.05).At 6 months after intervention,the blood pressure(systolic pressure/diastolic pressure)of the patients were(141.60±6.43),(76.00±4.85)mmHg in the experimental group,those of which were(151.60± 5.94),(85.40±4.56)mmHg in the control group,the differences were statistically significant (t=-2.555,-3.158,P<0.05).There were differences in systolic and diastolic pressure in patients with different groups at different time,and the differences were statistically significant(F=18.668,18.174,P<0.01).The total score of self-efficacy at 3 months and 6 months after intervention was(30.14±0.43),(32.56±0.23)points in the experimental group and (28.14 ± 0.15), (29.40 ± 0.19) points in the control group, the difference was statistically significant(t=9.736, 23.819, P<0.05). The differences in the total score of patients′self-efficacy in different groups at different time had statistically significance (F=2 631.551, P<0.01). Conclusions The WeChat+family centered health intervention model can significantly improve the self-efficacy of the elderly patients with high blood pressure in the General clinic.

BACKGROUNDThe pathogenesis of gastric cancer (GC) involves environmental and genetic factors. Recently, two genome-wide association studies found that phospholipase C epsilon 1 (PLCE1) polymorphisms might be related to GC risk, and several studies further validated this finding. However, these studies yielded inconsistent results.

METHODSA comprehensive database search was performed to identify eligible studies. Odds ratios with 95% confidence intervals were calculated to assess the strength of the association between PLCE1 rs2274223, rs753724, and rs11187842 and risk of GC. Subgroup analyses, publication bias, and sensitivity analyses were also conducted.

RESULTSEleven studies (12 cohorts) were included in the meta-analysis. Based on 13 676 cases and 23 569 controls, a significant association between PLCE1 rs2274223 and GC risk was detected under various genotypic models. In the subgroup analyses, the association was significant for cardia GC, but weak for non-cardia GC. The association under the heterozygote model was detected for PLCE1 rs753724 and rs11187842 based on three studies involving 2768 cases and 3890 controls.

CONCLUSIONSOur findings demonstrate that the presence of the G allele at rs2274223 of the PLCE1 gene may contribute to susceptibility to GC, especially cardia GC. PLCE1 rs753724 and rs11187842 are associated with GC risk under the heterozygote model. Further well-designed large studies are warranted to validate these findings.

Genetic Predisposition to Disease ; genetics ; Genome-Wide Association Study ; Humans ; Phosphoinositide Phospholipase C ; genetics ; Polymorphism, Genetic ; genetics ; Stomach Neoplasms ; genetics

0.05].Sao and Eao was significantly different between CHD group and the control group,but Aao has not significant different.③Sao positively correlated with ascending aortic distensibility coefficient(D)(r=0.73,P=0.03),and negatively correlated with aortic stiffness(?)(r=-0.68,P=0.03).Conclusion:Elastic properties of the aorta can directly be assessed by measuring the movements in the upper wall of the aorta with DTI.Reduced aortic S-velocity is significantly correlated with Ascending aortic distensibility coefficient(D) and stiffness index beta(?),which are important factors in assessing the changes of the aortic distensibility.

Objective To investigate the effect of pronuciferine on apoptosis of cultured human umbilical vein endothelium cells(HUVECs) induced by angiotensin Ⅱ(AngⅡ).Methods HUVECs cell line ECV304 was cultured in vitro,pretreated with Captopril(10 ?mol/L) or pronuciferine 10,1,0.1,0.01 ?mol/L for 30 min,respectively,then treated with AngⅡ(1 ?mol/L).Cell-morphosis was observed by light microscope.Cells viability was assessed by MTT assay.Production of nitric oxide(NO),activities of total nitric oxide synthase(tNOS),and inducible nitric oxide synthase(iNOS) were measured by colorimetry.Apoptosis rate was measured by Flow Cytometer(FCM).Results AngⅡ induced typical endothelial cell apoptosis and the apoptosis rates were significantly higher than those of the control group((P