فهرست مطالب sepideh omidvar

Due to the challenge of choosing the optimal treatment regimen as well as the accurate dose calculation algorithm (DCA), this study aimed to evaluate the DCAs to compare the conventional fractionation radiotherapy (CFRT) and hypofractionation radiotherapy (HFRT) of breast cancer (BC) in the prediction of cardio-pulmonary complications.
For 19 patients with left-sided BC, treatment regimens, CFRT (50Gy/25frs) vs. HFRT (42.5Gy/16frs), were simulated. Normal tissue complication probability (NTCP) and tumor control probability (TCP) values for each regimen using radiobiological models were calculated via Monte Carlo (MC) and Collapsed Cone Convolution (CCC) algorithms. For statistical comparison of the results obtained from the regimens and algorithms, the t-test and Wilcoxon test were used in SPSS Statistics. Statistical significance was defined as p<0.05.
The mean NTCP and TCP calculated in CFRT and HFRT were as follows: cardiac mortality (MC: CFRT=0.0374±0.0134 vs. HFRT=0.0173±0.0066; p<0.001) and (CCC: CFRT=0.0373±0.0134 vs. HFRT=0.0168±0.0064; p<0.001), pneumonitis (MC: CFRT=0.1201±0.0322 vs. HFRT=0.0756±0.0221; p<0.001) and (CCC: CFRT=0.1131±0.0310 vs. HFRT=0.0697±0.0120; p<0.010), and TCP (MC: CFRT=0.9979±0.0087 vs. HFRT=0.9997±0.0092; p=0.593) and (CCC: CFRT=0.9982±0.0029 vs. HFRT=0.9986±0.0016; p=0.821).
The comparison of CFRT and HFRT using MC and CCC algorithms showed that the risk of cardiac mortality and pneumonitis in CFRT was significantly higher than in HFRT, and TCP was not significantly different in the two regimens. Applications of MC-based DCAs along with suitable biological parameters can help physicists in the prediction of radiation-induced complications accurately and precisely.

The purpose of this research was to evaluate the services of the National Library and Archives of I. R. Iran on the basis of the EFQM excellence model and to predict the satisfaction of users with the services using an artificial neural network. Design/Methodology/Approach: This is an applied research project، carried out as a survey and with a quantitative approach. Questionnaires were distributed among a sample population of users; an artificial neural network was applied to evaluate the quality of the services of the National Library of Iran; and the most and least effective factors were identified through sensitivity analysis. Sensitivity analysis in the network showed that the most effective component and subcomponent were accordingly «products and services» (Mean Square Error - MSE = 0. 0613) and «up to date information» (MSE = 0. 1225). The least effective component and subcomponent was accordingly «fidelity» (MSE = 0. 0089) and «interest in introducing the library to others» (MSE = 0. 0132). From the viewpoint of library users، the «products and services» component was the most important component and «fidelity» the least effective component. Up to date information has a high priority from the users'' viewpoint.