%0 Journal Article
%T Modeling Skin Cancer Using Logistic Regression
%A Shler Katorani
%A Chia Sohrabnejad
%J specialty journal of medical research and health science
%@ 2521-3172
%D 2019
%V 4
%N 2
%P 10-16
%X Background and Aim: Environmental factors affect the development of skin cancer. Therefore, studying the geographical location of the area has a great influence on the identification of the parameters affecting the disease. The aim of this study was to determine the spatial modeling of skin cancer and to obtain the coefficients of the effect of the variables studied in the occurrence of the disease, as well as to provide a prediction map of the probability of occurrence of the disease in the future. Materials and Methods: In this study, 1017 patients in the cancer registry system of Kordestan province were collected from the provincial medical services medical center from 2005 to the end of 2010. In the next step, maps of independent and dependent variables were prepared and then in the final stage, using the logistic regression modeling method, we determined the coefficients of the effect of independent variables on the disease. Results: In this research, the coefficients for temperature variable +0.51083534, the number of frost days +0.06045922, sun hours +0.00631544,height -0.00194397, humidity -0.16770188, Direction -0.03983796, jobs-0.21915089 land us-0.25833941, Geology -0/00201768, and for the gradient of -0.11615703. The results showed that the temperature factor, the most effective parameter, and after that, were effective in land use, occupation, average humidity, gradient, freezing days, direction, sunny hours, geology and height in the occurrence of skin cancer. Conclusion: The results of the prediction map provided by the logistic regression model indicate that in the future, the most likely occurrence of skin cancer and the region susceptible to the occurrence of skin cancer is Sanandaj, Saravabad, Marivan, bardarasha, Baneh, Kamyaran and saghez.
%U https://sciarena.com/article/modeling-skin-cancer-using-logistic-regression