Abstarct: Logistic regression helps in the modeling of discrete statistical data. These models are very useful in medical sciences (such as death / life, the presence of disease / non-disease, etc.). The in diagnosis, doctors use various information sources, such as clinical history, medical examinations, laboratory tests, and so on. But placing people in both the sick and healthy groups is basically ambiguous, and obscure observations in clinical diagnosis are abundant. In such cases, logistic regression is not appropriate due to the lack of accuracy of the response variable. For fuzzy binary fuzzy response modeling, the fuzzy logistic regression model is presented in this thesis. To calculate the success of a fuzzy logistic model, the term linguistic terminology such as ..., low, medium, high, ... is defined. Then, using the principle of expansion, the transformation of the logic "odds ratio" is modeled on a set of observations of precise explanatory variables. also to estimate the coefficients of the proposed model, we use the probabilistic method and the least squares fuzzy method. We also offer two model selection methods to evaluate the model.‎