Incorporation of expert variability into breast cancer treatment recommendation in designing clinical protocol guided fuzzy rule system models

► We proposed a novel breast cancer treatment decision model. ► We used non-stationary fuzzy sets to model variability in the decisions obtained. ► Twelve fuzzy rules are generated from clinical protocol to characterize the clinician's knowledge. ► The membership functions are designed to adhere to clinical guidelines as accurately as possible. ► The proposed fuzzy model can significantly increase model performance to around 88.1%, p < 0.001.