Feature-Based classification and diagnosis of breast cancer using fuzzy inference system

Breast cancer is a major public health issue and currently the second leading cause of cancer death in women. Early detection on whether a tumor is benign or malignant is crucial to a patient's treatment and their odds of survival. There are already existing methods in place to detect breast cancer based on physical characteristics of tumors as well as identifying novel biomarker genes. However the accuracy still needs to be improved. The primary focus of our research is to produce a model that can predict the benign or malignant nature of a tumor with high accuracy. Since most biological systems are intrinsically fuzzy, fuzzy inference system is appropriate to classify and diagnose breast cancer based on physical characteristics of tumors. However, fuzzy inference system's performance is limited dues to the lack of in-depth understanding of the breast cancer tumor data. Therefore, in this paper, we introduce an advanced fuzzy inference system whose coefficients are optimized by statistical analysis methods. Using a normalized breast cancer dataset, we implement a Mamdani-Type fuzzy inference system to predict a benign or malignant state. The results show that the optimized fuzzy inference system can handle the complicated cancer prediction problem with higher accuracy.