Clinical databases are widely utilized for predicting different diseases. The medical analysis is primarily based on data received from various sources along with results of medical examination, patient previous records and other different information that physician consider in order to accomplishing a final diagnostic decision. Numerous studies have conveyed that a large number of people with breast cancer are undiagnosed due to uncertainty of medical data that contains redundant, incomplete, obscure and unpredictable. The intention of this study is to propose a hybrid metaheuristic approach which includes neural network, fuzzy logic techniques and genetic algorithm to diagnose breast cancer and its types. Neural network and fuzzy logic techniques utilized to identifying breast cancer and categorize the types. Genetic algorithm applied to compute the best fitness value of evaluating the diagnosing accuracy. The overall findings from this study scored very high which is 96.43 % accuracy. The proposed hybrid metaheuristic approach can serve as a supportive tool to assist medical experts and to train medical students and nurse for diagnose breast cancer disease.