A methodology of computer aided diagnostic system on breast cancer

In this paper, a new approach using ANFIS (adaptive neuro-fuzzy inference system) as a diagnosis system on Wisconsin breast cancer diagnosis (WBCD) problem is proposed. The automatic diagnosis of breast cancer is an important, real-world medical problem. It is occasionally difficult to attain the ultimate diagnosis even for medical experts due to the complexity and non-linearity of the relationships between the large measured factors. It is possibly resolved with a human like decision-making process using artificial intelligence (AI) algorithms. ANFIS is an AI algorithm which has the advantages of both fuzzy inference system and neural networks. Therefore, it can deal with ambiguous data and learn from the past data by itself. Considering these features, applying ANFIS as a diagnostic system was considered in our experiment. In addition, in real implementations, the performance of diagnosis system in computation is an important issue as well as the correctness of the output from the inference system. A couple of methods using recommended inputs generated by genetic algorithm, decision tree and correlation coefficient computation with ANFIS are proposed to reduce the computational overhead and they possibly enhance the performance by eliminating less-relevant input features