Improving female breast cancer prognosis by means of fuzzy rule induction with artificial immune systems

Breast cancer is the second most common cause of deaths from cancer among women in the United States. Even if significant steps have been made in the field of cancer treatment there’s still room for investigation when it comes to the modeling of metastatic behavior of tumors. In particular over-treatment avoidance of patient is currently a challenging area of research due to the positive effects it can have patients’ quality of life and clinical costs management. In this paper we propose a novel approach to gene signature finding aimed at improving prediction accuracy of the tumor recurrence. Our approach lays on a novel computational paradigm, namely Artificial Immune Systems (AIS). Based on AIS, our algorithm, IFRAIS (Induction of Fuzzy Rules with Artificial Immune Systems) mines the high density array data in order to extract useful knowledge, in the form of “IF-THEN” rules, easily interpretable by physicians and able to improve prediction accuracy for tumor recurrence.