Artificial Immunity and Features Reduction for effective Breast Cancer Diagnosis and Prognosis

The diagnosis and prognosis of breast cancer is an important, realworld medical problem. As an intrusion detection problem is one of the applications of artificial immune system, in this paper proposes a novel scheme that uses a robust immune system formed from clonal selection theory and principal component analysis for breast cancer diagnosis and Prognosis. Like the job done by Antigen Presenting Cells APCs in natural immune system, this work use PCA as an aided tool for immune cells in the selection for the most important features that can detect the cancer and forward them for the immune system in training phase which generates an artificial lymphocytes ALCs and save them as immune memory. It is important to note that the training phase was done on 20% of the dataset, whereas the testing phase was done on the remaining 80% of the data set which are considered as unknown cases for the ALCs. The study proved that the best results obtained when the PCA select minimum reasonable number of features, while in the training phase the diagnostic accuracy is 0.99 and the prognostic accuracy is 0.9, and the memories ALCs achieved in the testing phase a diagnostic accuracy 0.97 and prognostic accuracy 0.88.