Breast cancer lymph node analysis using artificial intelligence

Despite the fact that the most frequent type of cancer in the world is lung cancer the most frequent type of cancer in women is breast cancer. In breast cancer, the stage of the disease and accordingly along with the analysis of many cases concerning the method of treatment method to be used, the posture of armpit lymph node is of importance. The existence of armpit lymph case increases the risk of metastasis to other organs of the body in breast cancer. This condition, which requires surgical procedures, makes the seriousness of the disease important. In this study, clinical and pathological data, including armpit lymph case gathered from breast cancer patients in İstanbul Gaziosmanpaşa Taksim Education and Research Hospital, models were built by using k-means which is an Artificial Intelligence method, Adaptive Neuro Fuzzy Inference System, Genetic Algorithm and Artificial Neural Networks and it was analyzed for the usability of diagnosis of breast cancer lymph case. Two different hybrid models were formed using specified methods for the training of the models. A basic artificial neural network was used in order to see the efficiencies of these two methods claiming different characteristics. In an attempt to test the medical authenticity, comparison of the performance models that were created in this study was introduced using diagnostic tests and ROC Curve.