Breast cancer diagnosis using an enhanced Extreme Learning Machine based-Neural Network

Breast cancer has become one of the most deadly cancer among women all over the word. Fortunately, an early diagnosis of this type of cancer can considerably enhances the success of treatment. In this work, we propose a classification system of the breast cancer based on neural networks. The proposed system is a neural network with single hidden layer and trained using extreme learning machine algorithm. The main contribution of this work relies on the use of different activation functions for the hidden neurons and their optimization using genetic algorithm. To evaluate the performance of the proposed system, tests are carried out on Wisconsin Diagnostic Breast Cancer database. The obtained results show an important enhancement compared to the conventional extreme learning machine. On the other hand, the obtained results are promising compared to other state-of-the-art methods.