Optimization of MLP Neural Network Using the FinGrain Parallel Genetic Algorithm for Breast Cancer Diagnosis

Today, the use of intelligent systems in medical diagnosis is gradually increasing. These systems lead to a reduction in error, which may be experienced by inexperienced experts. In this study, the use of artificial intelligent systems in predicting and diagnosing breast cancer, which is one of the most common cancers among women, is being considered. In this research, the diagnosis of breast cancer is performed with a two-stage approach. In the first step, the two parameters of the effective properties and the number of secret layer nodes for optimizing the MLP neural network are simultaneously optimized by a genetic algorithm. Then, using selected features and number of hidden layer nodes, a MLP neural network modeling model is developed for diagnosis of breast cancer in the second step. Here, a FinGrain parallels genetic algorithm based on optimized parameters is used to adjust the weight of the MLP neural network. The evaluation of the experiments shows that the proposed method compared to the two GAANN and CAFS methods on the WBCD dataset yielded better results and reported an accuracy of 98.72% in the average time.