Breast Cancer Recognition Based on BP Neural Network Optimized by Improved Genetic Algorithm

In recent years, the incidence of breast cancer is increasing and becomes one of the main causes of female death. The BP neural network optimized by standard genetic algorithm has slow convergence speed and is prone to local optimization, which makes the diagnosis accuracy of breast cancer decrease. This paper uses the improved genetic algorithm to optimize BP neural network by improving the selection operator of the standard genetic algorithm. The population diversity was first increased, and the probability of crossover and mutation was adaptively adjusted. Then deep optimization was executed on the initial weight threshold of BP network to speed up the network’s convergence, and the number of iterations was reduced. Finally breast cancer diagnose was performed. The experiment results show that both the fitness of the improved genetic algorithm and the recognition accuracy of breast cancer are improved. The shortcomings of the standard genetic algorithm optimized BP neural network algorithm in breast cancer diagnosis are well solved.