The Segmentation of Skin Cancer Image Based on Genetic Neural Network

The segmentation of medical images is an important component of medical imaging technology, and the effect of which will impact the diagnosis and therapy directly. Taking the complexity and uncertainty of the medical images into consideration fully, we propose the genetic neural network to be used to segment the skin cancer images. Optimization of weights and thresholds in neural network based on genetic algorithm is executed to improve the convergence speed of the BP neural network. Compared with the standard BP neural network, the segmentation speed of the genetic neural network adopted in this paper is much higher. The skin cancer images segmented by this method have continuous edge and clear contour, which can be used in the quantitative analysis and identification of the skin cancer.