针对具体的乳腺癌诊断分类问题 ,研究了多层径向基函数 (RBP)网络的分类机理和初始化优化参数 ,采用动量法和学习率自适应调整两种策略方法 ,建立了基于BP神经网络和径向基 (RBF)神经网络的乳腺癌两种诊断模型。讨论了径向基神经网络的分类机理 ,同时对数据作了预处理。径向基(RBF)神经网络具有较强的非线性并行处理能力和容错能力。仿真结果表明 ,所设计的RBF网络模型性能稳定 ,训练时间短 ,分类效果较好 According to the diagnosis and classification of breast cancer, the classification mechanism and initialization optimization parameters of multi-layer Radial Basis Function (RBP) network are studied. Two methods are adopted, which are the momentum method and the adaptive adjustment of the learning rate. Based on BP neural network and Radial Basis Function (RBF) Neural Networks for Two Types of Breast Cancer Diagnosis. The classification mechanism of RBF neural network is discussed, and the data is preprocessed. Radial Basis (RBF) neural network has strong non-linear parallel processing ability and fault-tolerant ability. Simulation results show that the designed RBF network model has stable performance, short training time and good classification