基于遗传算法和BP神经网络的CTG识别研究

CTG recognition based on genetic algorithm and BP neural network

目的:针对正常类、非典型类和异常类3类胎心宫缩监护图(Cardiotocography,CTG),利用BP神经网络建立分类识别模型。方法:根据遗传算法(Genetic algorithm,GA)具有全局寻优的特点,引入遗传算法对神经网络权值和阈值进行优化,形成一种训练神经网络的混合算法(GA-BP算法),以克服BP算法易陷入局部最优解的缺陷,并以UCI数据库中的CTG数据集为例进行测试。结论:仿真结果表明,对于正常类、非典型类和异常类3类样本,BP算法的分类准确率分别为97.32%、71.97%和95.45%,而GA-BP算法的分类准确率分别提高到98.24%、82.67%和95.65%。可见GA-BP分类模型具有较强的学习能力和泛化能力,分类准确度更令人满意,所提出的方法是可行的。

Objective：To establish BP neural network（NN）classification model for normal,atypical and abnormal cardiotocography（CTG）.Methods：Genetic algorithm（GA）was used to optimize the initial weights and threshold value of the neural network based on the inherent local minimum problem of NN and the good global convergence and global search ability of GA.The UCI＇s CTG set was used to test the algorithm.Results：The simulative results indicated that the recognition rates of the normal,atypical and abnormal CTG reached 98.24%,82.67%,and 95.65% based on GA-BP,more than 97.32%,71.97%,and 95.45% based on BPNN.The model has good ability both in learning and generalization and the algorithm can also effectively be used in other signal processing fields.