论文部分内容阅读
阿尔茨海默病(AD)和轻度认知功能损伤(MCI)具有患者多、诊断难的特点,改进BP神经网络,提出自适应BP神经网络(ABP)进行100次AD和MCI诊断模拟,ABP神经网络的诊断正确率显著高于BP和RBF神经网络.采用留一法将101例正常人、200例MCI和90例AD患者的样本分为训练集和检测集,用ABP神经网络对其进行诊断模拟,总正确率达到73.91%.
Alzheimer’s disease (AD) and mild cognitive impairment (MCI) have many characteristics of patients with difficult diagnosis, improve BP neural network, put forward adaptive BP neural network (ABP) 100 AD and MCI diagnostic simulation, The diagnostic accuracy of ABP neural network was significantly higher than that of BP neural network and RBF neural network.The samples of 101 normal subjects, 200 MCI patients and 90 AD patients were divided into training set and test set using the one-leave method, and ABP neural network Diagnostic simulation, the total correct rate of 73.91%.