老年性白内障晶状体中金属元素的人工神经网分析

Classification of Human Senile Cataract Lenses Based on Metal Contents Using Neural Networks

由Ｋ、Ｎａ、Ｃａ、Ｍｇ、Ｆｅ、Ｃｕ、Ｚｎ和Ｍｎ在老年性晶状体中的含量，运用人工神经网法成功地将老年性白内障、白内障晶状体核和正常人晶状体划分为３类。同时讨论了神经网的结构（层数及每层的结点数）、初始权重等对神经网性能的影响。随机地将３０个晶状体分为训练集和测试集，其识别率及预测率均达到１００％。

Senile cataract lenses.nuclei from cataract lenses,and normal lenses were successfully separated into three classes using quasi-Newton neural networks. Tlie lenses were classified based on the concentrations of K,Na,Ca,Mg,Cu,Fe,Zn and Mn measured by atomic absorption spectroscopy.The 30 cataract lenses used in this study were randomly divided into a 23 member training set and 7 member test set.The architecture including the number of layers,the number of neurons in each layer,and the initial weights were varied in order to study their effects on the performance of the neural network.Once trained,the neural network correctly classified all 30 cataract lenses.