用径向基神经网络预测氯丙嗪的稳态血药浓度  

Predicting steady-state plasma concentration of chlorpromazine using radial basis function neural networks

作　　者：刘朝晖[1,2] 黄榕波[3] 陈庆强[4] 温预关[5] 李明亚[2] 

机构地区：[1]中山市中医院药剂科,广东中山528400 [2]广东药学院药科学院,广州510006 [3]广东药学院基础学院,广州510006 [4]广州市脑科医院药剂科,广州510370 [5]广州市脑科医院国家药品临床研究基地,广州510370

出　　处：《中国临床药理学杂志》2012年第7期536-538,共3页The Chinese Journal of Clinical Pharmacology

基　　金：国家自然科学基金资助项目(10926191);中山市科技计划基金资助项目(20102A024)

摘　　要：目的评价用径向基(RBF)神经网络所建立的预测氯丙嗪稳态血药浓度模型的预测性能。方法将数据分为训练集、校验集和测试集,来建立获取输出变量(37项参数)与输出变量(氯丙嗪稳态血药浓度)两者间关系的RBF网络模型,并评价其预测性能。结果当扩展速度(SP)值为2.8时,所建立的RBF网络模型,预测奋乃静稳态血药浓度的效果和泛化能力较好。结论 RBF网络用于预测氯丙嗪稳态血药浓度是可行的和有效的。Objective To evaluate the performance of a model for predicting the steady-state plasma concentration of chlorpromazine established by using radial basis function（RBF） neural network.Methods The data was divided into training set,validation set and test set to establish the RBF neural network model which had captured the relationships between the input variables（37 parametes） and the output variable（steady-state plasma concentration of chlorpromazine） and evaluate predictive performance of the model.Results When the SPREAD（SP） value was 2.8,the RBF neural network model had the better effect on predicting the steady-state plasma concentration of chlorpromazine and better generalization.Conclusion It is practical and valid for RBF neural network model to be applied to the study of steady-state plasma concentration prediction of chlorpromazine.

关 键 词：径向基神经网络 氯丙嗪 稳态血药浓度 

分 类 号：R969.1[医药卫生—药理学] R971.41[医药卫生—药学]