论文部分内容阅读
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 predictive performance of a predictive model of steady-state plasma concentration of chlorpromazine established by a radial basis function (RBF) neural network. Methods The data were divided into training set, test set and test set to establish a RBF network model for obtaining the relationship between output variables (37 parameters) and output variables (steady-state plasma concentration of chlorpromazine) Predict performance. Results When the expansion velocity (SP) value was 2.8, the RBF network model was established to predict the steady-state plasma concentration of perphenazine with good effect and generalization ability. Conclusions The RBF network is feasible and effective in predicting steady-state plasma concentration of chlorpromazine.