论文部分内容阅读
目的利用人工神经网络技术预测癫痫患儿服用丙戊酸后体内药物浓度。方法收集200例癫痫患儿服用丙戊酸后血药浓度监测结果、身高、体重及监测当日肝肾功能等15项相关指标,根据神经网络和遗传优化反向传播算法的基本原理,构建丙戊酸血药浓度预测模型,并用该浓度预测模型进行样本预测分析。结果 50个病例样本的预测结果表明,与实际测定浓度相比,误差小于10%的有29个浓度,误差在10%～15%的有10个浓度,误差在15%～20%的有7个浓度,误差大于20%的有4个浓度。误差小于15%的比率是78%。人工神经网络预测的血药浓度和实际测定浓度之间的相关系数为0.9476。结论用人工神经网络技术预测癫痫患儿服用丙戊酸后的血药浓度是可行的;有待将其广泛应用于个体化给药设计。
Objective To predict the drug concentration of valproic acid in children with epilepsy by artificial neural network. Methods Fifty children with epilepsy after taking valproate were enrolled in this study. Fifteen related indicators, including height, weight and liver and kidney function, were collected. Based on the basic principles of neural network and genetic algorithm to optimize backpropagation, Acid concentration prediction model, and use the concentration prediction model for sample prediction analysis. Results The predictive results of 50 case samples showed that there were 29 concentrations with error less than 10%, 10 with error within 10% ~ 15% and 7% with error within 15% ~ 20% compared with the actual measured concentration A concentration, the error is greater than 20% of the 4 concentrations. The error is less than 15% of the ratio is 78%. The correlation coefficient between the predicted blood concentration of artificial neural network and the actual measured concentration was 0.9476. Conclusion It is feasible to predict the plasma concentration of valproic acid in children with epilepsy by artificial neural network technology. It is to be widely applied in the design of individualized administration.