论文部分内容阅读
建立人工神经网络用于估算他克莫司血药浓度。收集37例肝移植受者口服他克莫司的176份稳态全血浓度数据,采用遗传算法配合动量法优化网络参数,建立人工神经网络。人工神经网络平均预测误差(MPE)与平均绝对误差(MAE)分别为(0.02±2.40)ng.mL 1和(1.93±1.37)ng.mL 1,84.6%血药浓度数据绝对预测误差≤3.0 ng.mL 1。人工神经网络的准确性及精密度优于多元线性回归。结果表明,人工神经网络预测的相关性、准确性和精密度较好,简便迅捷,可用于预测他克莫司血药浓度。
Establishment of artificial neural network for estimating tacrolimus plasma concentration. A total of 176 steady-state whole blood concentration data from 37 liver transplant recipients were collected. The genetic algorithm and the momentum method were used to optimize the network parameters to establish an artificial neural network. The average prediction error (MPE) and mean absolute error (MAE) of artificial neural network were (0.02 ± 2.40) ng.mL 1 and (1.93 ± 1.37) ng.mL respectively. The absolute prediction error of blood concentration data was less than or equal to 3.0 ng .mL 1. Artificial neural network accuracy and precision better than multiple linear regression. The results show that the ANN prediction has good correlation, accuracy and precision, is simple and quick and can be used to predict the plasma concentration of tacrolimus.