论文部分内容阅读
目的:建立人工神经网络用于估算西罗莫司血药浓度的方法。方法:收集56例肾移植患者口服西罗莫司的182份全血浓度数据,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,并对测试数据进行处理,验证测试结果。结果:人工神经网络平均预测误差(MPE)与平均绝对误差(MAE)分别为(0.31±1.14)、(0.89±0.77)ng·mL-1,32例/次(88.9%)血药浓度数据绝对预测误差≤2.0ng·mL-1。人工神经网络模型准确性及精密度优于多元线性回归及非线性混合效应模型。结论:人工神经网络模型可用于预测西罗莫司血药浓度,指导个体化给药。
OBJECTIVE: To establish an artificial neural network for estimating the plasma concentration of sirolimus. Methods: A total of 182 whole blood concentrations of sirolimus were collected from 56 renal transplant recipients. The genetic algorithm and the momentum method were used to optimize the network parameters. The artificial neural network was established and the test data was processed to verify the test results. Results: The average prediction error (MPE) and mean absolute error (MAE) of artificial neural network were (0.31 ± 1.14), (0.89 ± 0.77) ng · mL-1, and the absolute value of plasma concentration was 88% Prediction error ≤2.0ng · mL-1. Artificial neural network model accuracy and precision better than multivariate linear regression and nonlinear mixed effects model. Conclusion: Artificial neural network model can be used to predict the plasma concentration of sirolimus and guide the individualized administration.