基于差分进化算法优化多因素灰色模型的异丙酚血药浓度预测

发布时间：2018-05-18 20:59

  本文选题：差分进化算法 + 多因素灰色模型　； 参考：《药学学报》2017年10期

【摘要】：针对短效静脉麻醉药物异丙酚在药物代谢过程中存在强时变性、复杂非线性等特点,以及传统的群体药代非线性混合效应法在建模方法中存在工作繁杂、人为因素多等缺陷,本研究利用差分进化算法优化多因素时序灰色模型,建立基于灰色理论的异丙酚药代血药浓度预测模型,并与非线性混合效应建模法(nonlinear mixed effects modeling,NONMEM)预测效果进行比较。结果显示,差分进化-多变量灰色模型(DE-MGM)的预测结果的偏离性(MDPE)为-4.6%,NONMEM为-12.13%;DE-MGM的预测结果的精确度(MDAPE)为13.19%,NONMEM为23.12%。基于差分进化优化多因素灰色模型能稳定预测异丙酚血药浓度,且准确度高。该方法原理简单,实现便捷,可适用于异丙酚等短效静脉麻醉药物的群体药代药效学研究和分析。
[Abstract]:In view of the characteristics of propofol in the course of drug metabolism, such as strong time-varying and complex nonlinearity, the traditional method of group pharmacokinetic nonlinear mixing effect has many defects, such as complicated work and many human factors in the modeling method of short-term intravenous anesthetic drug propofol. In this study, the multi-factor time-series grey model was optimized by using differential evolution algorithm, and the prediction model of propofol drug concentration was established based on grey theory, and compared with nonlinear mixed effects modeling method of NONMEM (nonlinear mixed effects modeling method). The results show that the deviant MDPEs of the DE-MGMGM prediction results are -4.6NMEM -12.13DE-MGM, the accuracy of the prediction results is 13.19NONMEM is 23.120.The results show that the deviation of MDPEs is -4.6NMEM and -12.13DE-MGM, and the accuracy of the prediction results is 13.19NMEM (23.12NMEM). Multi-factor grey model based on differential evolution can predict propofol concentration stably and has high accuracy. The method is simple in principle and convenient in realization. It can be applied to the study and analysis of population pharmacodynamics of propofol and other short-acting intravenous anesthetic drugs.
【作者单位】： 中南大学湘雅医院麻醉科;中南大学湘雅医院医学工程中心;中南大学数学与统计学院;
【基金】：国家自然科学青年基金资助项目(81601728)
【分类号】：R969
，

本文编号：1907186