计算机辅助人工神经网络参数优化软件及在缓释制剂处方优化中的应用.pdfVIP

药学学报 Acta Pharmaceutica Sinica 2009, 44 ( 10) : 1159 1164 ·1159 · 计算机辅助人工神经网络参数优化软件及其 在缓释制剂处方优化中的应用 1, 2 2* 2 3* 张星一 , 陈大为 , 金 杰 , 卢 炜 ( 1. 国家食品 品监督管理局 品审评中心, 北京 100038; 2. 沈阳药科大学药学院, 辽宁 沈阳 110016; 3. 北京大学 学院, 北京 100083) 摘要: 人工神经网络是一种多目标同时优化方法, 它需要使用者具备统计学基础, 并熟练掌握其原理以进 行正确的参数设置。为简化其复杂的设置和调试过程, 本文采用 VB 语言编制了人工神经网络参数优化软件 (ANNPOS)。以盐酸川芎嗪为模型 物, 在其缓释制剂的处方设计过程中, 采用ANNPOS 预先进行节点数、动量 项、速率等网络参数的优化, 再使用Matlab 5.0 神经网络工具箱进行多目标同时优化。实践表明, ANNPOS 可以 提高神经网络的优化效率。 关键词: 人工神经网络; 处方优化; 缓释制剂 中图分类号: R943 文献标识码:A 文章编号: 05 13-4870 (2009) 10-1159-06 Artificial neural network parameters optimization software and its application in the design of sustained release tablets 1, 2 2* 2 3* ZHANG Xing-yi , CHEN Da-wei , JIN Jie , LU Wei (1. Center for Drug Evaluation, SFDA, Beij ing 100038, China; 2. College of Pharmaceutics, Sheny ang Pharmaceutical University, Sheny ang 110016, China; 3. School of Pharmaceutical Science, Peking University, Beij ing 100083, China) Abstract: Artificial neural network (ANN) is a multi-obj ective optimization method that needs mathematic and statistic knowledge which restricts its application in the pharmaceutical research area. An artificial neural network parameters optimization software (ANNPOS) programmed by the Visual Basic language was developed to overcome this shortcoming. In the design of a sustained release formulation, the suitable parameters of ANN were estimated by the ANNPOS. And t