论文部分内容阅读
目的采用人工神经网络对脑脉通有效部位进行组分配伍优化。方法采用均匀设计法将脑脉通中的五大有效部位10个不同水平的配比组合进行药效考察,除假手术组外其他各组制备大鼠大脑中动脉闭塞(MCAO)模型,以脑梗死面积和坏死区体密度作为药效指标,将获得的实验结果作为训练集,建立人工神经网络药效预测模型。结果基于建立的人工神经网络模型进行预测可知脑脉通组分最佳配伍为:大黄总蒽醌180 mg·kg~(-1),人参总皂苷70 mg·kg~(-1),葛根总黄酮450 mg·kg~(-1),川芎总酚酸27 mg·kg~(-1),川芎油110μL·kg~(-1)。并经实验验证优化的脑脉通组分配伍对各药效指标改善作用显著。结论脑脉通各组分配比均能不同程度的改善缺血性脑中风的症状,基于均匀设计结合人工神经网络优化组分配伍的方法是可行的。
Objective To optimize the compatibility of components in the effective part of Naomaitong by using artificial neural network. Methods A uniform design method was used to study the effect of 10 different levels of the five effective parts in the brain-vein communication. The rats in each group except the sham-operation group were given MCAO model and cerebral infarction Area and necrotic area density as the pharmacodynamic parameters, the experimental results obtained as a training set, the establishment of artificial neural network pharmacodynamic prediction model. Results Based on the established artificial neural network model, the best compatibility of Naomaitong was identified as total anthraquinone 180 mg · kg -1, ginseng total saponin 70 mg · kg -1, Flavonoids 450 mg · kg -1, Ligusticum chuanxiong 27 mg · kg -1 and Chuanxiong oil 110 μL · kg -1. And the experimental verification of optimized combination of components of the Naomaitong significant effect on the improvement of various indicators. Conclusion Naomaitong group can improve the symptoms of ischemic stroke in varying degrees. It is feasible to optimize the compatibility of components based on uniform design and artificial neural network.