三七总皂苷提取物物理指纹图谱构建及预测方法

Construction and prediction of physical fingerprints of notoginseng total saponins extracts

【目的】为了丰富中药质量一致性评价手段,促进新原料的风险评估,提升中药原料的质量控制水平。【方法】以三七总皂苷提取物为研究对象,借鉴中药化学指纹图谱的概念,构建其物理指纹图谱,通过数据驱动的方法对不同批次原料之间的差异进行深入分析并探索关键物理属性参数之间的关系;根据主成分分析和层次聚类分析的结果划分样品类别,在此基础上建立分类模型并筛选出造成样品物理属性质量差异的重要变量;采用近红外光谱技术对不同原料的质量进行快速识别并建立重要物理属性参数的快速预测方法。【结果】首先清晰地阐明关键物理属性参数之间的关系,筛选出对样本分类贡献较大的物理属性参数;建立能够快速有效鉴别原料差异的分类模型;采用近红外光谱技术对不同原料的质量进行快速识别并建立重要物理属性参数的快速预测方法。【结论】所建方法可以对不同批次三七总皂苷提取物的物理属性质量进行准确分析,能够最大化地从物理属性参数数据中提取有效信息,从而提升药品生产过程中原料的质量控制水平。

Objective The current research on the quality consistency evaluation of traditional Chinese medicine mainly focuses on the chemical level,and often overlooks the examination of physical quality attributes.In the production process of traditional Chinese medicine formulations,the physical properties of raw materials significantly impact the formulation and molding quality.As a supplement to the evaluation of chemical properties,the construction of physical fingerprints can reflect the overall properties of raw materials,thereby providing data support for the application of design concepts in the quality control research of traditional Chinese medicine raw materials.Method This study takes the total saponin extract of Panax notoginseng saponins as the research object and evaluates the consistency of samples from different manufacturers from a physical perspective by constructing its physical fingerprint map.A NIR method was developed to address the issue of tedious and time-consuming determination process of physical property parameter,which is not conducive to real-time release during production.This method is used to quickly identify the physical property quality of different raw materials and establish a fast prediction model for important physical property parameters.Results This method can accurately analyze the physical quality of different batches of Panax ginseng total saponin extracts,maximizing the extraction of effective information from the physical property data and facilitating the selection of raw materials in the production process of formulations.It can be used by companies in the early development stage of process design.In addition,the data analysis method presented in this paper is effective in the process of product development and process optimization,as most results can be represented graphically,allowing production staff without a background in data modelling to easily understand the changing patterns of the data.Conclusion A NIR method has been developed to address the issue of tedious and time-consuming physical property parameter determination process,which is not conducive to real-time release during production.This method is used to quickly identify the physical property quality of different raw materials and establish a fast prediction model for important physical property parameters.The method developed in this study could accurately analyze the physical quality of different batches of Panax ginseng total saponin extracts,which could maximize the extraction of effective information from the physical property data and facilitate the selection of raw materials in the production process of formulations and could be used by companies in the early development stage of process design.In addition,the data analysis method presented in this paper is effective in the process of product development and process optimization as most of the results can be represented graphically so that production staff without a background in data modelling can easily understand the changing patterns of the data.This method can improve the production efficiency of enterprises and enhance the quality control of traditional Chinese medicine.At the same time,this research idea is not limited to the powder of Panax ginseng total saponin extract,but is also applicable to other raw materials of traditional Chinese medicine,as well as auxiliary materials and other powdered substances.