小波变换计算分光光度法可行性研究(Ⅱ.实验研究)  

Research into Computational Spectrophotometry Based on Wavelet Transform(Ⅱ.Experimently Study)

作　　者：王艳[1] 相秉仁[1] 

机构地区：[1]中国药科大学分析计算中心

出　　处：《药物分析杂志》2005年第10期1225-1229,共5页Chinese Journal of Pharmaceutical Analysis

摘　　要：目的:通过实验研究基于小波变换的计算分光光度法的可行性。方法:以紫外分光光度法测定双嘧达莫(λ_(max)=292nm)含量为例,设计三组实验,分别加入与双嘧达莫紫外吸收光谱重叠的鞣酸(λ_(max)=276 nm)、盐酸普萘洛尔(λ_(max)=289 nm)和盐酸普罗帕酮(λ_(max)=305nm),对每组的紫外吸收光谱数据进行连续小波变换,共研究了34个小波,提取只与双嘧达莫含量有关的特征小波系数。结果:每组中都存在很多特征小波系数,用该组的特征小波系数建立的线性回归方程,都可以准确地预测该组样品中双嘧达莫的含量,相对误差均小于4%。结论:基于小波变换的计算分光光度法可以应用于吸收光谱重叠的分析体系。Objective： To study the feasibility of computational spectrophotometry based on wavelet transform by experiment. Method：Chose dipyridamole （λmax = 292 nm）as the component wanted to be measured by ultraviolet spectrophotometry,and three groups are designed by three components with overlapping absorption spectrum are added respectively, tannin （ λmax = 276 nm）, propranol hydrochloride （ λmax = 289 nm） and propafenone hydrochloride （λmax =305 nm）. In each group,the ultraviolet spectra are manipulated by continuous wavelet transform to obtain eigen wavelet coefficients that only correlated with the concentration of dipyridamole. Results：There are many eigen wavelet coefficients could be obtained in each group, and simulative samples could be accurately determined with the relative error 〈 4% by linear regression equation established by eigen wavelet coefficients. Conclusion： Computational spectrophotometry based on wavelet transform can be applied to the analytical system with overlapping absorption spectra.

关 键 词：小波变换 连续小波变换 计算分光光度法 实验研 紫外吸收光谱 紫外分光光度法 双嘧达莫 盐酸普罗帕酮 盐酸普萘洛尔 线性回归方程 

分 类 号：R241.1[医药卫生—中医诊断学] R971.2[医药卫生—中医临床基础]