伸缩移动窗口的峰识别用于药品拉曼光谱分析

Scalable moving-window peaks identification for drug detection by Raman spectroscopy

目的提出一种基于伸缩移动窗口的拉曼光谱峰识别方法,并且将其应用到药品的真假判别中。方法采用一阶导数和插值法构建伸缩移动窗口,在移动窗口的同时进行窗口内的峰识别,利用该伸缩移动窗口的峰识别法对药品进行真假判别,根据药品的活性成分(API)的拉曼光谱峰宽设定移动到相应位置的窗口的大小,然后分别识别窗口内API以及药品的光谱峰,对比两者的峰值和波长点以及计算两者的相似度。结果可以直观看到API拉曼光谱的峰在药品光谱信号中的贡献。根据峰的贡献度辨别药品的真假,其中7种药品被判为真药,2种药品被判为假药,准确率为100%。结论伸缩移动窗口的拉曼光谱峰识别可以用于药品的真假判别分析,方法简单有效。

AIM To develop a Raman spectral peak identification method based on scalable moving-window and apply it for drug authenticity discrimination. METHODS A scalable moving-window was established by u- sing the first derivative and interpolation method. The peaks within the window were identified while the window was moving. The peaks identification method of the sealable moving-window was used to discriminate drugs. The window size was set according to the peak＇ s width of active pharmaceutical ingredient （API） spectrum at the corre- sponding position. Then, the peaks of both API and drug spectra within the window were detected. The peak＇ s val- ue and wavelength point of the two was compared, and the correlation coefficient between the two was computed. RESULTS The contribution of API spectral peaks in the drug spectra signal was apparent. According to the contri- bution rate of peaks, the authenticity of drugs was identified. Among them, seven drugs were judged as real drugswhile two drugs were fakes. The accuracy rate was 100%. CONCLUSION It is simple and effective to discrlmi-nate drugs by using scalable moving-window based peaks identification.