Quantitative Analysis of Thiram by Surface-Enhanced Raman Spectroscopy Combined with Feature Extraction Algorithms
ZHANG Bao-hua1, JIANG Yong-cheng1, SHA Wen1, ZHANG Xian-yi2, CUI Zhi-feng2*
1. School of Electronic and Information Engineering, Anhui University, Hefei 230601, China 2. Institute of Atomic and Molecular Physics, Anhui Normal University, Wuhu 241000, China
Abstract：Three feature extraction algorithms, such as the principal component analysis (PCA), the discrete cosine transform (DCT) and the non-negative factorization (NMF), were used to extract the main information of the spectral data in order to weaken the influence of the spectral fluctuation on the subsequent quantitative analysis results based on the SERS spectra of the pesticide thiram. Then the extracted components were respectively combined with the linear regression algorithm—the partial least square regression (PLSR) and the non-linear regression algorithm—the support vector machine regression (SVR) to develop the quantitative analysis models. Finally, the effect of the different feature extraction algorithms on the different kinds of the regression algorithms was evaluated by using 5-fold cross-validation method. The experiments demonstrate that the analysis results of SVR are better than PLSR for the non-linear relationship between the intensity of the SERS spectrum and the concentration of the analyte. Further, the feature extraction algorithms can significantly improve the analysis results regardless of the regression algorithms which mainly due to extracting the main information of the source spectral data and eliminating the fluctuation. Additionally, PCA performs best on the linear regression model and NMF is best on the non-linear model, and the predictive error can be reduced nearly three times in the best case. The root mean square error of cross-validation of the best regression model (NMF+SVR) is 0.045 5 μmol·L-1 (10-6 mol·L-1), and it attains the national detection limit of thiram, so the method in this study provides a novel method for the fast detection of thiram. In conclusion, the study provides the experimental references the selecting the feature extraction algorithms on the analysis of the SERS spectrum, and some common findings of feature extraction can also help processing of other kinds of spectroscopy.
[1] Lee S, Chon H, Lee J, et al. Biosens Bioelectron, 2014, 51: 238. [2] Bi L, Dong J, Xie W, et al. Anal. Chim. Acta, 2013, 805: 95. [3] Muller C, David L, Chis V, et al. Food Chem., 2014, 145: 814. [4] Bocklitz T, Walter A, Hartmann K, et al. Anal. Chim. Acta, 2011, 704(1-2): 47. [5] Temiz H T, Boyaci I H, Grabchev I, et al. Spectrochim Acta A Mol. Biomol. Spectrosc., 2013, 116: 339. [6] Kokot S, Czarnik-Matusewicz B, Ozaki Y. Biopolymers, 2002, 67(6): 456. [7] Rivera-Betancourt O E, Karls R, Grosse-Siestrup B, et al. Analyst, 2013, 138(22): 6774. [8] Phamila Y A V, Amutha R. Signal Processing, 2014, 95: 161. [9] Szatmary B, Szirtes G, Lorincz A, et al. Pattern Analysis & Applications, 2003, 6(3): 194. [10] WENG Shi-zhuang, CHEN Sheng, ZENG Xin-hua, et al(翁士状, 陈 晟, 曾新华, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2013, 33(9): 2438. [11] Moore B. IEEE Transactions on Automatic Control, 1981, 26(1): 17. [12] Rafiq Abuturab M. Optics and Lasers in Engineering, 2012, 50(10): 1383. [13] Phon-Amnuaisuk S. EURASIP Journal on Audio, Speech, and Music Processing, 2012, 2012(1): 11. [14] Kachrimanis K, Braun D E, Griesser U J. J. Pharm. Biomed. Anal., 2007, 43(2): 407. [15] Smola A J, Schlkopf B. Statistics and Computing, 2004, 14(3): 199. [16] Saute B, Narayanan R. Analyst, 2011, 136(3): 527.
