Application of Partial Robust M-Regression in Noninvasive Measurement of Human Blood Glucose Concentration with Near-Infrared Spectroscopy
LI Qing-bo1, YAN Hou-lai1, LI Li-na1, WU Jin-guang2, ZHANG Guang-jun1*
1. Key Laboratory of Precision Opto-mechatronics Technology, Ministry of Education, School of Instrument Science and Opto-Electronics Engineering, Beihang University, Beijing 100191, China 2. College of Chemistry and Molecular Engineering, Peking University, Beijing 100871, China
摘要： 采用偏稳健M回归方法有效地解决了人体血糖浓度近红外无创检测研究过程中由于样本奇异值影响模型稳健性的问题。该方法源于现有的迭代变权偏最小二乘法，计算快、易于实现，具有M估计的所有性质，且当权函数选择合适时，能降低奇异值的影响，建立具有稳健性的校正模型。采用该方法对近红外光谱实验数据进行了处理，并与传统的偏最小二乘(partial least squares，PLS)建模方法进行了比较。结果表明，与PLS相比，该方法可建立稳健的校正模型提高预测精度，更适合复杂样品建模，对于人体血糖浓度近红外无创检测的进一步研究具有应用价值。
关键词：偏稳健M回归;偏最小二乘;稳健性;近红外光谱;血糖浓度
Abstract：In the study of non-invasive measurement of human blood glucose concentration with near-infrared spectroscopy, the partial robust M-regression (PRM) is proposed in the present paper to solve the robustness of calibration model affected by outliers existing in the spectra data set. While keeping the good properties of M-estimators if an appropriate weighting scheme is chosen, PRM inherits the speed of computation and easy realization of the iterative reweighted partial least squares (IRPLS) algorithm, but is robust to all types of outliers. With the pretreatment of spectra based on PRM, the root mean square error of prediction (RMSEP) of calibration model was presented and compared with partial least squares (PLS). Experimental results show that the robust calibration model PRM produces better prediction of glucose than the model of PLS when the components of the samples increase which is significant for non-invasive prediction of blood glucose levels.
李庆波1，阎侯赖1，李丽娜1，吴瑾光2，张广军1* . 偏稳健M回归在人体血糖浓度近红外无创检测中的应用 [J]. 光谱学与光谱分析, 2010, 30(08): 2115-2119.
LI Qing-bo1, YAN Hou-lai1, LI Li-na1, WU Jin-guang2, ZHANG Guang-jun1* . Application of Partial Robust M-Regression in Noninvasive Measurement of Human Blood Glucose Concentration with Near-Infrared Spectroscopy . SPECTROSCOPY AND SPECTRAL ANALYSIS, 2010, 30(08): 2115-2119.
[1] WANG Hong, LIU Ze-yi, LI Qing-bo, et al(王 宏, 刘则毅, 李庆波, 等). Chinese Journal of Analytical Chemistry(分析化学), 2003, 31(4): 440. [2] XU Ke-xin, GAO Feng, ZHAO Hui-juan(徐可欣, 高 峰, 赵会娟). Biomedical Photonics(生物医学光子学). Beijing: Science Press(北京：科学出版社), 2007. 145. [3] Heise H M, Marbach R, Koschinsky T H, et al. J. Artif. Org., 1994, 18: 439. [4] Kim Y J, Hahn S, Yoon G. Appl. Opt., 2003, 42(4): 745. [5] Coté G L, Cameron B D. J. Biomedical Optics, 1997, 2(3): 275. [6] Berger A J, Koo T W, Itzkan I, et al. Appl. Opt., 1999, 38(13): 2916. [7] LU Wan-zhen(陆婉珍主编). Modern Near Infrared Spectroscopy Analytical Technology(Second Edition)(现代近红外光谱分析技术, 第2版). Beijing: China Petrochemical Press(北京：中国石化出版社), 2007. 1. [8] LI Qing-bo, XU Ke-xin, WANG Yan(李庆波, 徐可欣, 汪 曣). Journal of Tianjin University(天津大学学报), 2003, 36(2): 139. [9] WANG Hong, LI Qing-bo, LIU Ze-yi, et al(王 宏, 李庆波, 刘则毅, 等). Chinese Journal of Analytical Chemistry(分析化学), 2002, 30(7): 779. [10] Wold S, Sjstrm M, Eriksson L. Chemometr. Intell. Lab. Syst., 2001, 58: 109. [11] Serneels S, Croux C, Filzmoser P, et al. Chemometr. Intell. Lab. Syst., 2005, 79: 55. [12] Cummins D J, Andrews C. J. Chemom, 1995, 9: 489. [13] Hubert M, Vanden K Branden. J. Chemom, 2003, 17: 537. [14] ZHANG Guang-jun, LI Li-na, LI Qing-bo, et al(张广军，李丽娜，李庆波, 等). J. Infrared Millim. Waves(红外与毫米波学报), 2009, 28(2): 107.
