NIR Spectroscopy Combined with Stability and Equivalence MW-PLS Method Applied to Analysis of Hyperlipidemia Indexes
CHEN Jie-mei1, XIAO Qing-qing1, PAN Tao1*, YAN Xia1, 2, WANG Da-wei1, 2, YAO Li-jun1
1. Key Laboratory of Optoelectronic Information and Sensing Technologies of Guangdong Higher Educational Institutes, Jinan University, Guangzhou 510632, China 2. Second Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou 510120, China
Abstract:Moving window partial least square (MW-PLS) method was improved by considering the stability and equivalence, and was used for the wavelength optimization of reagent-free near-infrared (NIR) spectroscopic analysis of total cholesterol (TC) and triglycerides (TG) for hyperlipidemia. A random and stability-dependent framework of calibration, prediction, and validation was proposed. From all human serum samples (negative 145 and positive 158, a total of 303 sample), 103 samples (negative 44 and positive 59) were randomly selected for the validation set, the remaining samples (negative 101 and positive 99, a total of 200 sample) were used as modeling set; then the modeling set was randomly divided into calibration set (negative 51 and positive 49, a total of 100 sample) and prediction set (negative 50 and positive 50, a total of 100 sample) by 50 times. To produce modeling stability, the model parameters were optimized based on the average prediction effect for all divisions; the optimized models were validated by using the validation samples. The obtained optimal MW-PLS wavebands were 1 556~1 852 nm for TC and 1 542~1 866 nm for TG. In order to solve the problem that instrument design typically involves some limitations of position and number of wavelengths because of cost and material properties, the equivalent model sets were proposed, and a unique public waveband 1 542~1 852 nm of the equivalent model sets for TC, TG was found. The validation results show that: using the optimal MW-PLS wavebands, validation samples’ root mean square error of prediction (V_SEP) for TC, TG were 0.177, 0.100 mmol·L-1, the correlation coefficient of prediction (V_RP) for TC, TG were 0.988, 0.996, and the sensitivity and specificity for hyperlipidemia achieved 95.0%, 90.5%, respectively; using the public equivalent wavebands, the V_SEP for TC, TG were 0.177, 0.101 mmol·L-1), the V_RP for TC, TG were 0.988, 0.996, and the sensitivity and specificity achieved 92.7%, 90.3%, respectively. Conclusion: NIR spectroscopy combined with the stability and equivalence-improvement MW-PLS method can provide a potential tool for detecting hyperlipidemia for large population.
[1] Chen H Z, Pan T, Chen J M. Chemometr. Intell. Lab, 2011, 107(1): 139. [2] Pan T, Li M M, Chen J M. Appl. Spectrosc., 2014, 68(3): 263. [3] PAN Tao, WU Zhen-tao, CHEN Hua-zhou(潘 涛, 吴振涛, 陈华舟). Chinese J. Anal. Chem.(分析化学), 2012, 40(6): 920. [4] Liu Z Y, Liu B, Pan T. Spectrochim. Acta A, 2013, 102: 269. [5] CAO Pu,PAN Tao, CHEN Xing-dan(曹 璞,潘 涛,陈星旦). Optics Preci. Eng.(光学精密工程), 2007,15(12): 1952. [6] Pan T, Chen Z H, Chen J M. Anal. Methods, 2012, 4(4): 1046. [7] CHU Xiao-li, YUAN Hong-fu, LU Wan-zhen(褚小立, 袁洪福, 陆婉珍). Prog. Chem.(化学进展), 2004, 16(4): 528. [8] CHU Xiao-li, XU Yu-peng, LU Wan-zhen(褚小立, 许育鹏, 陆婉珍). Chinese J. Anal. Chem.(分析化学), 2008, 36(5): 702. [9] Du Y P, Liang Y Z, Jiang J H. Anal. Chim. Acta, 2004, 501(2): 183. [10] XIE Jun, PAN Tao, CHEN Jie-mei(谢 军, 潘 涛, 陈洁梅). Chinese J. Anal. Chem.(分析化学), 2010, 38(3): 342. [11] LIU Zhen-yao, PAN Tao(刘振尧,潘 涛). Optics Preci. Eng.(光学精密工程), 2012, 20(10): 2171. [12] Pan T, Liu J M, Chen J M. Anal. Methods, 2013, 5(4): 4355. [13] Hazen K H, Arnold M A, Small G W. Anal. Chim. Acta, 1998, 371(2-3): 255. [14] CHEN Hua-cai, YANG Zhong-guo(陈华才, 杨仲国). Chinese J. Anal. Lab.(分析试验室), 2005, 24(7): 17.