11 October 2010 Optimal waveband and mathematical model for analysis of human whole blood glucose by near infrared transmission spectroscopy
Author Affiliations +
Abstract
The optimal waveband for quantitative analysis of the human whole blood glucose by near infrared transmission spectroscopy is discussed in this paper. First, whole blood samples of different thickness (0.5mm, 1mm, 2mm, 4mm ) of the near-infrared transmittance spectra were analyzed respectively. It shows that the sample thickness of 0.5mm is more suitable for spectral analysis. And then near infrared spectroscopy of 111 samples of 0.5mm thickness were collected. Finally, different pretreatment methods of scattering correction methods, derivatives and different modeling spectral regions were compared to find out their impact on mathematical prediction model. The result shows that the best prediction accuracy was obtained in the waveband of 1500~1900nm by using Standard Normal Variate(SNV),the second derivative spectra and Partial Least Square (PLS) regression. And the correlation coefficient(RP), the Root Mean Square Error of Prediction(RMSEP) and the Relative Root Mean square Error of Prediction(RRMSEP) for the corresponding model are 0.836, 0.271, 5.519%, respectively.
© (2010) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Furong Huang, Jianhui Yu, Shiping Li, Dongming Wang, Yunhan Luo, and Shifu Zheng "Optimal waveband and mathematical model for analysis of human whole blood glucose by near infrared transmission spectroscopy", Proc. SPIE 7656, 5th International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optical Test and Measurement Technology and Equipment, 76566W (11 October 2010); doi: 10.1117/12.864035; https://doi.org/10.1117/12.864035
PROCEEDINGS
7 PAGES


SHARE
Advertisement
Advertisement
Back to Top