Spectroscopy and Spectral Analysis, Volume. 41, Issue 9, 2754(2021)

Application of SG-MSC-MC-UVE-PLS Algorithm in Whole Blood Hemoglobin Concentration Detection Based on Near Infrared Spectroscopy

Dai-qing SUN1、*, Li-rong XIE1、1; *;, Yan ZHOU2、2;, Yu-tao GUO1、1;, and Shao-min CHE2、2;
Author Affiliations
  • 11. School of Electrical Engineering, Xinjiang University, Urumqi 830047, China
  • 22. School of Energy & Power Engineering, Xi'an Jiaotong University, Xi'an 710049, China
  • show less
    Figures & Tables(4)
    R2 curves for the original spectra of whole blood signals(a): Transmission spectrum; (b): Absorbance spectrum
    R2* vs. the width of the SG-MSC method(a): R2*vs. the width; (b): R2 after preporcessing by SG-MSC
    • Table 1. Comparison of R2* values of different preprocessing methods

      View table
      View in Article

      Table 1. Comparison of R2* values of different preprocessing methods

      Method原始透射原始吸收中心化标准化SNVMSCSG-MSC
      R2*0.003 50.005 00.005 00.005 00.070 60.105 20.944 1
    • Table 2. Determination results for PLS models

      View table
      View in Article

      Table 2. Determination results for PLS models

      SignalsNWR2RMSEPMAEMRE
      原始透射PLS7000.676 30.898 12.426 10.140 24
      原始吸收PLS7000.896 10.464 21.630.094 2
      SG-MSC- PLS7000.972 80.229 00.494 30.028 6
      SG-MSC-MC-UVE-PLS1910.979 10.220 30.411 20.023 8
      二阶导数+UVE+PLS2090.967 60.300 3--
    Tools

    Get Citation

    Copy Citation Text

    Dai-qing SUN, Li-rong XIE, Yan ZHOU, Yu-tao GUO, Shao-min CHE. Application of SG-MSC-MC-UVE-PLS Algorithm in Whole Blood Hemoglobin Concentration Detection Based on Near Infrared Spectroscopy[J]. Spectroscopy and Spectral Analysis, 2021, 41(9): 2754

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: Research Articles

    Received: Sep. 8, 2020

    Accepted: --

    Published Online: Oct. 29, 2021

    The Author Email: SUN Dai-qing (1248429197@qq.com)

    DOI:10.3964/j.issn.1000-0593(2021)09-2754-05

    Topics