Laser & Optoelectronics Progress, Volume. 56, Issue 4, 041701(2019)

Comparison of Multi-Factor-Considered Blood Glucose Prediction Models by Near-Infrared Spectroscopy

Xiaofei Wang*, Xinyi Zhang, and Xinhe Xu
Author Affiliations
  • School of Instrumentation Science and Optoelectronic Engineering, Beijing Information Science and Technology University, Beijing 100192, China
  • show less
    Figures & Tables(6)
    Experiment scheme for measuring blood glucose
    RBF modeling results with non-measurement-component considered. (a) Predicted values and true values; (b) relative errors
    Linear kernel function modeling results with non-measurement-component considered. (a) Predicted values and true values; (b) relative errors
    Modeling results without non-measurement-component considered. (a) Predicted values and true values; (b) relative errors
    Relative errors of SVM model
    • Table 1. Model parameters

      View table

      Table 1. Model parameters

      ModelingTraining setPrediction set
      RRMSECRRMSEP
      With non-measurement-component considered0.99930.020.96270.13
      Without non-measurement-component considered0.93440.170.86550.23
    Tools

    Get Citation

    Copy Citation Text

    Xiaofei Wang, Xinyi Zhang, Xinhe Xu. Comparison of Multi-Factor-Considered Blood Glucose Prediction Models by Near-Infrared Spectroscopy[J]. Laser & Optoelectronics Progress, 2019, 56(4): 041701

    Download Citation

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

    Category: Medical Optics and Biotechnology

    Received: Sep. 4, 2018

    Accepted: Sep. 18, 2018

    Published Online: Jul. 31, 2019

    The Author Email: Wang Xiaofei (wangxiaofei@bistu.edu.cn)

    DOI:10.3788/LOP56.041701

    Topics