• DocumentCode
    3022674
  • Title

    Measurement of pulse oximetry based on dynamic spectroscopy

  • Author

    Wenjun Zhao ; Xiaofei Wang

  • Author_Institution
    Beijing Inf. Sci. & Technol. Univ., Beijing, China
  • fYear
    2013
  • fDate
    20-22 Dec. 2013
  • Firstpage
    1013
  • Lastpage
    1016
  • Abstract
    In the paper, a new way to measure pulse oxygen saturation based on Dynamic spectroscopy (DS) is validated. DS method can promote the accuracy of pulse oximetry evidently. It can eliminate the interference of individual discrepancy and measuring condition in principle. There were 60 patients in the ICU to be collected spectroscopic data and do biochemical analysis. The spectroscopic data was processed to extract DS with D-value estimation method. The support vector regression (SVR) algorithm was used to establish the calibration model of subjects´ oxygen saturation values measured by the biochemical analysis against the DS data including 660nm and 940nm wavelengths. The correlation coefficients and relative error were 0.8138 and ±0.0145 for validation set comparing 0.6230 and ±0.1164 for the same samples measured by the monitor, which is based on the traditional two-wavelength test method. The experimental results establish the efficacy of the proposed method.
  • Keywords
    biochemistry; calibration; correlation methods; data acquisition; data analysis; feature extraction; medical signal processing; oximetry; oxygen; regression analysis; spectral analysis; spectrochemical analysis; support vector machines; D-value estimation; DS extraction; ICU patients; O2; SVR algorithm; biochemical analysis; calibration model; correlation coefficients; dynamic spectroscopy; interference elimination; measuring condition; oxygen saturation value; pulse oximetry accuracy; pulse oxygen saturation measurement; relative error; spectroscopic data collection; spectroscopic data processing; support vector regression algorithm; traditional two-wavelength test method; validation set; wavelength 660 nm; wavelength 940 nm; Computers; Mechatronics; D-value estimation; Dynamic spectrum; Oxygen saturation;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Mechatronic Sciences, Electric Engineering and Computer (MEC), Proceedings 2013 International Conference on
  • Conference_Location
    Shengyang
  • Print_ISBN
    978-1-4799-2564-3
  • Type

    conf

  • DOI
    10.1109/MEC.2013.6885209
  • Filename
    6885209