Optimization of mid-infrared noninvasive blood-glucose prediction model by support vector regression coupled with different spectral features

• Published in: Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy Vol. 321; p. 124738
• Main Authors: Song, Liying, Han, Zhiqiang, Lau, Woon-Ming
• Format: Journal Article
• Language: English
• Published: England Elsevier B.V 15.11.2024
• Subjects: ATR-FTIR; Blood glucose; Blood Glucose - analysis; Humans; Male; Noninvasive; Regression Analysis; Spectrophotometry, Infrared - methods; Spectroscopy, Fourier Transform Infrared - methods; Support Vector Machine; Support vector regression; Blood glucose; Support vector regression; ATR-FTIR; Noninvasive

[Display omitted] •The accuracy of noninvasive blood-glucose test is improved to meet FDA standards.•Hypothenar is the optimal body-spot for collecting subcutaneous blood-glucose signals.•Exploiting the minimization of spectral interference yields the lowest prediction error. Mid-infrared spectral a...