Advanced Statistical Techniques for Noninvasive Hyperglycemic States Detection in Mice Using Millimeter-Wave Spectroscopy Articles uri icon

publication date

  • May 2020

start page

  • 237

end page

  • 245

issue

  • 3

volume

  • 10

International Standard Serial Number (ISSN)

  • 2156-342X

Electronic International Standard Serial Number (EISSN)

  • 2156-3446

abstract

  • In this article, we discuss the use of advanced statistical techniques (functional data analysis) in millimeter-wave(mm-wave) spectroscopy for biomedical applications. We employ aW-band transmit&-receive unit with a reference channel to acquirespectral data. The choice of the W-band is based on a tradeoffbetween penetration through the skin providing an upper boundfor the frequencies and spectral content across the band. The dataobtained are processed using functional principal component logitregression (FPCLoR), which enables to obtain a predictive modelfor sustained hyperglycemia, typically associated with diabetes.The predictions are based on the transmission data from noninvasive mm-wave spectrometer at W-band. We show that there existsa frequency range most suitable for identification, classification,and prediction of sustained hyperglycemia when evaluating thefunctional parameter of the functional logit model (beta). This allowsfor the optimization of the spectroscopic instrument in the aim toobtain a compact and potential low-cost noninvasive instrumentfor hyperglycemia assessment. Furthermore, we also demonstratethat the statistical tools alleviate the problem of calibration, which is a serious obstacle in similar measurements at terahertz and IRfrequencies.

keywords

  • functional data analysis (fda); functional principal components logit regression (fpclor); millimeter-wave (mm-wave) spectroscopy; noninvasive diagnostics; sustained hyperglycemia; w-band reflectometer; w-band transmit/receive unit