Multivariate Determination of Glucose in Whole Blood Using Partial Least-Squares and Artificial Neural Networks Based on Mid-Infrared Spectroscopy

Abstract

The infrared (IR) spectra of whole blood EDTA samples, in the range between 1500 and 750 cm−1, obtained from the patient population of a general hospital, were used to compare different multivariate calibration techniques for quantitative glucose determination. Ninety-six spectra of whole undiluted blood samples with glucose concentration ranging between 44 and 291 mg/dL were used to create calibration models based on a combination of partial least-squares (PLS) and artificial neural network (ANN) methods. The prediction capabilities of these calibration models were evaluated by comparing their standard errors of prediction (SEP) with those obtained with the use of PLS and principal component regression (PCR) calibration models in an independent prediction set consisting of 31 blood samples. The optimal model based on the combined PLS-ANN produced smaller SEP values (15.6 mg/dL) compared with those produced with the use of either PLS (21.5 mg/dL) or PCR (24.0 mg/dL) methods. Our results revealed that the combined PLS-ANN models can better approximate the deviations from linearity in the relationship between spectral data and concentration, compared with either PLS or PCR models.

Infrared Spectroscopy of Biofluids in Clinical Chemistry and Medical Diagnostics

“…Mid-IR investigations [26][27][28][29][30][31][32] have typically yielded assays with standard errors of prediction of approximately 0.6 mmol/L, independent of the experimental details (e.g. ATR vs. transmission, liquid vs. dry film).…”

Vibrational Spectroscopy Applications in Clinical Chemistry

“…One solution is to use attenuated total reflection (ATR) spectroscopy rather than transmission measurements. A series of publications from Heise and co-workers has demonstrated that this approach can provide the basis to quantify native plasma, 3,4 whole blood, 5,6 and urine 7 constituents. A second approach is to eliminate the water absorptions by simply drying the specimens, and carrying out transmission measurements on the dry films.…”

Glucose, In Vivo Assay of

“…37 A different approach, which is able to model nonlinearities explicitly, is the application of neural networks. 33,38 As in other multivariate methods, calibration by ''learning'' is essential, but such an approach tends to be more susceptible to overfitting, and more difficult to handle and interpret than the linear approaches discussed before.…”