Native State of Metals in Non-Digested Tissues by Partial Least Squares Regression Analysis of Visible and Near-Infrared Spectra

Abstract

-A procedure has been used for the classification and quantification of metals on the basis of a chemometric analysis of visible and near-infrared (Vis-NIR) spectra of metals such as Cu, Mn and Fe in the brain, liver, kidney and testis of mice without digestion. Transmittance spectra in the 600-to 1000-nm region subjected to partial least-squares (PLS) regression analysis and leave-out cross-validation facilitated development of chemometrics models for predicting metal concentration. From the models, Cu, Mn and Fe yielded the coefficients of determination in cross-validation (R 2 VAL ) as 0.8013, 0.9021 and 0.8295 with standard errors of cross-validation (SECV) of 3.399, 0.8237 and 76.512 μg per g tissue, respectively. The respective detection limits of Cu, Mn and Fe were 12.19, 2.616 and 266.32 μg per g tissue. Furthermore, the regression coefficients of the models showed specific patterns for the respective metals. These results suggest that Vis-NIR spectroscopy may have a great potential for analysis of native state of metals in tissues.

Aquaphotomics: Dynamic Spectroscopy of Aqueous and Biological Systems Describes Peculiarities of Water

“…further on, it was possible to predict even the particle size of sugar and salt diluted in water when using the respective vis-nIr spectral model of the solution. Moreover, contrary to the common understanding of overtone spectroscopy, as the absorbance in the nIr range was very low, it showed that even very small concentrations (ppb level) [37][38][39][40]53,54 of the solute could be measured with vis-nIr. In another work, the result 55 obtained when using only characteristic absorbance bands to measure a component of interest (1620 nm was used as the polystyrene absorbance band) showed that the accuracy decreased with the decrease of concentration.…”

“…In a series of experiments, vis-nIr spectra were collected and regression models of respective perturbations were analysed in order to see if there were common water bands in various biological systems such as living cells under oxidative stress or environmental hormones added to cell cultures, 57 animal body (organ) tissue with low metal concentrations, 58 in vivo monitoring of plants inoculated with mosaic virus, 59,60 body fluids of healthy and mastitic cows, living tissue of prion inoculated mice 61 etc. It was found that a group of WaMacS repeatedly occurred in different combinations (water absorbance patterns) in the spectral models predicting the investigated perturbations.…”

The Multivariate Regression Statistics Strategy to Investigate Content-Effect Correlation of Multiple Components in Traditional Chinese Medicine Based on a Partial Least Squares Method

“…Because there was only one dependent variable, the partial least squares regression one prediction model was selected. Partial least squares regression analysis was employed to process the acquired content-effect data matrix following established methodology [ 23 , 24 , 25 , 26 ].…”

A Sparse Classification Based on a Linear Regression Method for Spectral Recognition

“…In addition, y i andf −k λ (x i ) denote the one dimensional real value and estimated value modeled by dividing by the k-th subset, respectively. As can be seen from Equation (12), the selection of λ affects the value of C(λ). Here, the value of C(λ) was the average of the quadratic sum of the error of each variable.…”

“…Sparse algorithms have previously been used for quantitative spectral analyses [11,12], but these studies make improvements based on the framework of common quantitative analyses, such as partial least squares or ridge regression. LRSC algorithm, introduced in our study, is applied to the field of recognition, and its framework is completely different from those of the currently available algorithms.…”