Measurement Data Processing in Spectrophotometric Analysers of Food

Abstract

Spectrometry, especially spectrophotometry, is getting more and more often the method of choice not only in laboratory analysis of (bio)chemical substances, but also in the off-laboratory identification and testing of physical properties of various products, in particular - of various organic mixtures including food products and ingredients. Specialised spectrophotometers, called spectrophotometric analysers, are designed for such applications. This paper is on the state of the art in the domain of data processing in spectrophotometric analysers of food (including beverages). The following issues are covered: methodological background of food analysis, physical and metrological principles of spectrophotometry, the role of measurement data processing in spectrophotometry. General considerations are illustrated with examples, predominantly related to wine and olive oil analysis.

Spectral Features Characterizing Rice Wine “<i>Sake</i>” Variety Using Mid-Infrared Spectroscopy

“…The manufacturing process has many steps and the process control requires the experience of experts [1]. Especially, the brewing process of Sake simultaneously involves two enzymatic reactions, i.e., the dissolution / saccharifica- Ken-ichiro SUEHARA, Takaharu KAMEOKA, Atsushi HASHIMOTO 280 they are rapid, not expensive and "as-it-is" methods [5][6][7][8][9][10][11][12]. In addition, the information about all the organic components in a food is theoretically contained in the absorption of the infrared spectrum.…”

“…Therefore, some chemometric techniques, such as the principal component analysis (PCA) and partial least squares (PLS), are used to extract the spectral features and to develop the calibration equation for the quantitative analysis of the target components [5][6][7][8][9]. However, information which is explained by the calculated principal components (PCs) using the chemometrics models is not clear because the PCA is used to reduce the dimensionality and because the calculated PC, which is the maximized distribution of the data, is not always based on the target component.…”

Use of optical skin phantoms for preclinical evaluation of laser efficiency for skin lesion therapy

“…Then the phantom parameters (μ a and μ 0 s ) are retrieved by an inverse adding-doubling method from photometric measurements. 23,24 Total reflectance and total transmission are measured with a spectrophotometric system 25 with integrating spheres (OL-750, Optronic Laboratories), while the sample thickness and refractive indices 26 are measured by optical coherence tomography (Hyperion OCT System, Thorlabs) and the Abbe refractometer (Atago, Japan), respectively. The μ 0 s of prepared skin phantoms at the target wavelength of 975 nm is, as shown in Fig.…”

Blood equivalent phantom vs whole human blood, a comparative study

“…The spectra were later pre-processed with common de-noising algorithms such as Savitzky-Golay and fast Fourier transform¯ltering. 29 The spectra were also normalized with a Standard Normal Variate normalization algorithm. The integration time was 200 ms with over 20 averages for blood and RBC samples.…”