No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Near-infrared spectra of serum albumin and -globulin and determination of their concentrations in phosphate buffer solutions by partial least squares regression
Abstract
Near-infrared (NIR) spectra have been measured for albumin and γ-globulin in a powder state as well as in phosphate buffer solutions. The second derivative spectra of powder samples have been used to make assignments of NIR bands of the proteins. The second derivative spectra of albumin and γ-globulin are significantly different from each other in the band frequencies and relative intensities. The NIR spectra in the 1300–1850 nm region of the solutions have been subjected to partial least squares (PLS) regression analysis to develop chemometrics models which predict the concentrations of the proteins. The calibration for the albumin solutions in the concentration range of 0.1–8.0 g/dl has yielded a correlation coefficient (R) of 0.9995 and a standard error of calibration (SEC) of 0.207 g/dl. For the γ-globulin solutions in the concentration range of 0.1–6.0 g/dl, R of 0.9946 and SEC of 0.128 g/dl have been obtained. Regression coefficients (RCs) for the calibration models have been calculated for the first four factors. These RCs reflect the spectral variations in bands due to the proteins and in a water band near 1400 nm caused by the dissolution of the proteins. Moreover, the RCs have been compared with the NIR spectra of the proteins in the powder state. The positions of peaks in the RCs correspond well to those of bands in the NIR spectra of the proteins in the powder state. This suggests that the chemometrics models can pick up effectively the information about albumin and γ-globulin, even if the models have been constructed from the spectra of the proteins in the dilute solutions.
... . Only a few applications could be found in biological samples [13][14][15][16][17][18][19]. Direct determination of proteins in serum samples by NIR 1 spectrometry is probably limited due to the strong interference of water (and other components) on protein bands that occur in the NIR spectral region. ...
... This technique was already employed for the determination of total protein content in several kinds of food sam-ples like grains [3,4], cheese [5], milk [6][7][8], sweet potato [9], bread [10], fish [11], and meat [12]. Only a few applications could be found in biological samples [13][14][15][16][17][18][19]. Direct determination of proteins in serum samples by NIR 1 spectrometry is probably limited due to the strong interference of water (and other components) on protein bands that occur in the NIR spectral region. ...
... A PLS2 algorithm was employed in order to achieve simultaneous determination of the analytes. In a previous work [14], the same research group studied the NIR spectroscopic behavior of these analytes in model phosphate buffer solutions. ...
In the present work, the determination of the total protein concentration in hyperimmune serum samples was performed through a partial least-squares near-infrared (NIR-PLS) method. The method was based on the chemometric treatment of the NIR spectra of samples. The influences of spectra preprocessing and spectral window utilized in the construction of PLS model were studied. Models were built using reference data of 19 samples selected through the use of hierarchical cluster analysis (HCA) of NIR spectra of samples and another 24 samples were employed for external validation of the method. A model with better prediction capacity was obtained after whole spectra preprocessing by multiplicative scattering correction (MSC) algorithm and using data in the spectral range of 2158-2209nm. Under optimized conditions a RMSEP of 0.21gdl(-1) and a quality coefficient value (QC) of only 5.8% were obtained for the prediction of total protein content in the samples used for external validation. Also, a determination coefficient, r(2), of 0.97 was obtained in the correlation of predicted and reference data of samples situated in the validation set.
... Aquaphotomics was also applied for various detection, quantification, monitoring and diagnostic purposes (11)(12)(13)(14)(15). This methodology also served for analysis of folding of prion proteins (16), study of nucleation of amyloid proteins (17), denaturation of albumin (18), as well as measuring of concentrations of albumin and gamma globulin in solutions and serum (19,20). ...
... Since the deposits on contact lenses are primarily composed of proteins (2, 6, 7), the importance of C6 water matrix coordinate for developed PCA and PLS-DA models suggest that this this band, which was shown to be important for water-protein interaction (9,16,17), is also important for the detection of proteins in spoiled contact lenses. The difference in NIR spectra of spoiled and new contact lenses detected with PCA, PLS-DA analysis and depicted in aquagrams suggests the possibility for application of NIR spectroscopy and Aquaphotomics for precise quantification of proteins indirectly, by studying changes in the water absorbance pattern and developments of calibration models based on comparative quantification of proteins by standard methods in controlled in-vitro conditions, which is further supported by literature data (19,20). The continuation of this preliminary study could also provide deeper understanding of the mechanism of protein conformational changes which is reported in Aquaphotomics literature recently (17), because these conformational changes are in the root of the entire process of their adsorption. ...
The aim of this study was to determine if it is possible to distinguish between the groups of spoiled and unspoiled soft contact lenses using near infrared spectroscopy and new analytical approach –Aquaphotomics.Using principal component analysis it was established that the absorbance spectra of worn and new contact lenses is differed at water absorption band related to hydration of proteins. Detection of proteins thus, was performed indirectly, by using vibrations of water molecules.This exploratory study showed that near infrared spectroscopy and Aquaphotomics have potential for non-invasive, chemical free detection of protein deposits on hydrated soft contact lenses.
... Recent studies have shown that biologically important molecules such as albumin [48][49][50], cholesterol [51,52], globulin [48][49][50], glucose [50,[53][54][55][56][57][58][59][60][61], protein [49,52,[62][63][64][65], urea [49,50,64], lipid [52], linoleic acid [52], collagen [52], DNA [52] and α-elastin [52] can be investigated by NIR spectroscopy. In addition, NIR spectroscopy has also been used to quantitatively determine creatine [64], lactate [59,66], triacetin [57], triglyceride [50], β-lipoprotein [62], Vibrio cholerae [67], Escherichia coli [68,69], yeast [70,71], ethanol [71,72], RNA [65], acetate [69], ammonia [59,69], glycerol [69] and glutamine [59]. ...
... Recent studies have shown that biologically important molecules such as albumin [48][49][50], cholesterol [51,52], globulin [48][49][50], glucose [50,[53][54][55][56][57][58][59][60][61], protein [49,52,[62][63][64][65], urea [49,50,64], lipid [52], linoleic acid [52], collagen [52], DNA [52] and α-elastin [52] can be investigated by NIR spectroscopy. In addition, NIR spectroscopy has also been used to quantitatively determine creatine [64], lactate [59,66], triacetin [57], triglyceride [50], β-lipoprotein [62], Vibrio cholerae [67], Escherichia coli [68,69], yeast [70,71], ethanol [71,72], RNA [65], acetate [69], ammonia [59,69], glycerol [69] and glutamine [59]. ...
... Water electromagnetic spectrum. (http://www1.lsbu.ac.uk/water/vibrat.html)Reference:[73] NOTE COPYRIGHT ON THIS FIGURE. ...
... NOTE COPYRIGHT ON THIS FIGURE. Need approval to reproduce.Table 1. Water matrix co-ordinates (WAMACS): characteristic water absorbance bands with a high contribution in respective spectral models developed under various perturbations: concentrations, temperature, illumination, disease, damage, host molecules in the structure or in solution, particles size.40,43,50,66,68,[72][73][74] ...
Aquaphotomics is a new discipline that provides a framework for understanding changes in water molecular system presented as a water spectral pattern, to mirror the rest of the solution and to give a holistic description related to system functionality. One of its main purposes is to identify water bands as main coordinates of future absorbance patterns to be used as a system biomarker. This chapter presents the Aquaphotomics methodology and illustrates a way to identify specific water bands using temperature change and addition of solutions of different ionic strength as perturbations. Rapid and precise measurement of low concentration solutes has been given as a strong evidence of the vast information that “the water spectral pattern as molecular mirror” approach provides. Few applications using near infrared spectroscopy and multivariate analysis as main tools of Aquaphotomics have been presented.
... Therefore, NIR spectroscopy has become a widely used analytical method in agricultural, pharmaceutical, chemical and petrochemical industries (3,15,19,28). Recently, biologically important molecules such as urea (8), protein (8,25,28), cholesterol (17), albumin (14) and γ-globulin (14) have been investigated by NIR spectroscopy. Globulin (5), lipoprotein (5) and total proteins (30) in human serum, protein (25), creatine (25), urea (25) and glucose (29) in urine and glucose in blood (9) have been quantitatively determined by NIR spectroscopy. ...
... Therefore, NIR spectroscopy has become a widely used analytical method in agricultural, pharmaceutical, chemical and petrochemical industries (3,15,19,28). Recently, biologically important molecules such as urea (8), protein (8,25,28), cholesterol (17), albumin (14) and γ-globulin (14) have been investigated by NIR spectroscopy. Globulin (5), lipoprotein (5) and total proteins (30) in human serum, protein (25), creatine (25), urea (25) and glucose (29) in urine and glucose in blood (9) have been quantitatively determined by NIR spectroscopy. ...
... Therefore, NIR spectroscopy has become a widely used analytical method in agricultural, pharmaceutical, chemical and petrochemical industries (3,15,19,28). Recently, biologically important molecules such as urea (8), protein (8,25,28), cholesterol (17), albumin (14) and γ-globulin (14) have been investigated by NIR spectroscopy. Globulin (5), lipoprotein (5) and total proteins (30) in human serum, protein (25), creatine (25), urea (25) and glucose (29) in urine and glucose in blood (9) have been quantitatively determined by NIR spectroscopy. ...
... Therefore, NIR spectroscopy has become a widely used analytical method in agricultural, pharmaceutical, chemical and petrochemical industries (3,15,19,28). Recently, biologically important molecules such as urea (8), protein (8,25,28), cholesterol (17), albumin (14) and γ-globulin (14) have been investigated by NIR spectroscopy. Globulin (5), lipoprotein (5) and total proteins (30) in human serum, protein (25), creatine (25), urea (25) and glucose (29) in urine and glucose in blood (9) have been quantitatively determined by NIR spectroscopy. ...
Presently, the diagnosis of virus infections is based mainly on serological assays. Although polymerase chain reaction (PCR) and enzyme-linked immunosorbent assay (ELISA) have been increasingly used for the diagnosis of such viral infections, the risk of transfusion-transmitted blood-borne viruses remains. Furthermore, PCR and ELISA are expensive and time-consuming, and sometimes cause falsepositive or false-negative results. Therefore, a rapid, accurate and cost-effective diagnostic procedure is needed. We subjected plasma from individuals infected with human immunodeficiency virus type-1 (HIV-1), the causative agent of acquired immune deficiency syndrome (AIDS), as well as plasma from uninfected individuals as a control to near-infrared (NIR) spectroscopy, which may provide a rapid diagnostic method for HIV-1 infection without using any reagent. NIR spectra in the 600-1,000 nm region for plasma from pre-serologically HIV-1-infected individuals and healthy donors were subjected to partial least squares (PLS) regression analysis and leave-out cross-validation to develop a multivariate model to estimate the concentration of HIV-1. Simultaneously, the same plasma samples were examined for HIV-1 p24 by ELISA. The results obtained by the NIR spectroscopy model for HIV-1 yielded a good correlation with those obtained by the reference method (HIV-1 p24 ELISA). These results suggest that NIR spectroscopy using plasma could provide a rapid, accurate, cost-effective tool for large-scale diagnosis of HIV-1 infection.
... Unfortunately, these powerful approaches lack the experimental simplicity for routine analysis of numerous samples. Recent efforts concerning simple protein assays have been based on syn-chronous fluorescence spectroscopy [7], Rayleigh light scattering (RLS) 2 spectroscopy [8,9], and near-infrared spectroscopy [10,11]. Fluorescence assays [7] rely on the spectral response of an organic fluorescence tag chemically attached to nanoparticles. ...
... RLS methods are based on a similar principle but extract their information from synchronous spectra, that is, spectra recorded at 0-nm difference between excitation and emission wavelengths [8]. The near-infrared approach [10,11] takes advantage of vibrationally resolved spectra with fingerprint information for protein identification. Because infrared transitions provide inherently weak spectral bands, peak assignment for qualitative and quantitative purposes is made possible with chemometric approaches that improve signal/noise ratio and minimize spectral interference from sample concomitants. ...
We investigate the feasibility of using the luminescence response of polymerized liposomes incorporating ethylenediaminetetraacetate europium(III) (EDTA-Eu(3+)) for monitoring protein concentrations in aqueous media. Quantitative analysis is based on the linear relationship between the luminescence enhancement of the lanthanide ion and protein concentration. Analytical figures of merit are presented for carbonic anhydrase, human serum albumin, gamma-globulins, and thermolysin. Qualitative analysis is based on the luminescence lifetime of the liposome sensor. This parameter, which follows well-behaved single exponential decays and provides characteristic values for each of the four studied proteins, demonstrates the selective potential for protein identification. Then partial least squares-1 and artificial neural networks are compared toward the quantitative and qualitative analysis of human serum albumin and carbonic anhydrase in binary mixtures without previous separation at the concentration levels found in aqueous humor.
... As a result, an index of oxygenated/deoxygenated Hb can be obtained [83]. In addition to Hb, other biologically important molecules such as albumin [84][85][86], and cholesterol [87,88] can also be investigated using near-infrared light. ...
The majority of potentially preventable mortality in trauma patients is related to bleeding; therefore, early recognition and effective treatment of hemorrhagic shock impose a cardinal challenge for trauma teams worldwide. The reduction in mesenteric perfusion (MP) is among the first compensatory responses to blood loss; however, there is no adequate tool for splanchnic hemodynamic monitoring in emergency patient care. In this narrative review, (i) methods based on flowmetry, CT imaging, video microscopy (VM), measurement of laboratory markers, spectroscopy, and tissue capnometry were critically analyzed with respect to their accessibility, and applicability, sensitivity, and specificity. (ii) Then, we demonstrated that derangement of MP is a promising diagnostic indicator of blood loss. (iii) Finally, we discussed a new diagnostic method for the evaluation of hemorrhage based on exhaled methane (CH4) measurement. Conclusions: Monitoring the MP is a feasible option for the evaluation of blood loss. There are a wide range of experimentally used methodologies; however, due to their practical limitations, only a fraction of them could be integrated into routine emergency trauma care. According to our comprehensive review, breath analysis, including exhaled CH4 measurement, would provide the possibility for continuous, non-invasive monitoring of blood loss.
... Near-infrared spectroscopy (NIRS) is a rapid, reliable technique that can be used in real-time for MUN measurements directly at the dairy facility [19][20][21][22][23]. The study presented in this chapter examines the feasibility of this method for estimating milk urea nitrogen (MUN) concentration in whole milk. ...
In this chapter, a method for the determination of milk urea nitrogen (MUN) using short wavelength near-infrared (NIR) analysis of non-homogenized raw milk will be presented. NIR spectra of 183 non-homogenized milk samples from seven early lactation cows with different health statuses (healthy and diagnosed with mastitis) were measured at 40 °C, in the NIR region 700–1,100 nm. All cows were housed in the same conditions and fed the same diet. Partial least squares regression (PLSR) analysis was used for quantification of MUN in raw milk, based on the acquired NIR spectra with leave-one-animal-out cross-validation. Animals which contracted mastitis during the trials were identified, and the results for MUN prediction were separated from the results for healthy animals. The results of PLSR analysis showed excellent correlation between the values of MUN estimated based on the developed model and reference MUN values measured by the enzymatic method (R = 0.906). A standard error of 1.09 mg/dl, which is suitable for real-time MUN monitoring of raw milk, was achieved in the analysis of the milk of healthy animals. The NIRS method may provide a valuable tool to monitor urea nitrogen in routine measurements of milk for feeding management in dairy.
... Consequently, the results of many studies proved that human and bovine serum albumins play a crucial role as closely homologues proteins [16,17]. Several methods have been employed for determination of albumin, such as resonance light scattering (RLS), spectrofluorimetry [18][19][20], dye and metal complex-binding spectrophotometry [21,22], anodic stripping voltammetry [23], Lowry assay [24], Bradford assay [25], size-exclusion chromatography [26], flow injection chemiluminescence [27], total internal reflected resonance light scattering [28] and near-infrared spectroscopy [29]. However, one common disadvantage of above mentioned methods is the lack of high sensitivity and specificity (selectivity) [22,23,30]. ...
We report on a new sensitive method, for bovine serum albumin quantification, which is based on the use of methylene blue as a labelling agent combined with photothermal lens detection. In the presence of sodium dodecyl sulfate, methylene blue forms a dimer which can react with bovine serum albumin producing dedimerization. We found that, the photothermal signal decreases proportional to the concentration of bovine serum albumin added to the system composed by dodecyl sulfate and methylene blue. Therefore, the change of the signal intensity is linearly proportional to the concentration of bovine serum albumin.
Under the optimized analytical conditions, used in this work, the photothermal signal is linearly dependent on the concentration of bovine serum albumin in the range of 0.5 × 10⁻⁶–7.5 × 10⁻⁵ gmL⁻¹ with a regression coefficient R² = 0.9954. The relative standard deviation for BSA determination at 3.5 × 10⁻⁵ gmL⁻¹ (n = 5) is 2.3% and the achieved detection limit is 3.5 × 10⁻⁷ gmL⁻¹.
... Measured wavelength and calculated wavenumbers of the bands found with PCA, SIMCA, OPLS-DA and PLSR methods and their assignment based on the corresponding references. [44] OH, 1st overtone, H+(H 2 O) 10 [40] C–H stretching, sucrose [45] OH, 1st overtone, OH -stretching mode [ OH, 1st overtone, H 15 O 7 + H-bonded OH stretch [47] NH, 1st overtone, amid [54] NH/OH, 1st overtone, N–H/O–H stretching [55] 1492 6700 6700/2 = 3350 OH, 1st overtone, hydrogen-bonded (S 4 ) [ 34] OH, 1st overtone, H 15 O 7 + [ 43] OH, 1st overtone, strongly H-bonded [56] NH, 1st overtone, N–H stretching [57,58] NH, 1st overtone, NH 2 's asymmetric stretch [ [44] C−H vibration [60] CH/CH2 combination band [61] H–O–H/O–H bending and translation/rotation combinations [62] 1819 5500 5500/2 = 2750 1st overtone IHB stretch (OH-(H 2 O) 3 ) [ 63] combinationν(C−H) + ν(O−D)free [64] doi:10.1371/journal.pone.0130698.t002 water. ...
Development of efficient screening method coupled with cell functionality evaluation is highly needed in contemporary microbiology. The presented novel concept and fast non-destructive method brings in to play the water spectral pattern of the solution as a molecular fingerprint of the cell culture system. To elucidate the concept, NIR spectroscopy with Aqua-photomics were applied to monitor the growth of sixteen Lactobacillus bulgaricus one Lac-tobacillus pentosus and one Lactobacillus gasseri bacteria strains. Their growth rate, maximal optical density, low pH and bile tolerances were measured and further used as a reference data for analysis of the simultaneously acquired spectral data. The acquired spectral data in the region of 1100-1850nm was subjected to various multivariate data analyses – PCA, OPLS-DA, PLSR. The results showed high accuracy of bacteria strains classification according to their probiotic strength. Most informative spectral fingerprints covered the first overtone of water, emphasizing the relation of water molecular system to cell functionality.
... The effects of absorption from the visible to NIR wavelengths were modeled using experimental data from a number of reports. [4][5][6][7][8] Unfortunately, absorption data for individual plasma protein absorption dispersion could not be found. Instead total protein (albumin, globulin, and fibrinogen) absorption was used. ...
Blood tests are an essential tool in clinical medicine with the ability
diagnosis or monitor various diseases and conditions; however, the
complexities of these measurements currently restrict them to a
laboratory setting. P&P Optica has developed and currently produces
patented high performance spectrometers and is developing a
spectrometer-based system for rapid reagent-free blood analysis. An
important aspect of this analysis is the need to extract the analyte
specific information from the measured signal such that the analyte
concentrations can be determined. To this end, advanced chemometric
methods are currently being investigated and have been tested using
simulated spectra. A blood plasma model was used to generate Raman, near
infrared, and optical rotatory dispersion spectra with glucose as the
target analyte. The potential of combined chemometric techniques, where
multiple spectroscopy modalities are used in a single regression model
to improve the prediction ability was investigated using unfold partial
least squares and multiblock partial least squares. Results show
improvement in the predictions of glucose levels using the combined
methods and demonstrate potential for multiblock chemometrics in
spectroscopic blood analysis.
... protein solutions, biomolecular water solutions, suspension of small particles in water and various biological systems like cells, plants, animal body fluids and tissues have been analysed with vis-nIr spectroscopy. [36][37][38][39][40][41][42][43][44][45][46][47][48][49][50][51][52] In these studies, systematically, already described water absorbance bands were registered under various perturbations (concentrations of solutes, concentrations of nano particles of various sizes, concentrations of molecules which don't absorb light in the vis-nIr range, temperature, light illumination etc.). Such dynamic spectra were acquired and analysed with multivariate methods. ...
... Absorption in the near infrared shortwave region depends on the composition and subordinated vibrating waves like stretching and bending hydrogen containing groups such as -CH, -OH, -NH. Recently, important biological molecules such as albumin and glucose [8, 9, 10], cholesterol and lipid [11, 12], protein [13] and also some bacteria [14, 15] have been investigated quantitatively using infrared spectroscopy. Moreover this method has been used for diagnosis of some diseases such as cancer, Alzheimer, diabetes and migraine [16, 17, 18, 19, 20]. ...
Aims: This study was performed to investigate the possibility of diagnosis of HIV-1 human plasma infection and to differentiate the infected plasma from healthy ones using near infrared (NIR) spectroscopy along with Support Vector Machines (SVM) analysis as a rapid, cost-effective and non-destructive method. Materials & Methods: 35 HIV-1 infected plasmas and 15 healthy ones were selected and scanned within the range of 600-1100 nm using spectrophotometer device and their absorbance was obtained. The given results were then subjected to SVM for conducting the training phase and calibration process. After determination of training structure of SVM, absorbance data obtained from 21 HIV-1 infected and 20 uninfected plasmas at the same wavelength range were subjected to SVM as the unknown samples for test process and diagnosis of infection or the lack of it. Results: Among 21 infected unknown samples, 2 cases were diagnosed as uninfected and the rest were diagnosed as infected and from 20 uninfected unknown samples, only 1 case was diagnosed by SVM as infected. The sensitivity of the method was estimated 91% and its specificity was 95%. Conclusion: NIR-spectroscopy along with SVM analysis can be used as a pre-diagnostic method for detection of human plasma infection with HIV-1.
... The results were compared to those from visual assessment and electrical methods (18), and visualization of facial skin hydration after moisturizer application (19,20). However, changes in the OH band near 1460 nm in skin are complicated (16) because there are two protein bands nearing close proximity (1400-1500 nm) (21,22). Therefore, detection of slight changes in the water content is difficult using the 1460 nm OH band. ...
The water content of hair can be evaluated by weighing, the Karl Fischer method, and from electrical properties. However, these methods cannot be used to study the distribution of water in the hair. Imaging techniques are required for this purpose. In this study, a highly sensitive near-infrared (NIR) imaging system was developed for evaluating water in human hair. The results obtained from NIR imaging and conventional methods were compared.
An extended indium–gallium–arsenide NIR camera (detection range: 1100–2200 nm) and diffuse illumination unit developed in our laboratory were used to obtain a NIR image of hair. A water image was obtained using a 1950-nm interference filter and polarization filter. Changes in the hair water content with relative humidity (20–95% RH) and after immersion in a 7% (w/w) sorbitol solution were measured using the NIR camera and an insulation resistance tester. The changes in the water content after treatment with two types of commercially available shampoo were also measured using the NIR camera.
As the water content increased with changes in the relative humidity, the brightness of the water image decreased and the insulation resistance decreased. The brightness in the NIR image of hair treated with sorbitol solution was lower than that in the image of hair treated with water. This shows the sorbitol-treated hair contains more water than water-treated hair. The sorbitol-treated hair had a lower resistance after treatment than before, which also shows that sorbitol treatment increases the water content. With this system, we could detect a difference in the moisturizing effect between two commercially available shampoos.
The highly sensitive imaging system could be used to study water in human hair. Changes in the water content of hair depended on the relative humidity and treatment with moisturizer. The results obtained using the NIR imaging system were similar to those obtained using a conventional method. Our system could detect differences in the moisturizing effects of two commercially available shampoos.
... A near-infrared method combined with PLS regression was reported for determining serum proteins (albumin and IgG) in phosphate buffer solutions [54]. A fluorescence method applying multivariate calibration in real serum samples was also developed for determining the same proteins (see below). ...
The determination of the contents of therapeutic drugs, metabolites and other important biomedical analytes in biological samples is usually performed by using high-performance liquid chromatography (HPLC). Modern multivariate calibration methods constitute an attractive alternative, even when they are applied to intrinsically unselective spectroscopic or electrochemical signals. First-order (i.e., vectorized) data are conveniently analyzed with classical chemometric tools such as partial least-squares (PLS). Certain analytical problems require more sophisticated models, such as artificial neural networks (ANNs), which are especially able to cope with non-linearities in the data structure. Finally, models based on the acquisition and processing of second- or higher-order data (i.e., matrices or higher dimensional data arrays) present the phenomenon known as “second-order advantage”, which permits quantitation of calibrated analytes in the presence of interferents. The latter models show immense potentialities in the field of biomedical analysis. Pertinent literature examples are reviewed.
Advanced glycation end products (AGE) are produced by non-enzymatic reaction between glucose and biomolecules including proteins. AGE accumulation is known to cause alternations of structure and function in proteins and to be related with an increased risk of diabetic complications, cardiovascular diseases, and aging processes. Conventionally, AGE accumulation has been estimated by measuring auto fluorescence level using ultraviolet (UV) light excitation. In this study, we investigated an alternative approach to estimate auto fluorescence level and thus AGE accumulation in in vivo human skin using NIR (Near-Infrared) spectroscopy. To examine spectral features attributed to glycation in proteins, we first analyzed in vitro NIR spectra from native and glycated protein. Then, we further examined NIR spectra of in vivo skin from human subjects, and estimated their auto fluorescence level using several multivariate regression approaches. Our analysis in in vitro spectra from native and glycated albumin revealed that glycation may affect -CH and -NH stretching. Furthermore, we elucidated that those bands for -CH and -NH may be responsible for the variation in auto fluorescence level in human skin NIR spectra. Finally, auto fluorescence level was estimated from those NIR spectra using several multivariate regression methods: principal component regression (PCR), partial least square regression (PLS-R) and support vector regression (SVR). Among the three methods, SVR showed the best performance. We demonstrated in this study that NIR spectroscopy can be used as an alternative non-invasive method to estimate AGE accumulation in in vivo human skin tissue without UV radiation on skin tissue.
Aquaphotomics is a novel scientific discipline involving the study of water and aqueous systems. Using light-water interaction, it aims to extract information about the structure of water, composed of many different water molecular conformations using their absorbance bands. In aquaphotomics analysis, specific water structures (presented as water absorbance patterns) are related to their resulting functions in the aqueous systems studied, thereby building an aquaphotome—a database of water absorbance bands and patterns correlating specific water structures to their specific functions. Light-water interaction spectroscopic methods produce complex multidimensional spectral data, which require data processing and analysis to extract hidden information about the structure of water presented by its absorbance bands. The process of extracting information from water spectra in aquaphotomics requires a field–specific approach. It starts with an appropriate experimental design and execution to ensure high-quality spectral signals, followed by a multitude of spectral analysis, preprocessing and chemometrics methods to remove unwanted influences and extract water absorbance spectral pattern related to the perturbation of interest through the identification of activated water absorbance bands found among the common, consistently repeating and highly influential variables in all analytical models. The objective of this paper is to introduce the field of aquaphotomics and describe aquaphotomics multivariate analysis methodology developed during the last decade. Through a worked-out example of analysis of potassium chloride solutions supported by similar approaches from the existing aquaphotomics literature, the provided instruction should give enough information about aquaphotomics analysis i.e. to design and perform the experiment and data analysis as well as to represent water absorbance spectral pattern using various forms of aquagrams—specifically designed aquaphotomics graphs. The explained methodology is derived from analysis of near infrared spectral data of aqueous systems and will offer a useful and new tool for extracting data from informationally rich water spectra in any region. It is the hope of the authors that with this new tool at the disposal of scientists and chemometricians, pharmaceutical and biomedical spectroscopy will substantially progress beyond its state-of-the-art applications.
Nowadays a quick and inexpensive method, which allows rapid, in vivo comprehensive probiotic bacteria identification, is needed. To elucidate a new concept to evaluate probiotic bacteria, near infrared spectroscopy with aquaphotomics were applied to monitor the growth of eight Lactobacillus bulgaricus and one Lactobacillus gasseri bacteria strains. Their resistance to low pH (1.8) in the presence of pepsin and bile were measured and further used as reference data for analysis of the simultaneously acquired spectral data. The acquired spectral data in the region of 1100–1300 nm were subjected to various methods for multivariate data analyses—principal component analysis, linear discriminant analysis, soft independent modeling of class analogy, and partial least squares regression. The results showed high accuracy of bacteria strains classification according to their resistance and the potential of the tested wavelength region for rapid selection and prediction of some basic phenotypic characteristics of probiotic candidates. Results of the current study also revealed different suitability of each growth stage when using near infrared spectra for the classification of bacteria strains.
Near-infrared spectroscopy with interval and synergy interval partial least squares were used to determine total proteins, albumin, and globulin in human serum. Quantitative models of the relationships between the near-infrared spectra and these proteins were established. The root mean square error of prediction (RMSEP) was used to evaluate the models. The results demonstrate that synergy interval partial least squares was superior to interval partial least squares. The optimum wavelength ranges using synergy interval partial least squares for total proteins, albumin, and globulin in serum were 1668–1772 and 1772–1874 nm; 812–1000 and 1636–1800 nm; and 1228–1364 and 1642–1772 nm, respectively. The RMSEP of these proteins were 0.6905, 0.6686, and 0.5355 g/L. Waveband optimization was necessary to enhance the accuracy. The synergy interval partial least square model for total proteins, albumin, and globulin in serum offered higher accuracy and overcomes shortcomings of interval partial least squares. Furthermore, the optimized spectrum region may allow the design of near-infrared spectrophotometers for specific applications.
Near-infrared (NIR) absorbance spectra of glucose and human serum albumin (HSA) mixtures in phosphate-buffered solution (PBS) (pH=7.4) were investigated as the basis of reagentless blood constituent measurements. Diffuse reflectance spectra of powdered glucose and HSA were also studied. Differential absorbance spectra taken relative to the spectrum of a reference PBS were measured in the wavelength range of 750-2, 500 nm. In the differential spectra of mixture sample solutions, positive peaks corresponding to the HSA concentration could be observed, particularly in the range of 2,150-2,350 nm; however, peaks corresponding to the glucose concentration could not be observed. Some peaks visible in the diffuse reflectance spectra of powdered HSA could not be observed in the differential spectra. Measured NIR spectra were subjected to partial least-squares regression and good predictions of both HSA and glucose concentrations were obtained. Calibration for glucose in the concentration range of 0-300 mg/dl yielded a correlation coefficient of 0.9965 and a standard error of prediction (SEP) of 8.3 mg/dl with a latent variable of five. Calibration of HSA within the range of 2-4 g/dl yielded a correlation coefficient of 0.9996 with a SEP of 0.023 g/dl and a latent variable of five. These results suggest that reagentless measurements of glucose and HSA concentrations in human serum are possible with NIR spectroscopy.
A sensitive spectrofluorimetric method using constant-energy synchronous fluorescence technique is proposed for the determination of human albumin without separation. In this method, no reagent was used for enhancement of the fluorescence signal of albumin in the solution. Effects of some parameters, such as energy difference between excitation and emission monochromators (ΔE), emission and excitation slit widths and scan rate of wavelength were studied and the optimum conditions were established. For this purpose factorial design and response surface method were employed for optimization of the effective parameters on the fluorescence signal. The results showed that the scan rate of the wavelength has no significant effect on the analytical signal. The calibration curve was linear in the range 0.1–220.0 µg mL–1 of albumin with a detection limit of 7.0 × 10–3 µg mL–1. The relative standard deviations (RSD) for six replicate measurements of albumin were calculated as 2.2%, 1.7% and 1.3% for 0.5, 10.0 and 100.0 µg mL–1 albumin, respectively. Furthermore the proposed method has been employed for the determination of albumin in human serum and urine samples. Copyright © 2014 John Wiley & Sons, Ltd.
NIR offers multiple advantages for serum analysis, permitting a fast and direct determination of several parameters simultaneously, with low sample handling and without the need for reagents during the measurement step. The aim of this paper was to provide an evaluation of this technique in a real world scale, for the simultaneous determination of several parameters and based on a considerable number of samples. Direct near infrared (NIR) absorbance measurements were used to determine the concentration of clinical parameters in human serum that are required in routine biochemical tests. Total protein, albumin, total cholesterol, high-density lipoprotein (HDL cholesterol), low-density lipoprotein (LDL cholesterol), and very low-density lipoprotein (VLDL cholesterol), triglycerides, urea and glucose were determined in 447 serum samples obtained randomly from the clinical laboratory of the University Hospital Doctor Peset in Valencia (Spain). NIR spectra from 12 500 to 4000 cm(-1) obtained with a 1 mm optical path length were evaluated by using partial least squares regression models (PLS) built from the spectra of samples with known concentrations provided by the hospital. Root mean square error cross-validation (RMSECV) was used for selecting a number of factors, spectral regions and spectral preprocessing considered to build the models, that were evaluated from their prediction capability using the relative root mean square error of prediction (RRMSEP) of a series of around 30 independent samples, not used for calibration. For some analytes such as total protein, albumin, total cholesterol and triglycerides, errors obtained were 2.3, 4.4, 5.1, and 6.2% respectively, evidencing that the proposed methodology could compete with the enzymatic reference methodologies. However in the case of urea, glucose, HDL and LDL, average errors obtained were 16.0, 16.2, 18.0 and 11.0% respectively, and therefore the NIR methodology proposed is limited as a screening tool. With the use of a considerable number of samples for calibration, this study confirms that the proposed green and cost-effective methodology is ready for scaling up from the bench to the real world.
In order to illustrate the possibilities of principal component analysis in determining the number of components in a system of heavily overlapped spectra, several numerical spectral models were formed of bands with very close parameters. The models consisted of three bands, whose peak positions were locally shifted and noise added. For all the cases the relations between eigenvalues and principal component loadings were considered. It was shown that for those complex spectra for which peak positions and band halfwidths can be determined with high accuracy, eigenvalues criteria could easily indicate the number of components. For all analyzed models, the consideration of the shape of loadings was proven to have a high importance. In limited number of cases the shape of a loading can make the results of eigenvalue analysis more understandable. It has been shown that the noise can be treated as a main limitation in the application of the method to this type of the spectra.
Near-infrared diffuse reflectance spectroscopy (NIRDRS) has attracted more and more attention in analyzing the components in samples with complex matrices. However, to apply this technique to micro-analysis, there are still some obstacles to overcome such as the low sensitivity and spectral overlapping associated with this approach. A method for fast determination of bovine serum albumin (BSA) in micro-volume samples was studied using NIRDRS with sample spots and chemometric techniques. 10 μL of sample spotted on a filter paper substrate was used for the spectral measurements. Quantitative analysis was obtained by partial least squares (PLS) regression with signal processing and variable selection. The results show that the correlation coefficient (R) between the predicted and the reference concentration is 0.9897 and the recoveries are in the range of 87.4%–114.4% for the validation samples in the concentration range of 0.61–8.10 mg/mL. These results suggest that the method has the potential to quickly measure proteins in micro-volume solutions.
A weighted partial least squares (PLS) regression method for multivariate calibration of near infrared (NIR) spectra is proposed. In the method, the spectra are split into groups of variables according to the statistic values of variables, i.e., the stability, which has been used to evaluate the importance of variables in a calibration model. Because the stability reflects the relative importance of the variables for modeling, these groups present different spectral information for construction of PLS models. Therefore, if a weight which is proportional to the stability is assigned to each sub-model built with different group variables, a combined model can be built by a weighted combination of the sub-models. This method is different from the commonly used variable selection strategies, making full use of the variables according to their importance, instead of only the important ones. To validate the performance of the proposed method, it was applied to two different NIR spectral data sets. Results show that the proposed method can effectively utilize all variables in the spectra and enhance the prediction ability of the PLS model.
The feasibility of utilizing infrared spectroscopy for the prediction of haze formation in white wines resulting from heat and colloidal stability tests was investigated. One-hundred eleven white wines, representing multiple regions and varieties from the 2008 California vintage, were collected and analyzed. The near and mid-infrared spectra were measured and heat and colloidal (ethanol addition) stability tests were performed on the same wines. Partial-least squares regression analysis was then used to construct models predictive of the resulting nepholometric turbidity to the acquired spectra. Preliminary models obtained following application of spectral pretreatments today considered as “classical” (e.g., derivatives, standard normal variate, vector normalization, constant offset elimination) lacked robustness; two alternative algorithms designed to remove spectral information unrelated to the turbidity were then employed (orthogonal signal correction; direct orthogonal signal correction). While OSC pretreatment did not result in more robust models, DOSC considerably enhanced the goodness of the PLS model constructed to predict the ethanol test turbidity. Predictive modeling of the short-NIR spectra, following DOSC preprocessing, allowed the prediction of colloidal stability on an unknown test set with an R2=0.80 and a RMSEP=10.12 using three latent variables. When the data set was restricted to Chardonnay wines alone, the predictive ability improved, with R2=0.85 and RMSEP=8.90.
We present the analysis of H spectra obtained during a two-ribbon flare observed on the solar disk on May 16, 1981 at Yunnan Observatory with the Spectra- Spectroheliograph (SSHG). The complicated asymmetric pro- lesproducedbythepost-flareloopsoverlyingflareribbonsare analysed with a fast method based on the cloud model method. This method takes into account the bright background of the loopsandallowscomputationofthephysicalquantitiesofloops crossedbytheslit:velocityeld,opticalthickness,sourcefunc- tion and Doppler width. Using the scanning spectra of the re- gion, we obtain 2-D maps of these parameters. The validity of the method used is compared with other methods and the sensitivity of parameters to solution is discussed in details. The resultsobtainedwiththismethodarecomparedwiththosegiven by Heinzel et al. (1992) using a fully non-LTE approach.
Near infrared (NIR) spectra in the 10 000–4000 cm− 1 region were collected from control serum solutions mixed with cholesterol, glucose and urea solutions with various concentrations. The concentration ranges of cholesterol, glucose and urea were 46.08–395.39, 29.89–318.57 and 5.92–19.66 mg/dl, respectively, which cover the clinically important ranges. Moving window partial least squares regression (MWPLSR) method was applied to the NIR data to select informative regions for cholesterol, glucose and urea. Searching combination moving window partial least squares regression (SCMWPLSR) method was used to search for the optimized combinations of informative regions found by MWPLSR, and partial least squares calibration models were developed and compared for each spectral region proposed by MWPLSR and SCMWPLSR and whole region. The best PLS calibration model for each component was obtained from the spectral region optimized by SCMWPLSR. The best prediction results for cholesterol, glucose and urea have RMSEP of 6.68, 10.35 and 1.28 mg/dl, respectively. Of particular note is that the prediction result for urea is much better than that previously obtained by several research groups under similar conditions. SCMWPLSR can select the optimized combination spectral region for each blood component successfully within the complex blood matrix over the low concentration range.
Attenuated total reflectance mid-infrared spectra of serum and blood samples were obtained from 4,000 to 600 cm(-1). Models for the determination of albumin, immunoglobulin, total globulin, and albumin/globulin coefficients were established for serum samples, using reference data obtained by capillary electrophoresis. Based on the use of the amide bands I and II regions, the relative root mean square error of prediction (RRMSEP) was 4.9, 14.9, 4.5, and 7.1% for albumin, immunoglobulin, total globulin, and albumin/globulin coefficients, respectively, determined in an independent validation set of 120 samples using 200 samples for calibration. Additionally, the use of Kennard-Stone method for the selection of a representative calibration subset of samples provided comparable results using only 60 samples. For whole blood analysis, hemoglobin was determined in 40 validation samples using models built from 40 calibration independent samples with RRMSEP of 8.3, 5.5, and 4.9% with models built from direct spectra in the first case and from sample spectra recorded after lysis by sodium dodecyl sulfate and freezing, respectively, for the last two ones. The developed methodologies offer green alternatives for patient diagnosis in a few minutes, minimizing the use of reagents and residues and being adaptable for its use as a point-of-care method.
Near-infrared spectroscopy of human serum albumin (HSA) in the region of 750-2500 nm was subjected for developing reagentless measurement of HSA. Near-infrared diffuse reflectance absorption spectra of powder state HSA and differential absorbance from a reference phosphate buffer solution were measured for phosphate buffer solutions (pH=7.4) of HSA in the wavelength range of 750-2500nm. In the diffuse reflectance absorption spectra of powder state HSA, positive peaks can be observed near 1650-1750 nm and 2150-2350 nm. In the differential spectra of HSA solutions from the reference, peaks that follow HSA concentration were able to be clearly observed in the 1650-1750 nm and 2150-2350 nm region, and the peaks on 2150-2350 nm region was relatively greater. The near-infrared spectra band of the 750-1350nm, 1550-1850nm and 2052-2500nm regions were subjected to partial least squares (PLS) regression to obtain calibration models predicting the concentrations of the HSA. The calibration in the concentration of 0-5g/dl has yielded a correlation coefficient of 0.999 and a standard error of prediction of 0.0292 g/dl with latent variable of seven. These result suggests that the near-infrared differential spectra of HSA from the reference HSA solution and its calibration model constructed by using PLS regression can extract up effectively the information about albumin contained in phosphate buffer solution samples. These results are considered as a base of reagentless measurement of HSA in human serum and also in vivo HSA measurement.
Near-infrared differentialabsorbance spectra from pure water have been measured for phosphate buffer solutions (pH＝ 7.4) of human serum albumin (HSA) in the wavelength range of 750-2500 nm. The peaks that follow HSA concentration can be observed in near-infrared differential absorbance spectra ; in the 1650-1750 nm and 2150-2350 nm region. The near-infrared spectra in the 750-1350 nm, 1550-1850 nm and 2052-2500 nm regions of the solutions have been subjected to partial least squares (PLS) regression analysis to make calibration models predicting the concentrations of the HSA. The calibration in the concentration range of 0-5 g dlhas yielded a correlation coefficient of 0.999 and a standard error of prediction of 0.0283 g dl with latent variable of seven. This result suggests that the near-infrared differential spectra of HSA from pure water reference and its calibration model constructed by using PLS regression can pick up effectively the information about albumin contained in phosphate buffer solutions. This result can be a base of reagentless measurement of HSA in human serum and in vivo HSA measurement.
Near-infrared (NIR) spectra in the region of 5000–4000 cm−1 with a chemometric method called searching combination moving window partial least squares (SCMWPLS) were employed to determine the concentrations of human serum albumin (HSA), γ-globulin, and glucose contained in the control serum IIB (CS IIB) solutions with various concentrations. SCMWPLS is proposed to search for the optimized combinations of informative regions, which are spectral intervals, considered containing useful information for building partial least squares (PLS) models. The informative regions can easily be found by moving window partial least squares regression (MWPLSR) method. PLS calibration models using the regions obtained by SCMWPLS were developed for HSA, γ-globulin, and glucose. These models showed good prediction with the smallest root mean square error of predictions (RMSEP), the relatively small number of PLS factors, and the highest correlation coefficients among the results achieved by using whole region and MWPLSR methods. The RMSEP values of HSA, γ-globulin, and glucose yielded by SCMWPLS were 0.0303, 0.0327, and 0.0195 g/dl, respectively. These results prove that SCMWPLS can be successfully applied to determine simultaneously the concentrations of HSA, γ-globulin, and glucose in complicated biological fluids such as CS IIB solutions by using NIR spectroscopy.
Recently, near-infrared (NIR) imaging has been applied to detecting changes in skin hydration using the water OH band centered near 1460 nm. However, assigning changes in the intensity of the OH band near 1460 nm to changes in the skin's water content is complicated. Consequently, detection of small changes in facial skin water content is difficult. For highly sensitive imaging of facial skin water and oil, a near-infrared unit with a large detection range that includes the CH(3) and CH(2) stretching vibration modes at 1700-1800 nm and the strongest water bands centered near 1920 nm is required. In this study, an extended range indium gallium arsenide near-infrared camera was combined with a diffuse-illumination unit specifically developed for facial skin analysis. Images of water and oil in facial skin were obtained in real time using a combination of interference filters, such as 1950 ± 56 nm for water OH, 1775 ± 50 nm for oil CH, and 1300 ± 40 nm for background reflections. Clear near-infrared images were obtained with little mirror reflection. The water and oil content of facial skin could be evaluated even around the eyes, nose, and sides of the cheeks, which are areas that are difficult to analyze using current commercial devices. Differences were detected in the time-dependent changes of water and oil content in facial skin images obtained after the application of different types of moisturizer. The distribution of both water and oil in the facial skin could be visualized at the same time, and the images could be used to evaluate skin type and skin conditions.
The use of a combined spectral intensity and polarization signals optically scattered by tissue to determine analyte concentration in optically clear and turbid biological media was explored in a simulation study. Blood plasma was chosen as the biological model and glucose as the analyte of interest. The absorption spectrum and optical rotatory dispersion were modeled using experimental data and the Drude's equation, respectively, between 500 and 2000 nm. A polarization-sensitive Monte Carlo light-propagation model was used to simulate scattering media. Unfold partial least squares and multiblock partial least squares were used as regression methods to combine the spectral intensity and polarization signals, and to predict glucose concentrations in both clear and scattering models. The results show that the combined approaches produce better predictive results in both clear and scattering media than conventional partial least squares analysis, which uses intensity or polarization spectra independently. This improvement was somewhat diminished with the addition of scattering to the model, since the polarization signals were reduced due to multiple scattering. These findings demonstrate promise for the combined approach in clear or moderately scattering biological media; however, the method's applicability to highly scattering tissues is yet to be determined. The methodology also requires experimental validation.
A method is proposed for the simultaneous determination of albumin and immunoglobulin G (IgG1) with fluorescence spectroscopy and multivariate calibration with partial least squares regression (PLS). The influence of some instrumental parameters were investigated with two experimental designs comprising 19 and 11 experiments, respectively. The investigated parameters were excitation and emission slit, detection voltage and scan rate. When a suitable instrumental setting had been found, a minor calibration and test set were analysed and evaluated. Thereafter, a larger calibration of albumin and IgG1 was made out of 26 samples (0-42mugml(-1) albumin and 0-12.7mugml(-1) IgG1). This calibration was validated with a test set consisting of 14 samples in the same concentration range. The precision of the method was estimated by analysing two test set samples for six times each. The scan modes tested were emission scan and synchronous scan Delta60nm. The results showed that the method could be used for determination of albumin and IgG1 (albumin, root mean square error of prediction (RMSEP) <2, relative standard error of prediction (RSEP) <6% and IgG1, RMSEP <1, RSEP <8%) in spite of the overlapping fluorescence of the two compounds. The estimated precision was relative standard deviation (R.S.D.) <1.7%. The method was finally applied for the analysis of some sample fractions from an albumin standard used in affinity chromatography.
The water content of human nail plates was determined using a portable near-infrared (NIR) spectrometer with an InGaAs photodiode array detector. NIR diffuse reflectance (DR) spectra were collected from 108 cut nail plates with different relative humidity and in vivo from fingernails. Partial least-squares (PLS) regression was applied to the NIR spectra in the 1115-1645 nm region to develop calibration models that determine the water content in the cut nail plates and fingernails. A good correlation was obtained between the NIR spectra and the water content measured by nuclear magnetic resonance (NMR) for the NIR measurement of both cut nail plates and fingernails. The results indicate that the water content in the nails can be determined very rapidly (1 s) by means of the portable NIR spectrometer and PLS regression.
Near-infrared (NIR) spectra in the 12,000-4,000 cm(-1) region were measured for phosphate buffer solutions containing human serum albumin (HSA), gamma-globulin, and glucose with various concentrations at 37 degrees C. Five levels of full factorial design were used to prepare a sample set consisting of 125 samples of three component mixtures. The concentration ranges of HSA, gamma-globulin and glucose were 0.00-6.00 g dl(-1), 0.00-4.00 g dl(-1) and 0.00-2.00 g dl(-1), respectively. The 125 sample data were split into two sets, the calibration set with 95 data and the prediction set with 30 data. The most informative spectral ranges of 4648-4323, 4647-4255 and 4912-4304 cm(-1) were selected by moving window partial least-squares regression (MWPLSR) for HSA, [gamma]-globulin, and glucose in the mixtures, respectively. For HSA, the correlation coefficient (R) of 0.9998 and the root mean square error of prediction (RMSEP) of 0.0289 g dl(-1) were obtained. For [gamma]-globulin, R of 0.9997 and RMSEP of 0.0252 g dl(-1) were obtained. The corresponding statistic values of glucose were 0.9997 and 0.0156 g dl(-1), respectively. These statistical values obtained by MWPLSR are highly significant and better than those calculated by using the regions reported in the literature. The results presented here show that MWPLSR can select the informative regions with a simple procedure and increase the power of NIR spectroscopy for simultaneous determination of the concentrations of HSA, [gamma]-globulin and glucose in the mixture systems.
Two lanthanide complexes, namely 5-aminosalicylic acid ethylenediaminetetraacetate europium(III) (5As-EDTA-Eu3+) and 4-aminosalicylic acid ethylenediaminetetraacetate terbium(III), were evaluated for the analysis of carbonic anhydrase, human serum albumin (HSA), and gamma-globulin. Quantitative analysis is based on their luminescence enhancement upon protein binding and qualitative analysis on their lifetime capability to recognize the binding protein. Analytical figures of merit are presented for the three proteins. The limits of detection with 5As-EDTA-Eu3+ are at the parts per billion level. Partial least square regression analysis is used to determine HSA and gamma-globulin in binary mixtures without previous separation at the concentration ranges typically found in clinical tests of human blood serum.
The simultaneous determination of Cu(2+), Cd(2+) and Pb(2+) is demonstrated at four modified gold electrodes using N-PLS calibration. Three of the electrodes were modified with the peptides Gly-Gly-His, gamma-Glu-Cys Gly and human angiotensin I which were covalently attached to thioctic acid self-assembled monolayers and the fourth electrode was modified with thioctic acid only. Voltammetry at the modified electrodes in the presence of the three metal ions revealed one peak due to the reduction of copper and another due to the overlapping peaks of cadmium and lead which made quantification using conventional methods difficult. N-PLS was used to calibrate and predict trace concentrations (100 nM to 10 microM) of mixtures of Cu(2+), Cd(2+) and Pb(2+).
Noninvasive monitoring of blood/tissue glucose concentrations has been successfully accomplished in individual diabetic subjects by using near-infrared (NIR) spectroscopy coupled with chemometric methods. Three different spectrometer configurations were tested: a) a Fourier-transform infrared spectrometer with an indium antimonide detector; b) a grating monochromator equipped with a silicon (Si) array detector, without fiber optics; and c) a grating monochromator equipped with an Si detector, with fiber-optic sampling. NIR spectra were obtained from diabetic subjects by transmission through the finger during a meal-tolerance test. The maximum range of observed plasma glucose concentrations obtained from the blood samples was 2.5-27 mmol/L. The NIR spectra were processed by using the chemometric multivariate calibration methods of partial least squares and principal component regression. The best calibration yielded a cross-validated average absolute error in glucose concentration of 1.1 mmol/L. This predictive ability suggests that noninvasive glucose determinations by NIR/chemometrics is a viable analytical method.
Albumin is clearly an extraordinary molecule of manifold functions and applications. Although the exact function of albumin has been debated, much of the present data support the notion that the principal role of albumin in the circulatory system is to aid in the transport, metabolism, and distribution of exogenous and endogenous ligands. The ability of albumin to act as an important extracellular antioxidant (Halliwell, 1988) or impart protection from free radicals, and other harmful chemical agents (Emerson, 1989) agrees well with the increased susceptibility of analbuminemic rats to cancer (Kakizoe and Sugimura, 1988). The expression and delivery of albumin to the circulatory system by the liver therefore seem appropriate. An overview of the prolific ligand-binding properties of albumin is summarized in Fig. 21. The positions of known binding sites for important pharmaceutical markers such as diazepam, ibuprofen, aspirin, and warfarin are illustrated. In addition, the important endogenous markers tryptophan, octanoate, and bilirubin are also shown. With the exception of the definitive positions of the long-chain fatty acids, most albumin-ligand chemistry can now be explained by the atomic coordinates derived from crystal structures. Knowledge of the atomic structure coupled with the current applications of genetic engineering, such as site-directed mutagenesis, promises to provide an even greater understanding of albumin chemistry. It is widely accepted in the pharmaceutical industry that the overall distribution, metabolism, and efficacy of many drugs can be altered based on their affinity to serum albumin. In addition, many promising new drugs are rendered ineffective because of their unusually high affinity for this abundant protein. Obviously, an understanding of the chemistry of the various classes of pharmaceutical interactions with albumin can suggest new approaches to drug therapy and design, placing albumin in its rightful place as the 'second step in rational drug design.' Application of albumin in other therapeutic approaches is widely known. Some studies have suggested that modified serum albumin may be used as a selective contrast agent for tumor detection and/or therapy (Sinn et al., 1990). Other studies have demonstrated that albumin may be used to deliver toxic compounds for elimination of Mycobacterium tuberculosis via receptor-mediated drug delivery (Majumdar and Basu, 1991). Recently, chimeric albumin molecules such as HSA-CD4 (Yeh et al., 1992) and HSA-Cu,Zn-superoxide dismutase (Mao and Poznansky, 1989) have been utilized to increase the half-life and distribution, and reduce the immunogenicity, of these potential protein therapeutics. Albumin has now been cloned and expressed in several bacterial and fungal systems. The primary motivation for many of these studies has been the potential of recombinant albumin to serve as a serum replacement product that is free from unwanted viral contaminants, e.g., hepatitis, herpes, and human immunodeficiency virus (HIV). The most successful production has been achieved by extracellular expression in yeast (Etcheverry et al., 1986; Hinchcliffe and Kenney, 1986; Kalman et al., 1990; Okabayashi et al., 1986; Quirk, et al., 1989; Sijmons et al., 1990; Sleep et al., 1991). Clearly, future scientific and therapeutic applications of albumin appear limitless. In conclusion, albumin may be unique among proteins in that so many scientists have spent the largest portion of their professional careers studying very specific aspects of this protein. New appreciation for the complexity and potential applications presented by the structure of albumin promises to consume the careers of many more scientists.
In attempting to analyze, on digital computers, data from basically continuous physical experiments, numerical methods of performing familiar operations must be developed. The operations of differentiation and filtering are especially important both as an end in themselves, and as a prelude to further treatment of the data. Numerical counterparts of analog devices that perform these operations, such as RC filters, are often considered. However, the method of least squares may be used without additional computational complexity and with considerable improvement in the information obtained. The least squares calculations may be carried out in the computer by convolution of the data points with properly chosen sets of integers. These sets of integers and their normalizing factors are described and their use is illustrated in spectroscopic applications. The computer programs required are relatively simple. Two examples are presented as subroutines in the FORTRAN language.
Visible and near-infrared transmittance (T) spectra of unlysed blood samples were obtained with an NIRSystems Model 6500 spectrophotometer modified for an open cell and a vertical light path. Without rigid control of the pathlength and temperature, we were able to measure hemoglobin content within a standard error of 0.43 g/dL using a single-term second-derivative ratio of log(1/T) data at 1740 and 1346 nm. Calibration was done on a set of 104 samples (two spectra of blood from 52 patients) having hemoglobin levels of 6.1 to 19.2 g/dL. Validation was done on an independent set of 56 samples (two spectra of blood from 28 patients) having hemoglobin levels of 7.2 to 19.0 g/dL. The reproducibility of the measurement, tested by computing the coefficient of variability of the 28 duplicated results, was 0.63% (<0.1 g/dL). The results demonstrate a rapid simple diffuse transmittance measurement of hemoglobin in unlysed blood.
Noninvasive monitoring of glucose in diabetic patients is feasible with the use of near-infrared spectroscopic measurements. As a step toward the final goal of the development of a noninvasive monitor, the near-infrared spectra (4250 to 6600 cm−1) of glucose-doped whole blood samples were obtained along with reference glucose values. Glucose concentrations and spectra of blood samples obtained from four subjects were subjected to multivariate calibration with the use of partial least-squares (PLS) methods. The cross-validated PLS standard errors of prediction for glucose concentration based on data obtained from each individual subject's blood samples averaged 33 mg/dL over the range from 3 to 743 mg/dL. Cross-validated standard errors for glucose concentration from PLS calibrations based on data from all four subjects were 39 mg/dL. However, when PLS models based upon three subjects' data were used for prediction on the fourth, glucose prediction abilities were poor. It is suggested that blood chemistry differences were sufficiently different for the four subjects to require that a larger number of subjects be included in the calibration for adequate prediction abilities to be obtained from near-infrared spectra of blood from subjects not included in the calibration.
The present study aims to provide new insights into the temperature-dependent spectral variations in the near infrared (NIR) region of the spectrum of water by comparing chemometrics with spectroscopic analysis. Fourier transform (FT)-NIR spectra of water in the 9000–5500 cm ⁻¹ region have been measured over a temperature range of 5–85°C. The observed spectral changes have been analysed by both chemometrics, such as multilinear regression (MLR), principal component regression (PCR) and partial least squares (PLS) regression, and spectroscopic data analyses, such as second derivative, difference spectra and curve fitting. The second derivative of the NIR spectra of water suggests that an intense feature around 6900 cm ⁻¹ , due to the combination of antisymmetric and symmetric stretching modes of water, consists of at least five component spectra. Each component spectrum may be ascribed to the water species with no, one, two, three and four hydrogen bonds. Curve fitting has been performed for the 6900 cm ⁻¹ band and it has been found that the species with no hydrogen bonds increase largely with temperature, while those with more than two hydrogen bonds decrease. The temperature of water has been predicted by use of MLR, PCR and PLS regression. PCR and PLS regression loadings plots for Factor 1 of the models for the prediction of the temperature of water are almost identical with the difference spectrum of water between 5 and 85°C; both the loadings plots and the difference spectrum reflect strongly the changes in the hydrogen bonds of water. Loadings plots of Factor 1 of the PCR and PLS regression models are very similar to each other. It is very likely that since the temperature-dependent spectral variations of water in the NIR region are very regular, and the spectra have only very small noise and baseline changes, PCR and PLS regression select nearly identical factors.
In the present study we describe the relationship between laboratory values obtained with routinely used laboratory analytical methods and near infrared (NIR) spectral data of 126 whole blood and 228 blood serum samples. Spectra were measured with a SPECTRALYZER 1025 (PMC) computerised research analyser. The relationship among laboratory data and values of the second derivative of the log (1/ R) spectra measured at different wavelengths was determined by multiple linear regression (MLR) using three and four term linear summation equations, principal component regression (PCR) and partial least-squares (PLS) regression methods. Along with examples for qualitative detection of protein and lipid in human sera, as well as distinction of albumin and globulin dissolved in physiological saline solution, we describe mathematical models and evaluate their performance for the determination of protein and beta-lipoprotein (β-LP) content of serum as well as oxygen saturation and carbon dioxide pressure in whole blood. Validation of our results yielded a standard error of performance (SEP) of 2.47 g L ⁻¹ for protein content and 0.79 TU for β-LP content in blood serum, whereas SEP values of 5.41% for oxygen saturation and 5.27 mm Hg for carbon dioxide pressure in whole blood were found. Our results presented in this preliminary study indicate that NIR measurements can be related to analytical data of whole blood and serum. NIR spectroscopy is a rapid, accurate, cost effective method for determining quality parameters of whole blood and serum and might be a promising new tool in the field of automated clinical laboratory analysis.
Recent progress in spectroscopy and chemometrics have brought the reagentless analysis of blood substrates by near infrared spectroscopy into clinical reach. Results for the in-vitro analysis of several blood substrates in human blood plasma using multivariate calibration by partial-least squares are presented for 125 hospital samples. Whereas the relative meansquared prediction error for total protein (1.4 %) using short wave NIR data is comparable with previous results using conventional NIR spectroscopy, the errors found for total cholesterol (6.5 %) and triglycerides (13.8 %) are nearly a factor of two worse for this study.
We determined the human serum urea concentration from the near infrared spectrum of dried sera to assess whether the removal of water could provide an improvement over the direct analysis of unmodified samples. The results ( r = 0.989, SEP = 1.0 mmol L ⁻¹ ) obtained by the analysis of dried sera by using multiple linear least-squares regression were comparable to the results ( r = 0.993, SEP = 0.8 mmol L ⁻¹ ) obtained by the analysis of unmodified serum by using Partial Least-Squares regression. ¹ While the NIR determination of serum urea concentration in dried specimens simplifies the regression analysis it requires more technical competence and is more time-consuming.
Near-infrared (NIR) spectra of the human inner lip were obtained by using a special optimized accessory for diffuse reflectance measurements. The partial-least squares (PLS) multivariate calibration algorithm was applied for linear regression of the spectral data between 9000 and 5500 cm−1 (λ = 1.1-1.8 μm) against blood glucose concentrations determined by a standard clinical enzymatic method. Calibration experiments with a single person were carried out under varying conditions, as well as with a population of 133 different patients, with capillary and venous blood glucose concentration values provided. A genuine correlation between the blood glucose concentrations and the NIR-spectra can be proven. A time lag of about 10 min for the glucose concentration in the spectroscopically probed tissue volume vs. the capillary concentration can be estimated. Mean-square prediction errors obtained by cross-validation were in the range of 45 to 55 mg/dL. An analysis of different variance factors showed that the major contribution to the average prediction uncertainty was due to the reduced measurement reproducibility, i.e., variations in lip position and contact pressure. The results demonstrate the feasibility of using diffuse reflectance NIR-spectroscopy for the noninvasive measurement of blood glucose.
A digital Fourier filter is combined with partial least-squares (PLS) regression to generate a calibration model for glucose that is insensitive to sample temperature. This model is initially created by using spectra collected over the 5000 to 4000 cm-1 spectral range with samples maintained at 37°C. The analytical utility of the model is evaluated by judging the ability to determine glucose concentrations from a set of prediction spectra. Absorption spectra in this prediction set are obtained by ratioing single-beam spectra collected from solutions at temperatures ranging from 32 to 41°C to reference spectra collected at 37°C. The temperature sensitivity of the underlying water absorption bands creates large baseline variations in prediction spectra that are effectively eliminated by the Fourier filtering step. The best model provides a mean standard error of prediction across temperatures of 0.14 mM (2.52 mg/dL). The benefits of the Fourier filtering step are established, and critical experimental parameters, such as number of PLS factors, mean and standard deviation for the Gaussian shaped Fourier filter, and spectral range, are considered.
The multivariate calibration method of partial least-squares (PLS) was applied to the mid-infrared spectra of whole blood for quantitatively determining blood glucose concentrations. Separate calibration models were developed on the basis of spectra of whole blood obtained from six diabetic subjects from either in vitro glucose-supplemented blood or blood obtained from the same subjects in the post-prandial state during meal tolerance tests. The cross-validated PLS calibrations yielded average errors in glucose concentration of 11 and 13 mg/dL, respectively. It is desirable to use the calibration models based on the in vitro glucose-supplemented blood for determining glucose concentrations in unknown blood samples. However, when these multivariate calibration models based upon in vitro blood spectra were applied to the spectra of the post-prandial blood samples, a subject-dependent concentration bias was observed. The source of this bias was not identified, but when the glucose determinations were corrected for the bias, average concentration errors were found to be 14 mg/dL. Changes in spectrometer design or calibrations based on large numbers of subjects are expected to eliminate the presence of this bias. If these measures do not succeed in eliminating the bias, then methods are demonstrated that significantly reduce the bias while retaining the sensitive outlier detection capabilities of the PLS methods. These latter methods require that the infrared spectrum and reference glucose levels be obtained from a single blood sample from each subject.
Partial least-squares regression (PLS) and radial basis function (RBF) networks are used to compute calibration models for
non-invasive blood glucose determination by NIR diffuse reflectance spectroscopy. A model computation shows that even extremely
small deviations of the spectra induce increased prediction errors. Since the spectral contribution of blood glucose is much
smaller than deviations resulting from the non-invasive measuring process a method based on Pearson’s correlation coefficient
can be used for evaluating the quality of the recorded spectra during the prediction step. Another method is based on the
leverage values from the hat matrix of the RBF network. Both methods lead to a significant decrease in prediction error.
We have studied the standardization of total serum protein assay with the biuret reaction. Standard solutions were prepared from lyophilized preparations of human serum albumin and bovine serum albumin, with corrections made for volatile material and ash contents. These solutions and a solution of crystalline albumin standard were analyzed with a new stable biuret reagent, to establish absorptivity values (values for the absorbance of a 1 g/liter final reaction mixture). The mean values obtained were 0.302, 0.292, and 0.290 for human serum albumin, bovine serum albumin, and the crystalline albumin, respectively. We believe that the established absorptivity value will improve the accuracy of serum protein determinations. We studied the linearity of the relation between color produced and protein concentration, with use of the solutions described above and a serum pool. The color adheres to Beer's law up to the highest concentration tested: 3 g/liter for HSA and BSA, and 2.8 g/liter for serum in the final reaction mixture. The new biuret reagent has been stable for one year at room temperature. We recommend the use of bovine serum albumin as a primary standard for serum protein assays. It is inexpensive, easily available, and exhibits the best linearity in the biuret reaction.
The near-infrared (NIR) spectral region (700-2500 nm) is a fertile source of chemical information in the form of overtone and combination bands of the fundamental infrared absorptions and low-energy electronic transitions. This region was initially perceived as being too complex for interpretation and consequently was poorly utilized. Advances in chemometric techniques that can extract massive amounts of chemical information from the highly overlapped, complex spectra have led to extensive use of NIR spectrophotometry (NIRS) in the food, agriculture, pharmaceutical, chemical, and polymer industries. The application of NIRS in clinical laboratory measurements is still in its infancy. NIRS is a simple, quick, nondestructive technique capable of providing clinically relevant analyses of biological samples with precision and accuracy comparable with the method used to derive the NIRS models. Analyses can be performed with little or no sample preparation and no reagents. The success of NIRS in any particular case is determined by the complexity of the sample matrix, relative NIR absorptivities of the constituents, and the wavelengths and regression technique chosen. We describe the general approach to data acquisition, calibration, and analysis, using serum proteins, triglycerides, and glucose as examples.
A procedure is described for the measurement of clinically relevant concentrations of glucose in aqueous solutions with near-infrared (NIR) absorbance spectroscopy. A glucose band centered at 4400 cm-1 is used for this analysis. NIR spectra are collected over the frequency range 5000-4000 cm-1 with a Fourier transform spectrometer. A narrow-band-pass optical interference filter is placed in the optical path of the spectrometer to eliminate light outside this restricted range. This configuration provides a 2.9-fold reduction in spectral noise by utilizing the dynamic range of the detector solely for light transmitted through the filter. In addition, a novel spectral processing scheme is described for extracting glucose concentration information from the resulting absorbance spectra. The key component of this scheme is a digital Fourier filter that removes both high-frequency noise and low-frequency base-line variations from the spectra. Numerical optimization procedures are used to identify the best location and width of a Gaussian-shaped frequency response function for this Fourier filter. A dynamic area calculation, coupled with a simple linear base-line correction, provides an integrated area from the processed spectra that is linearly related to glucose concentrations over the range 1-20 mM. The linear calibration model accurately predicted glucose levels in a series of test solutions with an overall mean percent error of 2.5%. Based on the uncertainty in the parameters defining the calibration model and the variability of the magnitudes of the integrated areas, an overall uncertainty of 7.8% was estimated for predicted glucose concentrations.
This study describes the measurement of total protein in serum by near-infrared reflectance spectroscopy. With an algorithm, generated by the calibration procedure, the protein content of serum samples was calculated from absorbance data at various wavelengths in the near-infrared. A good correlation (r = 0.993) was found between near-infrared reflectance spectroscopy measurement of serum protein and analysis by the biuret reaction.
A new approach using near-infrared reflectance spectrometry for measuring serum cholesterol is reported. The method requires a calibration performed on 30 human sera samples containing cholesterol concentrations ranging from 3 to 12 mmol/L. A multiple linear regression makes it possible to select a filter combination that is characteristic of the matrix and its composition. The calibration constants calculated by the regression calculations are used to quantify serum cholesterol according to a mathematical equation. The method is reproducible and the results correlate very well with those obtained with the reference method (Liebermann-Burchard) giving a correlation coefficient of 0.963 and of 0.985 with the enzymatic method (Trinder assay). The present method provides excellent accuracy and gives a direct result without any reagent.
Research into noninvasive devices for self-monitoring of blood glucose is mainly based on near-infrared spectroscopy. Such a device is particularly desirable in the intensive therapy of patients with diabetes mellitus to achieve optimal metabolic control through frequent glucose testing. The state of noninvasive assay technology is presented. Using diffuse reflectance spectra of mucous lip tissue has advantages and drawbacks compared with tissue transmittance experiments. Different approaches have been proposed in the patent literature; however, current technology requires further significant improvements, particularly within the lower normal and hypoglycemic glucose concentration ranges.
Protocols are established for coupling digital filtering techniques with partial least-squares (PLS) regression for use in constructing multivariate calibration models from Fourier transform near-infrared absorbance spectra. Calibration models are developed to predict glucose concentrations in bovine plasma samples. Employing a calibration data set of 300 spectra collected from 55 plasma samples and 3 plasma lots, individual calibration models are developed based on four spectral ranges selected from the region 5000-4000 cm-1. A separate test set of 69 spectra collected from 14 plasma samples is used to evaluate the computed models. Gaussian-shaped bandpass digital filters are implemented by use of Fourier filtering techniques and employed to preprocess spectra to remove variation due to the background absorbance of the plasma matrix. PLS regression is used with the filtered spectra to compute calibration models for glucose. The optimization of the filter bandpass parameters is explored through the use of response surface methods. Through these optimization studies, calibration models are developed that achieve standard errors of estimate and standard errors of prediction in the range 0.4-0.5 mM across the concentration range of 2.5-25.5 mM. It is determined that the use of digital filtering as a preprocessing step significantly improves the performance of the resulting calibration models, minimizes the importance of spectral range in the calibration model development, and reduces the required number of PLS factors in each model.
A method is described for measuring clinically relevant levels of glucose in a protein matrix by near-infrared (near-IR) absorption spectroscopy. Results from an initial screening of major blood constituents identify protein as a major potential interference to the near-IR measurement of glucose in blood. The interference by protein is caused by relatively high concentrations coupled with strong near-IR absorption bands between 5000 and 4000 cm-1 (2.0-2.5 microns). Calibration models based on a simple univariate calibration procedure are not capable of providing accurate glucose concentrations from an independent set of prediction spectra. By use of the multivariate technique of partial least squares (PLS) regression, glucose concentrations can be determined with a 0.35 mM (6.3 mg/dL) standard error of prediction. The spectral range for this calibration model extends from 4600 to 4200 cm-1, and the optimum number of PLS factors is 14. In addition, calibration models based on a combination of digital Fourier filtering and PLS regression have been constructed and evaluated. Superior calibration models are obtained by using a preprocessing digital filtering step to remove spectral features not associated with glucose. The best overall calibration model was obtained by using a Gaussian-shaped Fourier filter defined by a mean position of 0.03f and standard deviation of 0.007f coupled with a 12-factor PLS regression computed over the spectral range from 4600 to 4200 cm-1. This model provided a standard error of prediction of 0.24 mM (4.3 mg/dL) for an independent set of prediction spectra.(ABSTRACT TRUNCATED AT 250 WORDS)
Selective calibration models are generated for glucose over the 1-20 nM concentration range by use of partial least-squares regression analysis of near-infrared spectra from 5000 to 4000 cm-1. Two spectral data sets are used to simulate triglyceride and protein variations in clinical samples. Triacetin is used in one data set to simulate variations in triglyceride levels, and bovine serum albumin (BSA) is used in the second data set to simulate variations in blood protein levels. Although these matrix components possess strong absorption bands that overlap and overshadow the absorption bands of glucose, successful calibration models can be generated with no evidence of prediction bias caused by the different levels of the matrix components. Furthermore, the benefits of using digital Fourier filtering as a preprocessing step are evaluated in terms of calibration performance. The resulting calibration models provide standard errors of prediction of 0.5 and 0.2 mM in triacetin and BSA matrices, respectively. Accurate glucose predictions are demonstrated from spectra that correspond to protein concentrations not present in the calibration data set. Lastly, digital Fourier filtering alone is shown to have only limited ability to isolate glucose signals from those of BSA and triacetin due to similarities in the widths of the absorption bands of the three species.
To determine the feasibility of near-infrared analysis for quantitating urea, creatinine, and protein in urine. Practical advantages of this method include ease of sample presentation and the absence of reagents or disposables.
The near-infrared methods were developed by first measuring the spectra of 123 different urine samples and, using independent clinical analyses, determining the protein, creatinine, and urea levels in each. Calibration models relating near-infrared spectroscopic features to those independently determined concentrations were optimized, and each model then validated using a set of 50 additional samples.
Standard errors of calibration were 14.4 mmol/L, 0.66 mmol/L, and 0.20 g/L, and standard errors of prediction 16.6 mmol/L, 0.79 mmol/L, and 0.23 g/L, respectively, for urea, creatinine, and protein.
Near-infrared urea quantitation is as accurate as the reference method, enzymatic (urease) conductivity, used here for calibration. Creatinine analysis is slightly less accurate relative to the reference (Jaffe rate) method; however, these errors can be minimized by careful attention to factors affecting precision. The accuracy of the near-infrared protein analysis cannot approach that of the reference method; nevertheless, the technique is potentially useful for coarse screening and for quantifying protein levels above 0.3 g/L.
Non-invasive instrumentation for metabolite monitoring can be based on near-infrared spectroscopy. The main research emphasis is on the development of glucose assays, which could be used for patient self-monitoring. Some histological and physiological variability during in-vivo measurements with integral tissue probing is discussed. The state of non-invasive monitoring is presented and compared with other competing methodologies. Different techniques have been proposed so far, which include transmission and diffuse reflectance measurements. Other methods employ photoacoustic laser spectroscopy or the measurement of changes in radiation scattering due to variations in glucose concentration. Currently available technology requires further improvements, in particular for the normal and hypoglycemic ranges, if it is to be used for devices for self- and bedside-monitoring or during surgery. The necessary selectivity for reliable glucose concentration prediction may only be gained by multivariate data analysis exploiting information from broad spectral ranges because of overlapping spectral features from several other biocomponents in the body tissue under investigation.
Lyophilized samples of two serum pools, prepared commercially, were sent to 200 Canadian clinical chemistry laboratories for estimations of total cholesterol, sodium, chloride, glucose, nonprotein or urea nitrogen, total protein, and inorganic phosphorus. The purpose of this evaluation study was to determine if these laboratories, as a group, were performing satisfactorily, and to help individual laboratories to evaluate their performances and detect gross errors in their results.
Values were reported from 170 laboratories. A summary of the complete data is presented in the form of tables, scatter diagrams, and frequency charts. It was found that there was a lack of accuracy and precision in many laboratories. Over 40% of the 3762 values reported fell outside of the allowable limits of errors and therefore were classified as unacceptable. These results indicated a need for improved performances in many of the participating laboratories.
Leaping Ahead with Near Infrared Spectroscopy
- J W Hall
- A Pollard
Leaping Ahead with Near Infrared Spectroscopy
- K H Norris
- J T Kuenstner
Near-Infrared Technology in Agricultual and Food Industries, Academic Association of Cereal Chemists
- P Williams
- K Norris