Influence of Temperature on Vibrational Spectra and Consequences for the Predictive Ability of Multivariate Models

Local Modeling by Adapting Source Calibration Models to Analyte Shifted Target Domain Samples Without Reference Values

Spectral multivariate calibration aims to derive models characterizing mathematical relationships between sample analyte amounts and corresponding spectral responses. These models are effective at predicting target domain sample analyte amounts when target samples are within the analyte and spectral calibration source domain. Models fail when target samples shift (analyte amounts and/or spectra) from the original calibration domain model. A total recalibration solution requires acquisition of new sample reference values and spectra. However, obtaining enough reference values to distinguish the target domain may be challenging or expensive. A simpler approach adapts the original model to the target domain using target sample spectra without analyte reference values (unlabeled). Analytical chemists have developed several machine learning algorithms using unlabeled regression domain adaptation processes. Unfortunately, prediction accuracy declines for these methods depending on how much the target domain analyte distribution has shifted from the calibration distribution, and regression transfer learning methods are instead needed. Regression domain adaptation and transfer learning are often referred to as model updating in analytical chemistry, but regression domain adaptation only applies to spectral shifts. The regression transfer learning method presented in this paper named null augmentation regression constant analyte (NARCA) leverages unlabeled repeat spectra of a single target sample to update an original calibration model to the shifted target domain sample. With sample repeat spectra, the analyte amount can be assumed constant or nearly constant for NARCA and because models are formed for one sample, NARCA operates as a local modeling method. The performance of NARCA as a regression transfer learning method is evaluated using five near-infrared data sets.

The Use of a Droplet Collar Accessory Attached to a Portable near Infrared Instrument to Identify Methanol Contamination in Whisky

The aim of this study was to evaluate the ability of a droplet collar accessory attached to a portable near-infrared (NIR) instrument to characterize the artificial contamination of methanol in commercial whisky samples. Unadulterated samples (n = 12) were purchased from local bottle shops where adulterated samples were created by adding methanol (99% pure methanol) at six levels (0.5%, 1%, 2%, 3%, 4% and 5% v/v) to the commercial whisky samples (controls). Samples were analyzed using a drop collar accessory attached to a MicroNIR Onsite instrument (900-1650 nm). Partial least squares (PLS) cross-validation statistics obtained for the prediction of all levels of methanol (from 0 to 5%) addition were considered adequate when the whole adulteration range was used, coefficient of determination in cross-validation (R2cv: 0.95) and standard error in cross of validation (SECV: 0.35% v/v). The cross-validation statistics were R2cv: 0.97, SECV: 0.28% v/v after the 0.5% and 1% v/v methanol addition was removed. These results showed the ability of using a new sample presentation attachment to a portable NIR instrument to analyze the adulteration of whisky with methanol. However, the low levels of methanol adulteration (0.5 and 1%) were not well predicted using the NIR method evaluated.

In-Field Implementation of Near-Infrared Quantitative Methods for Analysis of Medicines in Tropical Environments

Near-infrared (NIR) spectroscopy is actually a well-established technique that demonstrates its performance in the frame of detection of poor-quality medicines. The use of low-cost handheld NIR spectrophotometers in low-resource contexts can allow an inexpensive and more rapid detection compared to laboratory methods. Considering these points, it was decided to develop, validate, and transfer methods for the quantification of ciprofloxacin and metronidazole tablet samples using a NIR handheld spectrophotometer in transmission mode (NIR-M-T1) coupled to chemometrics such as partial least squares regression (PLSR) algorithm. All of the models were validated with the total error approach using an accuracy profile as a decision tool, with ±10% specifications and a risk α set at 5%. Quantitative PLSR models were first validated in Belgium, which is a temperate oceanic climate zone. Second, they were transferred to Cameroon, a tropical climate zone, where issues regarding the prediction of new validation series with the initial models were highlighted. Two augmentation strategies were then envisaged to make the predictive models robust to environmental conditions, incorporating the potential variability linked to environmental effects in the initial calibration sets. The resulting models were then used for in-field analysis of ciprofloxacin and metronidazole tablet samples collected in three cities in Cameroon. The contents results obtained for each sample with the two strategies were close and not statistically different. Nevertheless, the first one is easier to implement and the second is the best regarding model diagnostic measures and accuracy profiles. Two samples were found to be noncompliant in terms of content, and these results were confirmed using high-performance liquid chromatography taken as the reference method.

Physicochemical Responsive Integrated Similarity Measure (PRISM) for a Comprehensive Quantitative Perspective of Sample Similarity Dynamically Assessed with NIR Spectra

Determining sample similarity underlies many foundational principles in analytical chemistry. For example, calibration models are unsuitable to predict outliers. Calibration transfer methods assume a moderate degree of sample and measurement dissimilarities between a calibration set and target prediction samples. Classification approaches link target sample similarities to groups of similar class samples. Although similarity is ubiquitous in analytical chemistry and everyday life, quantifying sample similarity is without a straightforward solution, especially when target domain samples are unlabeled and the only known features are measurable, such as spectra (the focus of this paper). The process proposed to assess sample similarity integrates spectral similarity information with contextual considerations among source analyte contents, model, and analyte predictions. This hybrid approach named the physicochemical responsive integrated similarity measure (PRISM) amplifies hidden-but-essential physicochemical properties encoded within respective spectra. PRISM is tested on four near-infrared (NIR) data sets for four diverse application areas to show efficacy. These applications are the assessment of prediction reliability and model updating for model generalizability, outlier detection, and basic matrix matching evaluation. Discussion is provided on adapting PRISM to classification problems. Results indicate that PRISM collects large amounts of similarity information and effectively integrates it to produce a quantitative similarity evaluation between the target sample and a source domain. The approach is also useful for biological samples with additional physiochemical variations. While PRISM is dynamically tested on NIR data, parts of PRISM were previously applied to other data types, and PRISM should be applicable to other measurement systems perturbed by matrix effects.

SmartSpectrometer-Embedded Optical Spectroscopy for Applications in Agriculture and Industry

The ongoing digitization of industry and agriculture can benefit significantly from optical spectroscopy. In many cases, optical spectroscopy enables the estimation of properties such as substance concentrations and compositions. Spectral data can be acquired and evaluated in real time, and the results can be integrated directly into process and automation units, saving resources and costs. Multivariate data analysis is needed to integrate optical spectrometers as sensors. Therefore, a spectrometer with integrated artificial intelligence (AI) called SmartSpectrometer and its interface is presented. The advantages of the SmartSpectrometer are exemplified by its integration into a harvesting vehicle, where quality is determined by predicting sugar and acid in grapes in the field.

Unveiling non-linear water effects in near infrared spectroscopy: A study on organic wastes during drying using chemometrics

In the context of organic waste management, near infrared spectroscopy (NIRS) is being used to offer a fast, non-destructive, and cost-effective characterization system. However, cumbersome freeze-drying steps of the samples are required to avoid water's interference on near infrared spectra. In order to better understand these effects, spectral variations induced by dry matter content variations were obtained for a wide variety of organic substrates. This was made possible by the development of a customized near infrared acquisition system with dynamic highly-resolved simultaneous scanning of near infrared spectra and estimation of dry matter content during a drying process at ambient temperature. Using principal components analysis, the complex water effects on near infrared spectra are detailed. Water effects are shown to be a combination of both physical and chemical effects, and depend on both the characteristics of the samples (biochemical type and physical structure) and the moisture content level. This results in a non-linear relationship between the measured signal and the analytical characteristic of interest. A typology of substrates with respect to these water effects is provided and could further be efficiently used as a basis for the development of local quantitative calibration models and correction methods accounting for these water effects.

Feature selection using distributions of orthogonal PLS regression vectors in spectral data

Feature selection, which is important for successful analysis of chemometric data, aims to produce parsimonious and predictive models. Partial least squares (PLS) regression is one of the main methods in chemometrics for analyzing multivariate data with input X and response Y by modeling the covariance structure in the X and Y spaces. Recently, orthogonal projections to latent structures (OPLS) has been widely used in processing multivariate data because OPLS improves the interpretability of PLS models by removing systematic variation in the X space not correlated to Y. The purpose of this paper is to present a feature selection method of multivariate data through orthogonal PLS regression (OPLSR), which combines orthogonal signal correction with PLS. The presented method generates empirical distributions of features effects upon Y in OPLSR vectors via permutation tests and examines the significance of the effects of the input features on Y. We show the performance of the proposed method using a simulation study in which a three-layer network structure exists in compared with the false discovery rate method. To demonstrate this method, we apply it to both real-life NIR spectra data and mass spectrometry data.

Compensation of temperature effects on spectra through evolutionary rank analysis

Spectra measured in various ranges of temperature are usually slightly different from each other in shape and position of the bands. Although the displayed inconsistencies are rather small, yet may lead to incorrect analysis and interpretation of the collected spectrothermal data. Thus the unspecific spectral effects induced by temperature, in particular the thermal shifts and broadening of the bands, have to be compensated. In the paper, a simple two-step method of thermospectral dataset uniformisation is presented. Thermally induced 'movement' of the bands is approximated as a linear function of the difference of temperatures, so the co-shifting of the spectra is done linearly. Thermal broadening is mimicked by convoluting the low-temperature signal (spectrum) with a Gaussian or Lorentzian spreading filter. Proper widths (values of FWHM) of these filters, used to uniform the whole dataset, are assumed to depend on the difference of temperatures, in a form of one-parameter functions. This assumption, which has been empirically confirmed, is a fundamental premise of the method of Partial Compensation for Thermal Broadening (PCTB). Optimal values of the parameters of all the functions, used to compensate both thermal shifting and broadening, are found by the Evolutionary Rank Analysis (ERA) applied on an evolving data matrix. Efficiency of the proposed approach was verified on the UV-Vis thermospectral dataset of one-component model systems. In addition, since the method is aimed at making uniformed the thermospectral datasets of multi-component systems with similar spectral properties of individual components, the two-component conformer system of t-APE (trans-1-(2'-anthryl)-2-phenylethene) has also been analyzed.

Multilevel LASSO-based NIR temperature-correction modeling for viscosity measurement of bisphenol-A

Temperature variation will affect the prediction accuracy when using the near-infrared (NIR) analytical model to measure the viscosity of bisphenol-A. In order to correct the effect of temperature on the prediction performance of NIR model, a multilevel least-absolute shrinkage and selection operator (LASSO) is proposed in this paper. The multilevel LASSO algorithm combines LASSO with the multilevel simultaneous component analysis (MLSCA) to enhance the robustness of the model under temperature changes and external disturbances. MLSCA is applied to decompose the molecular spectral data into two parts. One part denotes the property caused by temperature, the other means the changes of concentration. LASSO, a sparse regression model, is used to select the variables and perform the regularization to further enhance the robustness and interpretability of the model. Experimental results demonstrate the effectiveness of the proposed model in measuring bisphenol-A viscosity, which provides a more stable prediction result compared with the existing ones without temperature corrections.

Nondestructive measurement of pectin polysaccharides using hyperspectral imaging in mulberry fruit

Pectin polysaccharide is an important phytochemical with potential biomedical applications. It is commonly measured by time-consuming destructive chemical methods. This work demonstrates the feasibility of using visible and near-infrared hyperspectral imaging (HSI) techniques to rapidly measure pectin polysaccharides in intact mulberry fruits. Based on spatial information provided by HSI images, the representative spectrum of each whole mulberry was accurately extracted without background. The effects of storage temperature on two varieties of mulberries for model establishment were studied. The performances of two spectral ranges obtained by Si and InGaAs CCD detectors for pectin prediction were compared. The best predictions were obtained from dilute alkali soluble pectin and total soluble pectin in Dashi mulberry fruit stored at room temperature, with residual predictive deviation values of 2.317 and 1.935, respectively. Our results show that HSI is a promising alternative to the chemical method to rapidly and nondestructively measure the pectin content.

Non-Destructive Soluble Solids Content Determination for 'Rocha' Pear Based on VIS-SWNIR Spectroscopy under 'Real World' Sorting Facility Conditions

In this paper we report a method to determine the soluble solids content (SSC) of 'Rocha' pear (Pyrus communis L. cv. Rocha) based on their short-wave NIR reflectance spectra (500-1100 nm) measured in conditions similar to those found in packinghouse fruit sorting facilities. We obtained 3300 reflectance spectra from pears acquired from different lots, producers and with diverse storage times and ripening stages. The macroscopic properties of the pears, such as size, temperature and SSC were measured under controlled laboratory conditions. For the spectral analysis, we implemented a computational pipeline that incorporates multiple pre-processing techniques including a feature selection procedure, various multivariate regression models and three different validation strategies. This benchmark allowed us to find the best model/preproccesing procedure for SSC prediction from our data. From the several calibration models tested, we have found that Support Vector Machines provides the best predictions metrics with an RMSEP of around 0.82 ∘ Brix and 1.09 ∘ Brix for internal and external validation strategies respectively. The latter validation was implemented to assess the prediction accuracy of this calibration method under more 'real world-like' conditions. We also show that incorporating information about the fruit temperature and size to the calibration models improves SSC predictability. Our results indicate that the methodology presented here could be implemented in existing packinghouse facilities for single fruit SSC characterization.

Temperature correction strategy for improving concentration predictions with near-infrared spectra of aqueous-based samples

Concentration predictions from near-infrared spectra are used across a range of application areas. When aqueous samples are employed, the extreme temperature sensitivity of underlying water absorption bands can lead to significant errors in predicted analyte concentrations, even when efforts are made to control sample temperatures. To address this issue, a temperature-correction procedure was developed on the basis of modeling the systematic error that occurs in predicted concentrations as a function of variation in sample temperature. With this approach, a quantitative calibration model was developed for samples at a fixed temperature. This model was subsequently applied to the spectra of a second set of samples with known analyte concentrations collected under conditions of varying temperature. Using either measured temperatures or those estimated from a spectral temperature prediction model, a least-squares polynomial fit was performed between concentration residuals and temperature. Going forward, for a given sample temperature, the polynomial model was used to estimate the concentration residual at that temperature. The estimated residual was then used to correct the predicted concentration. For spectra collected in the 5000-4000 cm^-1 near-infrared region, this methodology was tested for samples of glucose in buffer and mixture samples of glucose and lactate in buffer over the temperature range of 20.0-40.5 °C.

Assessing the influence of temperature on NIRS prediction models for the determination of sodium content in dry-cured ham slices

Temperature fluctuations are a key factor in the development of prediction models using near infrared spectroscopy (NIRS). In the present study, this influence has been investigated and a methodology has been proposed to reduce the effect of sample temperature on NIRS model prediction of the sodium content in dry-cured ham slices. Spectra were taken directly from the slices using a remote measurement probe (for non-contact analysis) at three different temperature ranges: -12 °C to -5°C, -5°C to 10 °C and 10 °C to 20 °C. Local and global temperature compensation methods were established. Partial-least squares (PLS) regression was used as a chemometrics tool to perform the calibrations. The results showed that local models were sensitive to changes in temperature, while a global temperature model using sample spectra over the entire temperature range showed good prediction ability, reducing the error caused by temperature fluctuations to acceptable levels for practical applications.

Monitoring Process Water Quality Using Near Infrared Spectroscopy and Partial Least Squares Regression with Prediction Uncertainty Estimation

Reuse of process water in dairy ingredient production-and food processing in general-opens the possibility for sustainable water regimes. Membrane filtration processes are an attractive source of process water recovery since the technology is already utilized in the dairy industry and its use is expected to grow considerably. At Arla Foods Ingredients (AFI), permeate from a reverse osmosis polisher filtration unit is sought to be reused as process water, replacing the intake of potable water. However, as for all dairy and food producers, the process water quality must be monitored continuously to ensure food safety. In the present investigation we found urea to be the main organic compound, which potentially could represent a microbiological risk. Near infrared spectroscopy (NIRS) in combination with multivariate modeling has a long-standing reputation as a real-time measurement technology in quality assurance. Urea was quantified Using NIRS and partial least squares regression (PLS) in the concentration range 50-200 ppm (RMSEP = 12 ppm, R²= 0.88) in laboratory settings with potential for on-line application. A drawback of using NIRS together with PLS is that uncertainty estimates are seldom reported but essential to establishing real-time risk assessment. In a multivariate regression setting, sample-specific prediction errors are needed, which complicates the uncertainty estimation. We give a straightforward strategy for implementing an already developed, but seldom used, method for estimating sample-specific prediction uncertainty. We also suggest an improvement. Comparing independent reference analyses with the sample-specific prediction error estimates showed that the method worked on industrial samples when the model was appropriate and unbiased, and was simple to implement.

Quantitative determination based on the differences between spectra-temperature relationships

In the Near-infrared (NIR) spectral measurement it is not always possible to keep the experimental conditions constant. The fluctuations in external variables, such as temperature, will result in a nonlinear shift and a broadening of the spectral bands. In this study, the temperature-induced spectral variation coefficient (TSVC) was obtained by using loading space standardization (LSS). The relationship between TSVC and normalized squared temperature was quantitatively analyzed and applied to the quantitative determination of the compositions in mixtures. NIR spectra of peanut-soy-corn oil mixtures measured at seven temperatures were analyzed. It was found that, the relationship between TSVC and normalized squared temperature can be established by using LSS. Furthermore, the quantitative determination of the compositions in a mixture can be achieved by using the difference between the relationships, i.e., the slope of the relationship. The calibration curves between slope and composition volume are found to be reliable with the correlation coefficients (R(2)) as high as 0.9992. Quantitative determination by the calibration curves were also validated. Therefore, the method can be an effective tool for investigating the effect of temperature and quantitatively analysis.

Chemometric optimization of the robustness of the near infrared spectroscopic method in wheat quality control

A chemometric approach was applied for the optimization of the robustness of the NIRS method for wheat quality control. Due to the high number of experimental (n=6) and response variables to be studied (n=7) the optimization experiment was divided into two stages: screening stage in order to evaluate which of the considered variables were significant, and optimization stage to optimize the identified factors in the previously selected experimental domain. The significant variables were identified by using fractional factorial experimental design, whilst Box-Wilson rotatable central composite design (CCRD) was run to obtain the optimal values for the significant variables. The measured responses included: moisture, protein and wet gluten content, Zeleny sedimentation value and deformation energy. In order to achieve the minimal variation in responses, the optimal factor settings were found by minimizing the propagation of error (POE). The simultaneous optimization of factors was conducted by desirability function. The highest desirability of 87.63% was accomplished by setting up experimental conditions as follows: 19.9°C for sample temperature, 19.3°C for ambient temperature and 240V for instrument voltage.

A Process Analytical Technology (PAT) approach to control a new API manufacturing process: development, validation and implementation

Pharmaceutical companies are progressively adopting and introducing Process Analytical Technology (PAT) and Quality-by-Design (QbD) concepts promoted by the regulatory agencies, aiming the building of the quality directly into the product by combining thorough scientific understanding and quality risk management. An analytical method based on near infrared (NIR) spectroscopy was developed as a PAT tool to control on-line an API (active pharmaceutical ingredient) manufacturing crystallization step during which the API and residual solvent contents need to be precisely determined to reach the predefined seeding point. An original methodology based on the QbD principles was designed to conduct the development and validation of the NIR method and to ensure that it is fitted for its intended use. On this basis, Partial least squares (PLS) models were developed and optimized using chemometrics methods. The method was fully validated according to the ICH Q2(R1) guideline and using the accuracy profile approach. The dosing ranges were evaluated to 9.0-12.0% w/w for the API and 0.18-1.50% w/w for the residual methanol. As by nature the variability of the sampling method and the reference method are included in the variability obtained for the NIR method during the validation phase, a real-time process monitoring exercise was performed to prove its fit for purpose. The implementation of this in-process control (IPC) method on the industrial plant from the launch of the new API synthesis process will enable automatic control of the final crystallization step in order to ensure a predefined quality level of the API. In addition, several valuable benefits are expected including reduction of the process time, suppression of a rather difficult sampling and tedious off-line analyses.

Development and comparative assessment of Raman spectroscopic classification algorithms for lesion discrimination in stereotactic breast biopsies with microcalcifications

Microcalcifications are an early mammographic sign of breast cancer and a target for stereotactic breast needle biopsy. Here, we develop and compare different approaches for developing Raman classification algorithms to diagnose invasive and in situ breast cancer, fibrocystic change and fibroadenoma that can be associated with microcalcifications. In this study, Raman spectra were acquired from tissue cores obtained from fresh breast biopsies and analyzed using a constituent-based breast model. Diagnostic algorithms based on the breast model fit coefficients were devised using logistic regression, C4.5 decision tree classification, k-nearest neighbor (k -NN) and support vector machine (SVM) analysis, and subjected to leave-one-out cross validation. The best performing algorithm was based on SVM analysis (with radial basis function), which yielded a positive predictive value of 100% and negative predictive value of 96% for cancer diagnosis. Importantly, these results demonstrate that Raman spectroscopy provides adequate diagnostic information for lesion discrimination even in the presence of microcalcifications, which to the best of our knowledge has not been previously reported.

Short-wave near-infrared spectrometer for alcohol determination and temperature correction

A multichannel short-wave near-infrared (SW-NIR) spectrometer module based on charge-coupled device (CCD) detection was designed. The design relied on a tungsten lamp enhanced by light emitting diodes, a fixed grating monochromator and a linear CCD array. The main advantages were high optical resolution and an optimized signal-to-noise ratio (0.24 nm and 500, resp.) in the whole wavelength range of 650 to 1100 nm. An application to alcohol determination using partial least squares calibration and the temperature correction was presented. It was found that the direct transfer method had significant systematic prediction errors due to temperature effect. Generalized least squares weighting (GLSW) method was utilized for temperature correction. After recalibration, the RMSEP found for the 25°C model was 0.53% v/v and errors of the same order of magnitude were obtained at other temperatures (15, 35 and 40°C). And an r(2) better than 0.99 was achieved for each validation set. The possibility and accuracy of using the miniature SW-NIR spectrometer and GLSW transfer calibration method for alcohol determination at different temperatures were proven. And the analysis procedure was simple and fast, allowing a strict control of alcohol content in the wine industry.

Simultaneous polarographic determination of 2-nitrophenol and 4-nitrophenol by differential pulse polarography method and support vector regression

A differential pulse polarography (DPP) for the simultaneous determination of 2-nitrophenol and 4-nitrophenol was proposed. It was found that under optimum experimental conditions (pH = 5, scan rate = 5 mV/s, pulse amplitude = -50 mV), 2-nitrophenol and 4-nitrophenol had well-defined polarographic reduction waves with peak potentials at -317 and -406 mV, respectively. In the mixture of two compounds overlapping polarographic peaks were observed. In this study, support vector regression (SVR) was applied to resolve the overlapped polarograms. Furthermore, a comparison was made between the performance of SVR and partial least square (PLS) on data set. The results demonstrated that SVR is a better well-performing alternative for the analysis and modeling of DPP data than the commonly applied PLS technique. The proposed method was used for the determination of 2-nitrophenol and 4-nitrophenol in industrial waste water.

Wavelength selection-based nonlinear calibration for transcutaneous blood glucose sensing using Raman spectroscopy

While Raman spectroscopy provides a powerful tool for noninvasive and real time diagnostics of biological samples, its translation to the clinical setting has been impeded by the lack of robustness of spectroscopic calibration models and the size and cumbersome nature of conventional laboratory Raman systems. Linear multivariate calibration models employing full spectrum analysis are often misled by spurious correlations, such as system drift and covariations among constituents. In addition, such calibration schemes are prone to overfitting, especially in the presence of external interferences that may create nonlinearities in the spectra-concentration relationship. To address both of these issues we incorporate residue error plot-based wavelength selection and nonlinear support vector regression (SVR). Wavelength selection is used to eliminate uninformative regions of the spectrum, while SVR is used to model the curved effects such as those created by tissue turbidity and temperature fluctuations. Using glucose detection in tissue phantoms as a representative example, we show that even a substantial reduction in the number of wavelengths analyzed using SVR lead to calibration models of equivalent prediction accuracy as linear full spectrum analysis. Further, with clinical datasets obtained from human subject studies, we also demonstrate the prospective applicability of the selected wavelength subsets without sacrificing prediction accuracy, which has extensive implications for calibration maintenance and transfer. Additionally, such wavelength selection could substantially reduce the collection time of serial Raman acquisition systems. Given the reduced footprint of serial Raman systems in relation to conventional dispersive Raman spectrometers, we anticipate that the incorporation of wavelength selection in such hardware designs will enhance the possibility of miniaturized clinical systems for disease diagnosis in the near future.

Investigation of the specificity of Raman spectroscopy in non-invasive blood glucose measurements

Although several in vivo blood glucose measurement studies have been performed by different research groups using near-infrared (NIR) absorption and Raman spectroscopic techniques, prospective prediction has proven to be a challenging problem. An important issue in this case is the demonstration of causality of glucose concentration to the spectral information, especially as the intrinsic glucose signal is smaller compared with that of the other analytes in the blood-tissue matrix. Furthermore, time-dependent physiological processes make the relation between glucose concentration and spectral data more complex. In this article, chance correlations in Raman spectroscopy-based calibration model for glucose measurements are investigated for both in vitro (physical tissue models) and in vivo (animal model and human subject) cases. Different spurious glucose concentration profiles are assigned to the Raman spectra acquired from physical tissue models, where the glucose concentration is intentionally held constant. Analogous concentration profiles, in addition to the true concentration profile, are also assigned to the datasets acquired from an animal model during a glucose clamping study as well as a human subject during an oral glucose tolerance test. We demonstrate that the spurious concentration profile-based calibration models are unable to provide prospective predictions, in contrast to those based on actual concentration profiles, especially for the physical tissue models. We also show that chance correlations incorporated by the calibration models are significantly less in Raman as compared to NIR absorption spectroscopy, even for the in vivo studies. Finally, our results suggest that the incorporation of chance correlations for in vivo cases can be largely attributed to the uncontrolled physiological sources of variations. Such uncontrolled physiological variations could either be intrinsic to the subject or stem from changes in the measurement conditions.

Quantitative determination by temperature dependent near-infrared spectra: a further study

Quantitative spectra-temperature relationship (QSTR) between near-infrared (NIR) spectra and temperature has been studied in our previous work (Talanta, 2010, 82, 1017-1021). In this study, applicability of the QSTR model for quantitative determination is further studied using the spectra of aqueous ethanol samples in the temperature range of 31-40°C and the concentration range of 1-99%. The results show that QSTR model can be built by using the spectra in a small temperature range and the quantitative analysis can be achieved by only two spectra at different temperatures. Moreover, calibration curves for different concentration ranges (1-5%, 20-70%, 95-99%, v/v) are investigated by using linear and nonlinear curve fitting, respectively. Both of the linear and nonlinear curves are found to be applicable within these concentration ranges. Therefore, the temperature dependent NIR spectra may provide a new way for quantitative determination and may have high potential in bio-fluids analysis or industrial practices.

Development of robust calibration models using support vector machines for spectroscopic monitoring of blood glucose

Sample-to-sample variability has proven to be a major challenge in achieving calibration transfer in quantitative biological Raman spectroscopy. Multiple morphological and optical parameters, such as tissue absorption and scattering, physiological glucose dynamics and skin heterogeneity, vary significantly in a human population introducing nonanalyte specific features into the calibration model. In this paper, we show that fluctuations of such parameters in human subjects introduce curved (nonlinear) effects in the relationship between the concentrations of the analyte of interest and the mixture Raman spectra. To account for these curved effects, we propose the use of support vector machines (SVM) as a nonlinear regression method over conventional linear regression techniques such as partial least-squares (PLS). Using transcutaneous blood glucose detection as an example, we demonstrate that application of SVM enables a significant improvement (at least 30%) in cross-validation accuracy over PLS when measurements from multiple human volunteers are employed in the calibration set. Furthermore, using physical tissue models with randomized analyte concentrations and varying turbidities, we show that the fluctuations in turbidity alone causes curved effects which can only be adequately modeled using nonlinear regression techniques. The enhanced levels of accuracy obtained with the SVM based calibration models opens up avenues for prospective prediction in humans and thus for clinical translation of the technology.