A Review of Calibration Transfer Practices and Instrument Differences in Spectroscopy.

Calibration transfer for use with spectroscopic instruments, particularly for near-infrared, infrared, and Raman analysis, has been the subject of multiple articles, research papers, book chapters, and technical reviews. There has been a myriad of approaches published and claims made for resolving the problems associated with transferring calibrations; however, the capability of attaining identical results over time from two or more instruments using an identical calibration still eludes technologists. Calibration transfer, in a precise definition, refers to a series of analytical approaches or chemometric techniques used to attempt to apply a single spectral database, and the calibration model developed using that database, for two or more instruments, with statistically retained accuracy and precision. Ideally, one would develop a single calibration for any particular application, and move it indiscriminately across instruments and achieve identical analysis or prediction results. There are many technical aspects involved in such precision calibration transfer, related to the measuring instrument reproducibility and repeatability, the reference chemical values used for the calibration, the multivariate mathematics used for calibration, and sample presentation repeatability and reproducibility. Ideally, a multivariate model developed on a single instrument would provide a statistically identical analysis when used on other instruments following transfer. This paper reviews common calibration transfer techniques, mostly related to instrument differences, and the mathematics of the uncertainty between instruments when making spectroscopic measurements of identical samples. It does not specifically address calibration maintenance or reference laboratory differences.

Pattern recognition assisted infrared library searching.

Pattern recognition methods have been used to develop search prefilters for infrared (IR) library searching. A two-step procedure has been employed. First, the wavelet packet tree is used to decompose each spectrum into wavelet coefficients that represent both the high and low frequency components of the signal. Second, a genetic algorithm for pattern recognition analysis is used to identify wavelet coefficients characteristic of functional group. Even in challenging trials involving carboxylic acids, compounds that possess both carbonyl and hydroxyl functionalities can be readily differentiated from carboxylic acids. The proposed search prefilters allow for the use of more sophisticated and correspondingly more time-consuming algorithms in IR spectral library matching because the size of the library can be culled down for a specific match using information from the search prefilter about the presence or absence of specific functional groups in the unknown.

Direct measurements of blood glucose concentration in the presence of saccharide interferences using slope and bias orthogonal signal correction and Fourier transform near-infrared spectroscopy.

Saccharide interferences such as Dextran, Galactose, etc. have a great potential to interfere with near infrared (NIR) glucose analysis since they have a similar spectroscopic fingerprint and are present physiologically at large relative concentrations. These can lead to grossly inappropriate interpretation of patient glucose levels and resultant treatment in critical care and hospital settings. This study describes a methodology to reduce this effect on glucose analysis using an NIR Fourier transform spectroscopy method combined with a multivariate calibration technique (PLS) using preprocessing by orthogonal signal correction (OSC). A mathematical approach based on the use of a single calibration based bias and slope correction was applied in addition to a standard OSC was investigated. This approach is combined with a factorial interferent calibration design to accommodate for interference effects. We named this approach as a slope and bias OSC (sbOSC). sbOSC differs from OSC in the way it handles the prediction. In sbOSC, statistics on slope and bias obtained from a set of calibration samples are then used as a validation parameter in the prediction set. Healthy human volunteer blood with different glucose (80 to 200 mg/dL) and hematocrit (24 to 48 vol.%) levels containing high expected levels of inteferents have been measured with a transmittance near-infrared Fourier transform spectrometer operates in the broadband spectral range of 1.25-2.5 µm (4000-8000 cm(-1)). The effect of six interferents compounds used in intensive care and operating rooms, namely Dextran, Fructose, Galactose, Maltose, Mannitol, and Xylose, were tested on blood glucose. A maximum interference effect (MIE) parameter was used to rank the significance for the individual interferent type on measurement error relative to the total NIR whole blood glucose measurement error. For comparison, a YSI (Yellow Springs Instrument) laboratory reference glucose analyzer and NIR data were collected at the same time as paired samples. MIE results obtained by sbOSC were compared with several standard spectral preprocessing approaches and show a substantial reduced effect of saccharide interferences. NIR glucose measurement results are substantially improved when comparing standard error of prediction from validation samples; and resulting MIE values are small.