Abstract: A process for enhancing a multivariate calibration through optimization of a calibration data set operates on a large calibration set of samples that includes measurements and associated reference values to automatically select an optimal sub-set of samples that enables calculation of an optimized calibration model. The process is automatic and bases sample selection on two basic criteria: enhancement of correlation between a partner variable extracted from the independent variable and the dependent variable and reduction of correlation between the dependent variable and interference. The method includes two fundamental steps: evaluation, assigning a measurement of calibration suitability to a subset of data; and optimization, selecting an optimal subset of data as directed by the measurement of suitability.

Abstract: An optical sampling interface system minimizes and compensates error resulting from sampling variations and measurement site state fluctuations. Components include: An optical probe placement guide having an aperture wherein the optical probe is received, facilitates repeatable placement accuracy on surface of a tissue measurement site with minimal, repeatable disturbance to surface tissue. The aperture creates a tissue meniscus that minimizes interference due to surface irregularities and controls variation in tissue volume sampled; an occlusive element placed over the tissue meniscus isolates the meniscus from environmental fluctuations, stabilizing hydration at the site and thus, surface tension; an optical coupling medium eliminates air gaps between skin surface and optical probe; a bias correction element applies a bias correction to spectral measurements, and associated analyte measurements. When the guide is replaced, a new bias correction is determined for measurements done with the new placement.

Abstract: An intelligent system for detecting errors and determining failure modes operates on an absorbance spectrum of in vivo skin tissue. Application of the system results in improved prediction accuracy through rejection of invalid and poor samples. System components include a noninvasive blood glucose meter, such as a near IR spectrometer, an error detection system (EDS); a system for diagnosing and mitigating errors; and a reporting method. In the EDS, a pattern classification engine and hierarchy of levels analyzes, detects and diagnoses instrument, interface and sample errors manifested in the spectrum to determine suitability of an absorbance spectrum for blood glucose measurement. The final component of the system evaluates the error condition, diagnoses the specific mode of failure (if necessary) and reports actions to be taken. Sub-components and levels of the EDS can operate independently of the other system elements to the benefit of a noninvasive glucose measurement system.

Abstract: Methods and apparatus for estimating and removing spectral interference improve precision and robustness of non-invasive analyte measurement using Near-Infrared (NIR) spectroscopy. The estimation of spectral interference is accomplished, either, through multivariate modeling or discrete factor analysis, using a calibration set of samples in which the interference is orthogonal to the analyte signal of interest, or where the shape of the interference is known. Each of the methods results in a multivariate model in which the spectral interference is estimated for a new sample and removed by vector subtraction. Independent models based on classes of sample variability are used to collapse spectral interference and determine more accurately which model is best equipped to estimate the signal of interference in the new sample. Principal components analysis and other commonly known analytical techniques can be used to determine class membership.

Abstract: The invention provides a class of samples that model the human body. This family of samples is based upon emulsions of oil in water with lecithin acting as the emulsifier. These solutions that have varying particle sizes may be spiked with basis set components (albumin, urea and glucose) to simulate skin tissues further. The family of samples is such that other organic compounds such as collagen, elastin, globulin and bilirubin may be added, as can salts such as Na+, K+ and Cl−. Layers of varying thickness with known index of refraction and particle size distributions may be generated using simple crosslinking reagents, such as collagen (gelatin). The resulting samples are flexible in each analyte's concentration and match the skin layers of the body in terms of the samples reduced scattering and absorption coefficients, &mgr;ms and &mgr;ma. This family of samples is provided for use in the medical field where lasers and spectroscopy based analyzers are used in treatment of the body.