Surgery – Diagnostic testing – Measuring or detecting nonradioactive constituent of body...
Reexamination Certificate
2002-06-25
2004-02-24
Winakur, Eric F. (Department: 3736)
Surgery
Diagnostic testing
Measuring or detecting nonradioactive constituent of body...
C600S473000, C356S300000, C250S339110, C250S341800
Reexamination Certificate
active
06697654
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates to the estimation and subtraction of interference from a NIR spectral measurement. More particularly, this invention relates to an apparatus and methods for determining targeted orthogonal interference and the development of models for removing unwanted spectral variation from a NIR measurement using techniques such as multivariate regression and discrete factor analysis.
2. Description of Related Art
Near-infrared (NIR) tissue spectroscopy is a promising noninvasive technology that bases measurements on the irradiation of a tissue site with NIR energy in the 700-2500 nanometer wavelength range. The energy is focused onto an area of the skin and propagates according to the scattering and absorption properties of the skin tissue. Therefore, the reflected or transmitted energy that escapes and is detected provides information about the tissue volume encountered. Specifically, the attenuation of the light energy at each wavelength is a function of the structural properties and chemical composition of the tissue. Tissue layers, each containing a unique heterogeneous particulate distribution, affect light absorbance through scattering. Chemical components such as water, protein, fat and analytes absorb light proportionally to their concentration through unique absorption profiles or signatures. The measurement of tissue properties, characteristics or composition is based on detecting the magnitude of light attenuation resulting from the respective scattering and/or absorption properties of the tissue sampled.
BIOLOGICAL ANALYTE MEASUREMENT
While global noninvasive measurement of biological constituents, such as glucose concentration, has been pursued through NIR spectroscopy, the reported success and product viability has been limited by the lack of a system for compensating for structural variations that occur over time in an individual and those that are present between individuals. This variation produces dramatic changes in the optical properties of the sampled tissue and inhibits the measurement of the signal related to the target constituents. See, for example, O. Khalil, Spectroscopic and Clinical Aspects of Non-invasive Glucose Measurements,
Clinical Chemistry
, Vol. 45, pp. 165-177, (1999) or J. Roe, B. Smoller, Bloodless Glucose Measurements,
Critical Reviews in Therapeutic Drug Carrier System
, Vol. 15 (3), pp. 199-241 (1998). With this problem being recognized, research has lead in the direction of developing analyte measurement models for individuals. Although variations in the optical properties of skin are reduced, the variations resulting from inconsistent sampling methods, for example, from coupling of the sample to the measurement device and slight variations in choice of sampling site, result in limited success. Furthermore, fluctuations in the physiological state of the individual, for example, changes in skin temperature, skin hydration levels, weight loss or weight gain, still limit the success of these models. Some of these variations can be reduced with refined experimental design and elaborate control strategies, but the resources required for such development are considerable. Therefore, variations in sampling technique and fluctuations in the physiological state of the individual are significant obstacles to overcome in the development of effective models for noninvasive measurement of analytes through NIR spectral absorbance.
The related application, T. Ruchti, T. Blank, A Multi-tier Approach to
Building Classification Models on Individuals for Noninvasive Measurement of Blood Glucose
, U.S. patent application Ser. No. 09/825,687 describes a method for substantially reducing spectral interference due to structural variations between individuals by classifying subjects according to major skin tissue characteristics prior to analyte measurement prediction. However, the subject application does not describe methods or apparatus to reduce variation across successive spectral measurements on the same individual.
In living subjects, inadequate sampling procedures and uncontrollable changes in skin tissue characteristics have been discovered to add significant interference to spectral measurements. This becomes increasingly important when attempting to estimate trace levels of analytes non-invasively. Therefore, an automated method for the estimation and removal of spectral interference prior to analyte measurement can provide increased measurement precision and accuracy.
SUMMARY OF THE INVENTION
The invention provides an apparatus and methods for modeling and removing targeted interfering signals from noninvasive spectral measurements such as NIR spectra. The invented methods are of utility in several areas, including analyte measurement and signal characterization. By grouping spectral measurements according to similar characteristics representing spectral variability, nonlinear variation is reduced and the determination and removal of interfering signals becomes easier, resulting in more accurate measurement of analytes. The invention finds particular application in the reduction of variation due to spectral interference across successive spectral measurements on the same individual.
A spectroscopic apparatus is used in conjunction with an optical interface to measure tissue properties and characteristics that are manifested spectrally and vary differently according to the sampled tissue and physiological changes in an individual.
Methods for determining the Net Analyte Signal (NAS) of a known specific spectral interference source utilize multivariate modeling and sample classification to remove unwanted spectral variation from future samples, thus yielding increased measurement precision and accuracy of analyte measurement.
A procedure for estimating known spectral interferences utilizes an empirically derived calibration model consisting of NIR tissue measurements from a set of exemplary samples and the measurements corresponding to a signal of interest that is to be removed. The model comprises a set of parameters and computer-generated code implemented to estimate the interfering signal of interest. The estimated signal reveals information relating to the property magnitude that the interference adds at any particular wavelength. Such properties include but are not limited to skin temperature, tissue hydration, sampled tissue site, pressure at apparatus interface, and day-to-day changes in an individual's physiological state.
The invention further provides a multi-tier approach to building classification models for specific interferences and its application to estimation of the true signal of interference in a new sample with greater accuracy.
REFERENCES:
patent: 5379238 (1995-01-01), Stark
patent: 5435309 (1995-07-01), Thomas et al.
patent: 5953381 (1999-09-01), Tsukahara
patent: 6175602 (2001-01-01), Gustafsson et al.
Blank Thomas B.
Lorenz Alexander D.
Ruchti Timothy L.
Glenn Michael A.
Glenn Patent Group
Sensys Medical Inc.
Winakur Eric F.
LandOfFree
Targeted interference subtraction applied to near-infrared... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Targeted interference subtraction applied to near-infrared..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Targeted interference subtraction applied to near-infrared... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3345071