Surgery – Diagnostic testing – Measuring or detecting nonradioactive constituent of body...
Reexamination Certificate
2000-07-14
2002-11-05
Winakur, Eric F. (Department: 3736)
Surgery
Diagnostic testing
Measuring or detecting nonradioactive constituent of body...
C600S365000, C600S322000
Reexamination Certificate
active
06477392
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to instruments for the non-invasive quantitative measurement of constituents, such as blood glucose levels in blood. More specifically, this invention provides improvements in methods and apparatus for near-infrared quantitative analysis.
BACKGROUND AND RELATED PRIOR ART
The use of quantitative near-infrared (“NIR”) analysis for the determination of chemical and/or physical characteristics of products is relatively well known in the art. See “An Introduction to Near Infrared Quantitative Analysis,”by Robert D. Rosenthal, presented at the 1977 Annual Meeting of the American Association of Cereal Chemists, (1978). See also U.S. Pat. No. 4,286,327 issued to Rosenthal et al. on Aug. 25, 1981.
Another well-known application of NIR analysis relates to the quantitative measurement of analytes in mammals, such as quantitative analysis of glucose in the blood. Information concerning the chemical composition of blood is widely used to assess the health characteristics of both people and animals. More specifically, analysis of the glucose content of blood provides an indication of the current status of the metabolism. Blood analysis, by the detection of above or below normal levels of various substances, also provides a direct indication of the presence of certain types of diseases and dysfunctions.
In particular, the non-invasive NIR quantitative measurement apparatus has particular application for use by diabetics in monitoring the level of glucose in the blood. See U.S. Pat. No. 5,028,787, Rosenthal et al., issued Jul. 2, 1991, the subject matter of which is hereby incorporated by reference in its entirety.
Quantitative NIR analysis is based on the principle that most organic (and some inorganic) substances absorb radiation in the near-infrared range, with the different substances having different absorption characteristics over specific NIR wavelength ranges. These different characteristics are then used to formulate specific measurement algorithms for obtaining quantitative information regarding the presence of such substances in the subject sample, product or patient.
The above-cited '327 patent teaches the use of infrared emitting diodes (IREDs) as sources of near-infrared radiation. As shown in
FIG. 1
, a plurality (eight in the figure) of IREDs
10
is arranged over a sample WS to be illuminated for quantitative analysis. Near-infrared radiation emitted from each IRED impinges upon an accompanying optical filter
12
. Each optical filter
12
is a narrow bandpass filter that passes NIR radiation at a different wavelength, and light baffles
14
are provided between IREDs to prevent an IRED's light from being transmitted through an adjacent filter. In the illustrated example, the sample WS is held in a holder
16
having a transparent bottom
18
. NIR radiation passing through the sample and the holder is detected by a detector
20
, such as a silicon photodetector, and converted to an electrical signal. The electrical signal is processed by processing circuitry, including an amplifier
22
, logarithmic amplifier
23
, and analog-to-digital converter
24
, and inputted to microprocessor
11
. The microprocessor processes the data from the detector, using preprogrammed algorithms to obtain a quantitative measurement of the analytes of interest in the sample, and outputs the result on a display
26
.
FIG. 2
illustrates another known NIR instrument for non-invasive measurement of blood analytes, as disclosed in U.S. Pat. No. 5,077,476, issued to Rosenthal on Dec. 31, 1991. The subject matter of the '476 patent is also incorporated by reference herein in its entirety. In brief summary, the instrument
1
uses a number of IREDs (
50
,
60
as shown in
FIG. 2
) for irradiating a body part, such as the finger, with NIR radiation at selected wavelengths. Narrow bandpass optical filters
160
and
170
are positioned at the output of the IREDs to pass NIR radiation at a selected wavelength. The radiation passes through a window
140
, through the subject, and is detected by a detector
80
. A light baffle
40
is provided to isolate the various IREDs to prevent radiation from one IRED passing through the optical filter associated with a different IRED.
The detector
80
outputs a signal to a microprocessor
100
through amplifier
90
. The microprocessor calculates the concentration of analytes at issue (such as blood glucose) using preprogrammed algorithms and outputs the results to a display device
180
. In this instrument, timing and control circuitry
110
is provided to sequentially and individually turn on and off each IRED, one at a time, so that the absorption by the blood analytes and other substances may be measured at each particular wavelength specified in the measurement algorithm.
In almost all known NIR measurement devices, such as those described above, the preprogrammed algorithms for interpreting the quantitative measurements are based upon Beer's Law, which provides that the amount of chemical constituent(s) to be measured is linearly related to D, where
D
=
log
(
1
I
)
,
(
Equation
1
)
in which I can be the fraction of light that is either transmitted through an object, reflected off an object or interacted with the object. This linear concept has worked quite well in many NIR measurement applications. For example, U.S. Pat. No. 4,928,014, issued May 22, 1990 to Rosenthal, provides for the measurement of body fat in the human body using an NIR measurement device.
These historical successes using Beer's Law have been in applications where the change in constituent concentration has been modest. For example, in the measurement of body fat, as taught by the '014 patent, the percentage of body fat varies from a minimum in the neighborhood of 10% to a maximum of approximately 40%, a four to one change. In contrast, a much larger range of concentration change occurs in certain blood analytes. For example, blood glucose molality can vary from 20 mg/dL to more than 500 mg/dL, a change of twenty-five to one. Over such large ranges of values, the basic assumption in Beer's Law of linearity may be invalid. As a result, conventional Multiple Linear Regression (“MLR”) or factor analysis does not provide sufficient accuracy to be meaningful. In fact, the D values, as defined in Equation 1, from the NIR measurement of blood glucose values generally have a highly nonlinear relationship.
Attempts to use an NIR device to measure blood glucose have encountered many problems. The NIR measurement devices must be calibrated for each individual user, with more accurate calibrations resulting in more accurate readings. It is generally tedious and time consuming to accurately calibrate the NIR measurement device. During the calibration, measurements of the NIR device are compared with other measurements known to be accurate. The NIR device is then adjusted to produce measurements that correspond with the measurements known to be accurate. The conventional approach for the NIR calibration is to evaluate absorption wavelengths and reference wavelengths. The absorption wavelengths are the portions of the electromagnetic spectrum with high absorbance due to the particular constituents of interest, and reference wavelengths are the portions of the spectrum that are insensitive to the particular constituents. However, in the very near infrared portion of the spectrum, from 700 nm to 1100 nm, the absorptions are quite broad, and thus, independent selection of wavelengths is difficult. Moreover, in the case of in vivo measurement of various blood analytes through a body part, the problem is even more difficult. Approximately ninety-nine percent of the organic content of a finger is due to the presence of water, fat and muscle (protein). The remaining 1 percent includes all of the blood analytes to be measured, as well as other materials. Even further complicating the NIR measurement is that blood analytes, such as blood glucose, are very weak NIR absorbers. There thus exists
Honigs David E.
Rosenthal Robert D.
Futrex Inc.
Kremer Matthew
Rothwell Figg Ernst & Manbeck
Winakur Eric F.
LandOfFree
Calibration of near infrared quantitative measurement device... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Calibration of near infrared quantitative measurement device..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Calibration of near infrared quantitative measurement device... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2992806