Method for building an algorithm for converting spectral information

• US 20080045820A1
• Filed: 10/10/2007
• Published: 02/21/2008
• Est. Priority Date: 02/11/2002
• Status: Active Grant

• Current Assignee: Ascensia Diabetes Care Holdings AG (PHC Holdings Corporation)
• Original Assignee: Bayer Healthcare LLC (Bayer AG)
• Inventors: Brown, Steven, Rebec, Mihailo, Tan, Hu-Wei, Smous, James
• Granted Patent: US 8,452,359 B2
• Time in Patent Office: Days
• US Class Current: 600/316
• CPC Class Codes: A61B 5/14532 for measuring glucose, e.g. by tissue impedance measurement
• CPC Class Codes: A61B 5/1455 using optical sensors, e.g. spectral photometrical oximeters
• CPC Class Codes: A61B 5/1495 Calibrating or testing of in-vivo probes
• CPC Class Codes: A61B 5/726 using Wavelet transforms
• CPC Class Codes: G01N 21/274 Calibration, base line adjustment, drift correction
• CPC Class Codes: G01N 21/359 using near infrared light
• CPC Class Codes: G01N 2201/067 Electro-optic, magneto-optic, acousto-optic elements

Abstract

A system for determining the concentration of an analyte in at least one body fluid in body tissue comprises an infrared light source, a body tissue interface, a detector, and a central processing unit. The body tissue interface is adapted to contact body tissue and to deliver light from the infrared light source to the contacted body tissue. The detector is adapted to receive spectral information corresponding to infrared light transmitted through the portion of body tissue being analyzed and to convert the received spectral information into an electrical signal indicative of the received spectral information. The central processing unit is adapted to compare the electrical signal to an algorithm built upon correlation with the analyte in body fluid, the algorithm adapted to convert the received spectral information into the concentration of the analyte in at least one body fluid.

Claims

1-97. -97. (canceled)

98. A method for building an algorithm for converting spectral information obtained from body tissue into the concentration of an analyte in at least one body fluid in the body tissue, the method comprising the acts of:

modulating the concentration of the analyte in at least one body fluid in the body tissue to a plurality of predetermined levels within a range of predetermined analyte concentration levels;

directing infrared light onto the body tissue of at least one test subject being analyzed during an analyte clamping test;

collecting spectral information from the body tissue;

collecting data other than spectral information from body tissue of the at least one test subject during the modulating of the concentration of the analyte in at least one body fluid in the body tissue;

modeling spectral information corresponding to analyte concentration levels outside the range of analyte concentration levels; and

filtering the collected and modeled spectral information to remove non-analyte spectral information using one or more chemometric techniques selected from the group consisting of wavelets analysis, orthogonal signal correction, partial least squares regression, principal components analysis, and genetic algorithms.

99. The method of claim 98 further comprising recording invasively determined analyte concentration levels of the body tissue of the at least one test subject in a invasively determined analyte data set.

100. The method of claim 99 further comprising recording the collected spectral information and the modeled spectral information in a spectral data set.

101. The method of claim 100 further comprising normalizing the data of the spectral data set.

102. The method of claim 101 further comprising applying an orthogonal signal correction analysis to the normalized data set to create a corrected spectral data set.

103. The method of claim 102 further comprising filtering the corrected spectral data set by applying a wavelet analysis to form a filtered spectral data set.

104. The method of claim 103 further comprising applying a partial least squares regression analysis to the filtered spectral data set and the invasively determined analyte data set to build a analyte calibration algorithm.

105. The method of claim 98 wherein modulating further comprises conducting a analyte clamping experiment.

106. The method of claim 98 wherein the data are selected from the group consisting of invasively determined analyte concentration levels of the body tissue, temperature of the body tissue, pressure applied to the body tissue, frequency of the infrared light directed onto the body tissue, and time of data collection.

113. The method of claim 98 wherein the analyte is glucose.

107. A method for building an algorithm for converting spectral information obtained from body tissue into the concentration of analyte in at least one body fluid in the body tissue, wherein the analyte concentration level of the at least one body fluid in body tissue of a plurality of test subjects including a first test subject and a second test subject is modulated to a plurality of predetermined analyte concentration levels within a range of analyte concentration levels, the method comprising the acts of:

invasively determining the analyte concentration levels of the at least one body fluid of each of the test subjects during the modulating of analyte concentration levels of the at least one body fluid in body tissue;

directing infrared light onto the body tissue of the test subjects being analyzed during the analyte clamping test;

collecting spectral information from the body tissue of the test subjects;

normalizing the information collected from the body tissue of the test subjects;

combining a portion of the normalized spectral information collected from the first test subject with a portion of the normalized spectral information collected from the second test subject to create a combined spectral data set;

mean centering the combined spectral data set;

creating a partial least squares regression model having at least one orthogonal signal correction component from the combined spectral data set, the partial least squares regression model being adapted to filter out non-analyte related spectral information from the spectral data set;

applying an orthogonal signal correction analysis to the normalized spectral information collected from the first test subject to form a first corrected data set and to the normalized spectral information collected from the second test subject from a second corrected data set; and

applying the partial least squares regression model to first corrected data set and the invasively determined analyte concentration levels collected from the first test subject to form the algorithm for converting spectral information obtained from body tissue into the concentration of analyte in at least one body fluid in the body tissue.

108. The method of claim 107 further comprising applying the algorithm to the second corrected data set for predicting the analyte concentration levels of the at least one body fluid in the body tissue of the second test subject.

109. The method of claim 108 further comprising comparing the predicted analyte concentration levels of the at least one body fluid in the body tissue of the second test subject to the invasively determined the analyte concentration levels collected from the second test subject during the analyte clamping test.

110. The method of claim 107 further comprising filtering the first and the second corrected spectral data set by applying a wavelet analysis prior to applying the partial least squares regression model.

111. The method of claim 107 further comprising collecting data other than spectral information during the during the modulating of analyte concentration levels, the data being selected from the group consisting of a target body fluid analyte concentration for each test subject, temperature of the body tissue for each test subject, pressure applied to the body tissue of each test subject, frequency of the infrared light directed onto the body tissue of each test subject, and time of data collection.

112. The method of claim 107 further comprising:

modeling spectral information corresponding to analyte concentration levels outside the range of analyte concentration levels; and

combining the modeled spectral information with the collected spectral information from the body tissue of the test subjects.

114. The method of claim 107 wherein the analyte is glucose.