Method of characterizing spectrometer instruments and providing calibration models to compensate for instrument variation

• US 7,038,774 B2
• Filed: 06/16/2004
• Issued: 05/02/2006
• Est. Priority Date: 07/22/1999
• Status: Expired due to Term

• Current Assignee: GLT Acquisition Corporation
• Original Assignee: Sensys Medical, Inc. (Sensys Medical Limited)
• Inventors: Ruchti, Timothy L., Hazen, Kevin H., Blank, Thomas B., Monfre, Stephen
• Primary Examiner(s): Turner, Samuel A.
• Assistant Examiner(s): Stock, Jr., Gordon J.
• Application Number:
  US10/870,727
• Publication Number:
  US 20040223155A1
• Time in Patent Office: 685 Days
• Field of Search: 356300-334
• US Class Current: 356/326
• CPC Class Codes: A61B 2560/0223 of calibration, e.g. protocols for calibrating sensors
• CPC Class Codes: A61B 2560/0233 Optical standards
• CPC Class Codes: A61B 5/0075 by spectroscopy, i.e. measuring spectra, e.g. Raman spectroscopy, infrared absorption spectroscopy A61B5/0071 takes precedence
• CPC Class Codes: A61B 5/1075 for measuring dimensions by non-invasive methods, e.g. for determining thickness of tissue layer A61B8/0858 takes precedence
• 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 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: G01J 2003/2866 Markers; Calibrating of scan
• CPC Class Codes: G01J 3/28 Investigating the spectrum using colour filters G01J3/51
• CPC Class Codes: G01N 21/274 Calibration, base line adjustment, drift correction
• CPC Class Codes: G01N 21/3563 for analysing solids; Preparation of samples therefor
• CPC Class Codes: G01N 21/359 using near infrared light
• CPC Class Codes: G01N 21/4785 Standardising light scatter apparatus; Standards therefor
• CPC Class Codes: G01N 21/49 within a body or fluid
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

Spectrometer instruments are characterized by classifying their spectra into previously defined clusters. The spectra are mapped to the clusters and a classification is made based on similarity of extracted spectral features to one of the previously defined clusters. Calibration models for each cluster are provided to compensate for instrumental variation. Calibration models are provided either by transferring a master calibration to slave calibrations or by calculating a separate calibration for each cluster. In one embodiment, a simplified method of calibration transfer maps clusters to each other, so that a calibration transferred between clusters models only the difference between the two clusters, substantially reducing the complexity of the model.

Claims

1. A method for generating an analyte prediction, comprising the steps of:

collecting a sample spectrum;

providing a plurality of predefined clusters with corresponding calibration models, wherein each cluster is characterized according to spectral features representative of states of at least one spectrometer instrument;

mapping said sample spectrum to one of said predefined clusters; and

applying said corresponding calibration model to generate said analyte prediction.

2. The method of claim 1, wherein said states comprise any of:

variation of a single spectrometer over time; and

variation between spectrometers.

3. The method of claim 1, further comprising the step of:

classifying said at least one spectrometer instrument into at least one of said clusters.

4. The method of claim 3, said step of classifying comprising:

extracting features; and

classifying said features according to a classification model and decision rule.

5. The method of claim 4, wherein said feature extraction step comprises:

applying a mathematical transformation to enhances a particular aspect or quality of data that is useful for interpretation.

6. The method of claim 4, wherein individual features are divided into two categories, said categories comprising:

abstract features, wherein said features are extracted using various computational methods; and

simple features that are derived from an a priori understanding of a system, wherein said feature is directly related to an instrument parameter or component.

7. The method of claim 1, wherein said step of mapping said sample spectrum to one of said predefined clusters comprises:

associating said sample spectrum to one of said pre-defined clusters.

8. The method of claim 7, further comprising the step of:

reporting said sample spectrum as an outlier if a matching cluster is not found.

9. An apparatus for generating an analyte prediction, comprising:

means for collecting a sample spectrum;

means for a plurality of predefined clusters having corresponding calibration models, wherein each cluster is characterized according to spectral features representative of states of at least one spectrometer instrument;

means for mapping said sample spectrum to one of said predefined clusters; and

means for applying a corresponding calibration model to generate said analyte prediction.

10. The apparatus of claim 9, wherein said states comprise any of:

variation of a single spectrometer over time; and

variation between spectrometers.

11. The apparatus of claim 9, further comprising:

means for classifying said at least one spectrometer instrument into at least one of said clusters.

12. The apparatus of claim 11, said means for classifying further comprising:

means for extracting features; and

means for classifying said features according to a classification model and decision rule.

13. The apparatus of claim 12, wherein said feature extraction means comprises:

a mathematical transformation for enhancing a particular aspect or quality of data that is useful for interpretation.

14. The apparatus of claim 12, wherein individual features are divided into two categories, said categories comprising:

abstract features, wherein said features are extracted using various computational methods; and

simple features that are derived from an a priori understanding of a system, wherein said feature is directly related to an instrument parameter or component.

15. The apparatus of claim 9, wherein said mapping means comprise:

associating said sample spectrum to one of said pre-defined clusters.

16. The apparatus of claim 15, further comprising:

means for reporting said sample spectrum as an outlier if a matching cluster is not found.

17. An apparatus for characterizing a spectrometer instrument, comprising:

a plurality of predefined clusters having corresponding calibration models;

wherein each cluster is characterized according to spectral features representative of states of at least one spectrometer instrument; and

means for mapping a sample spectrum obtained by said spectrometer instrument to one of said predefined clusters.