Two stage calibration and analyte measurement scheme for spectrophotomeric analysis

• US 5,891,024 A
• Filed: 04/09/1997
• Issued: 04/06/1999
• Est. Priority Date: 04/09/1997
• Status: Expired due to Term

• Current Assignee: Datex-Ohmeda Incorporated (General Electric Company)
• Original Assignee: Ohmeda, Inc. (Linde Plc)
• Inventors: Jarman, Kristin, Pologe, Jonas A.
• Primary Examiner(s): Hindenburg, Max
• Assistant Examiner(s): WINAKUR, ERIC FRANK
• Application Number:
  US08/835,289
• Time in Patent Office: 727 Days
• Field of Search: 600/310, 600/322, 600/323, 600/326, 600/328, 600/473, 600/476, 356/39, 356/41
• US Class Current: 600/323
• CPC Class Codes: A61B 5/1455 using optical sensors, e.g. spectral photometrical oximeters
• CPC Class Codes: G01N 21/314 with comparison of measurements at specific and non-specific wavelengths dual wavelength spectrometry G01J3/427
• CPC Class Codes: G01N 33/721 Haemoglobin
• CPC Class Codes: G01N 33/726 Devices

Abstract

A two-stage statistical calibration and measurement method and system is disclosed for performing photoplethysmographic measurement of blood analyte concentrations. Concentrations in a tissue sample of MetHb, O₂ Hb, RHb and COHb are estimated by first estimating a concentration of MetHb (in a first stage) and subsequently, if the concentration of MetHb is within a predetermined range, then the estimated concentration of MetHb is assumed to be accurate and this estimated concentration of MetHb is utilized as a "known value" in determining the concentrations of the remaining analytes O₂ Hb, RHb and COHb (in a second stage). That is, by eliminating one "unknown" from the system of equations, these remaining analytes can be calculated with increased accuracy. Each stage is performed using data obtained by transmitting light through the tissue sample (typically a finger or earlobe). The transmitted light is generated by four discrete light emitters, each emitter having a distinct spectral content.

Claims

1. An apparatus for performing spectrophotometric analyte concentration measurement comprising:

means for communicating beams of light each having a distinct spectral content across a tissue sample for obtaining values related to a measurement of the concentration of each of a plurality of analytes in the tissue sample;

means for measuring the beams of light received across the tissue sample for obtaining light measurements;

first estimating means for estimating a concentration of a collection of at least one of said plurality of analytes using said measured light measurements;

means for determining whether said first estimated concentration for each of the analytes in the collection satisfies a predetermined one or more constraints; and

second estimating means for estimating a concentration of the remaining analytes of said plurality of analytes when said means for determining indicates that said predetermined one or more constraints have been met;

wherein said second estimating means determines an estimated measurement of the remaining analytes using said measured light and using said first estimated concentration as a substantially exact measurement of said collection of analytes.

2. An apparatus as claimed in claim 1, wherein the means for communicating light comprises a plurality of emitters each emitting light having a spectral content characterized by a distinct center wavelength.

3. An apparatus as claimed in claim 2, wherein each said distinct center wavelength is in a range of 600 nanometers to 1000 nanometers.

4. An apparatus as claimed in claim 2, wherein said emitter is a laser diode.

5. An apparatus as claimed in claim 1, wherein the measurement of each of said plurality of analytes is a concentration of the analyte in the blood of the tissue sample.

6. An apparatus as claimed in claim 1, wherein said plurality of analytes is selected from the group consisting of:

methemoglobin, oxyhemoglobin, deoxyhemoglobin and carboxyhemoglobin.

7. An apparatus as claimed in claim 1, wherein said means for determining determines whether said first estimated concentration satisfies a constraint that said first estimated concentration is less than a predetermined concentration in the tissue sample.

8. An apparatus as claimed in claim 1, wherein said collection of at least one of said plurality of analytes includes only methemoglobin.

9. An apparatus as claimed in claim 8, wherein said means for determining determines whether methemoglobin is less than approximately one percent of a total amount of hemoglobin in the blood of the tissue sample.

10. An apparatus as claimed in claim 1, further including a third estimating means for estimating a concentration of at least one of said remaining analytes, wherein said third estimating means is used when said means for determining determines that an estimated concentration for one or more analytes in said collection does not satisfy one of said predetermined one or more constraints.

11. An apparatus as claimed in claim 10, wherein said third estimating means uses regression coefficients for estimating the remaining analytes, wherein said regression coefficients are the result of a statistical calibration on an exactly determined system.

12. An apparatus as claimed in claim 1, wherein said remaining analytes are selected from the group consisting of:

oxyhemoglobin, deoxyhemoglobin and carboxyhemoglobin.

13. An apparatus as claimed in claim 1, wherein the total number of said distinct spectral contents of said communicated light is one of:

(a) equal to the total number of analytes in said plurality of analytes, and (b) greater than the total number of analytes in said plurality of analytes.

14. An apparatus as claimed in claim 1, wherein said first estimating means determines an estimate of the concentration of said collection using regression coefficients.

15. An apparatus as claimed in claim 14, wherein said regression coefficients are determined using one of:

least squares regression, partial least squares regression, principle component regression and Fourier regression.

16. A method for determining regression coefficients for a photoplethysmographic analyte measurement system, comprising:

first collecting noninvasive photoplethysmographic data from a plurality of test subjects using a predetermined number of light emitters, each said light emitter providing a distinct spectral content of light;

second collecting invasive concentration measurements for each of a plurality of blood analytes from each of the test subjects;

determining a first set of regression coefficients, wherein said first set;

(A) has regression coefficients for each analyte of a first collection of said plurality of analytes, said first collection including at least a first analyte; and

(B) is determined using at least one of;

(a) said noninvasive photoplethysmographic data, and (b) at least some of said invasive concentration measurements;

determining a second set of regression coefficients, said second set having regression coefficients for each analyte of a second collection of said remaining analytes not in said first collection, wherein (a) said second set is determined using at least a subcollection of said noninvasive photoplethysmographic data, and (b) said subcollection satisfies a predetermined constraint related to a concentration measurement of said first analyte obtained in said second collecting step; and

determining a third set of regression coefficients, said third set having regression coefficients for each analyte of said second collection using substantially all of said noninvasive photoplethysmographic data.

17. A method as claimed in claim 16, wherein, for each of said plurality of test subjects, said steps of first collecting and second collecting are performed substantially simultaneously on the test subject for obtaining corresponding noninvasive photoplethysmographic data and invasive concentration measurements.

18. A method as claimed in claim 17, wherein said predetermined constraint excludes at least some of said noninvasive photoplethysmographic data for a test subject having a corresponding invasive concentration measurement outside of a predetermined range.

19. A method as claimed in claim 18, wherein said predetermined range is between zero and approximately 1% when said first analyte is MetHb.

20. A method as claimed in claim 16, wherein each of said first, second and third sets of regression coefficients is determined using one of:

a least squares regression, a partial least square regression, a principle component regression and a Fourier regression.

21. A method as claimed in claim 16, wherein said first collection includes MetHb and said remaining collection includes O₂ Hb, RHb, and COHb.

22. A method for photoplethysmographic determination of analyte concentrations in a tissue comprising:

directing a plurality of beams of light through the tissue wherein each beam of light is characterized by a distinct spectral content;

measuring the light transmitted through the tissue;

estimating concentrations for at least one analyte in a first collection of a plurality of analytes using said light measurements;

comparing said estimated concentrations for the at least one analyte in said first collection with a predetermined criteria for said estimated initial concentrations;

if said estimated concentrations of said first collection of analytes satisfies said criteria then estimating a concentration for each of the remaining analytes not in said first collection of analytes using said light measurements and said estimated concentrations of said first collection of analytes as a substantially accurate estimate of the actual analyte concentration.

23. A method as claimed in claim 22, wherein said first collection of analytes comprises MetHb and said remaining collection of analytes is selected from the group consisting of:

O₂ Hb, Rhb and CoHb.

24. A method as claimed in claim 22, wherein the spectral content of each beam of light is characterized by a distinct center wavelength.