Method and system for improving photoplethysmographic analyte measurements by de-weighting motion-contaminated data

• US 6,408,198 B1
• Filed: 08/03/2000
• Issued: 06/18/2002
• Est. Priority Date: 12/17/1999
• Status: Expired due to Term

• Current Assignee: Datex-Ohmeda Incorporated (GE Healthcare Technologies, Inc.)
• Original Assignee: Datex-Ohmeda Incorporated (GE Healthcare Technologies, Inc.)
• Inventors: Hanna, D. Alan, Pologe, Jonas A.
• Primary Examiner(s): Winakur, Eric F.
• Assistant Examiner(s): KREMER, MATTHEW J
• Application Number:
  US09/632,153
• Time in Patent Office: 684 Days
• Field of Search: 600/309-310, 600/316, 600/322-324, 600/326, 600/328, 600/330, 600/336, 356/39-42
• US Class Current: 600/323
• CPC Class Codes: A61B 5/14551 for measuring blood gases
• CPC Class Codes: A61B 5/7214 using signal cancellation, e.g. based on input of two identical physiological sensors spaced apart, or based on two signals derived from the same sensor, for different optical wavelengths

Abstract

A system is provided for improving photoplethysmographic analyte measurements by de-emphasizing motion-contaminated data and/or emphasizing motion-free data. The system 100 is used in a measurement instrument which includes at least illumination source 104 for transmitting at least two light signals at two center wavelengths through a patient'"'"'s appendage 106 and a sensor 108 for converting the light signals transmitted through the appendage into electrical output signals. The system 100 includes a first buffer 116 configured to temporarily store at least one pulse cycle of output signals received from the sensor 108. A motion estimator 122 quantifies an amount of motion associated with the output signals stored in the first buffer 116 by performing a principal component analysis on differential absorption values derived from the output signals. Based on the amount of motion estimated, a weight application module 118 is configured to associate a weight with the output signals in the first buffer. The weighted output signals are transferred into a second buffer 120 which contains weighted output signals collected over multiple pulse cycles. An analyte computing module 126 computes blood analyte related values based on the weighted output signals stored in the second buffer 120.

Claims

1. An apparatus for use in a measurement instrument which includes at least one illumination source for generating a plurality of light signals at least at two different center wavelengths corresponding to at least two different channels and transmitting said light signals relative to a tissue of a test subject and a sensor configured to receive said light signals transmitted relative to the tissue and to convert the light signals received into output signals, said apparatus comprising:

a first data storage for storing input data sets representing multiple determined time periods of said output signals, each of said input data sets including first channel data including multiple values based on a first one of said light signals corresponding to a first one of said channels and second channel data including multiple values based on a second one of said light signals corresponding to a second one of said channels;

a first processor configured to receive said input data from said first data storage and perform a multichannel noise analysis involving said first channel data and said second channel data for each of said data sets, and to thereby process said input data sets to output processed data, where said processed data includes multiple output data sets, corresponding to a plurality of said determined time periods, which are selectively modified on a set-by-set basis based on said multichannel noise analysis;

a second data storage for storing said processed data output by said first processor corresponding to said plurality of said determined time periods; and

a second processor for computing a blood analyte related value of the test subject based on said processed data stored in said second data storage.

2. The apparatus of claim 1, wherein each said input data set stored in said first data storage represents at least a significant portion of one pulse cycle of said output signals produced by the sensor.

3. The apparatus of claim 1, wherein the measurement instrument is a pulse oximeter for noninvasively measuring a blood oxygen saturation in the tissue of the test subject.

4. The apparatus of claim 1, wherein said first processor is configured to estimate a degree of an artifactual component reflected in said input data and process said input data sets such that said blood analyte computation is dominated by one or more of said input data sets with a relatively small artifactual component.

5. The apparatus of claim 4, wherein said artifactual component estimated by said first processor is due to motion of the test subject where the sensor is attached.

6. The apparatus of claim 1, wherein said first processor is configured to estimate a degree of motion associated with said input data and generate a corresponding motion factor based on said degree of motion estimated.

7. The apparatus of claim 6, wherein said processed data output by said first processor represents said input data after said motion factor is used to relatively emphasize or de-emphasize said input data in proportion to the degree of motion estimated.

8. The apparatus of claim 6, wherein said processed data output by said first processor represents said input data after said motion factor is used to eliminate said input data from the blood analyte computation if the degree of motion estimated associated therewith exceeds a determined threshold.

9. The apparatus of claim 6, wherein the degree of motion is estimated based on an assumption that motion affects each of the first and second pulse waveforms substantially the same.

10. The apparatus of claim 1, wherein the output signals produced by the sensor are separated into first and second pulse waveforms respectively representative of the light signal emitted at first and second wavelengths after being absorbed by said tissue.

11. The apparatus of claim 10, further comprising means for computing a set of data points from first and second pulse waveforms, each data point representative of a pair of differential absorption values of said first and second pulse waveforms.

12. The apparatus of claim 11, wherein said first processor is configured to determine a degree of variation among said set of data points with respect to a best-fit linear regression line.

13. The apparatus of claim 11, wherein said first processor is configured to determine a first degree of variation among said set of data points in one direction and a second degree of variation among said set of data points in another direction perpendicular to said one direction, wherein said degree of motion associated with said at least one pulse cycle of output signals is proportional to a product of said first and second degree of variations.

14. The apparatus of claim 11, wherein said second processor computes the blood analyte related value by determining a slope of a line which best fits a plurality of sets of data points after each individual set of data points is emphasized or de-emphasized based on the degree of motion associated therewith and computing the blood analyte related value based on said slope of the best-fit line.

15. An apparatus for determining the reliability of output signals produced by a pulse oximeter sensor, comprising:

means for receiving output data from said pulse oximeter sensor corresponding to first and second determined time periods, said output data including a first data set including a plurality of first data values for said first time period and a second data set including a plurality of second data values for said second time period;

means for analyzing said first and second data sets to estimate corresponding degree of artifactual component reflected in said first and second data sets and apply respective first and second weights to said first and second data sets; and

a processor for using said first and second data sets and said first and second weights to make a determination of a blood analyte relate value, where said first and second data values are attributed different weights in said determination.

16. The apparatus of claim 15, wherein said means for analyzing further comprises:

means for separating the output signals produced by the pulse oximeter into first and second pulse waveforms respectively representative of light signals emitted at first and second center wavelengths after being absorbed by a tissue of a test subject;

means for obtaining a set of data points from first and second pulse waveforms, wherein each respective one of said data points represents a pair of differential absorption values derived from said first and second pulse waveforms;

means for determining a degree of variation among said set of data points with respect to a best-fit linear regression line.

17. The apparatus of claim 16, wherein said artifactual component is caused by motion of the test subject where the sensor is attached and said degree of artifactual component is estimated based on an assumption that motion affects each of the first and second pulse waveforms substantially the same.

18. The apparatus of claim 17, wherein said means for analyzing further comprises:

means for determining a first degree of variation among said set of data points in one direction; and

means for determining a second degree of variation among said set of data points in another direction perpendicular to said one direction, wherein said degree of artifactual component associated with said output signals is proportional to a product of said first and second degree of variations.

19. A pulse oximeter for noninvasively measuring a blood analyte related value in a tissue of a test subject, comprising:

at least one illumination source for generating at least two light signals at predetermined center wavelengths and for transmitting said light signals through the tissue;

a sensor configured to receive said light signals transmitted through the tissue and to convert the light signals received into output signals; and

a processor configured to receive a plurality of input data sets, each of said input data sets representing a determined time period of said output signals produced by said sensor and to preprocess each respective input data set by applying different weights to each said set to thereby establish a preprocessed data output reflecting said different weights;

wherein said blood analyte related value is computed based on said preprocessed data reflecting said different weights.

20. The pulse oximeter of claim 19, wherein said processor is configured to evaluate the quality of each of said input data sets individually based on a mathematical analysis thereof.

21. The pulse oximeter of claim 19, wherein the processor is configured to separate said output signals into first and second pulse waveforms and to estimate a quantity of motion associated with each of said input data sets based on an assumption that motion affects each of the first and second pulse waveforms substantially the same.

22. The pulse oximeter of claim 21, wherein a weight parameter is assigned to each of said input data sets based on said quantity of motion estimated, and said weight parameter is used to accomplish at least one of the following:

emphasize output signals that are not affected by motion and de-emphasize output signals that are affected by motion relative to said quantity of motion estimated.

23. A method for improving the accuracy of a blood analyte related value measured by a pulse oximeter having an illumination source for transmitting light signals of at least at two different predetermined center wavelengths relative to tissue of a test subject and a sensor configured to receive the light signals transmitted relative to the tissue and to convert the light signals received into output signals, said method comprising the steps of:

based on said output signals, storing a first data set corresponding to a first time period and a second data set corresponding to a second time period in a first memory;

processing said first and second data sets stored in said first memory to generate a processed data set, for storage in second memory, based on but different than said first and second data sets; and

;

computing said blood analyte related value for the tissue of the test subject based on said processed data set, where said first and second data sets are weighted differently in said step of computing.

24. The method of claim 23, wherein said processing step further comprises the step of identifying motion affected output signals based on a mathematical analysis of said determined time period of said output signals.

25. The method of claim 23, wherein said processing step further comprises the step of estimating a quantity of motion associated with said determined time period of said output signals based on said mathematical analysis thereof.

26. The method of claim 25, wherein said processing step further comprises at least one of the following steps:

emphasizing said output signals according to said corresponding quantity of motion estimated; and

de-emphasizing said output signals according to said corresponding quantity of motion estimated.

27. The method of claim 25, wherein said processing step further comprises the step of eliminating said determined time period of said output signals from the blood analyte related value computation if the quantity of motion estimated associated therewith exceeds a determined threshold.

28. The method of claim 23, wherein said processing step is accomplished with a processor in connection with said first memory.

29. The method of claim 28, further comprising the step of accumulating in a second memory a plurality of said processed data output by said processor, wherein said blood analyte related value is computed based on said plurality of said processed data stored in said second memory.

30. The method of claim 23, further comprising the step of separating said output signals into first and second pulse waveforms respectively representative of light signals emitted at first and second wavelengths after being absorbed by said tissue.

31. The method of claim 30, wherein the processing step further comprises the step of evaluating the reliability of said output signals based on a mathematical analysis of said first and second pulse waveforms.

32. The method of claim 31, wherein the reliability of said output signals is evaluated on the basis of a quantity of motion associated with said output signals.

33. The method of claim 32, wherein the quantity of motion associated with said output signals is estimated based on an assumption that motion affect each of the first and second pulse waveforms substantially the same.

34. The method of claim 32, wherein said evaluating step further comprises the steps of:

deriving a set of data points from said first and second pulse waveforms, each data point representative of a pair of differential absorption values of said first and second pulse waveforms; and

estimating said quantity of motion associated with said set of data points based on a degree of variation among said set of data points with respect to a best-fit linear regression line.

35. The method of claim 34, wherein the evaluating step further comprises the steps of:

using said first and second waveforms to acquire pairs of differential absorption values dA_x and dA_y;

approximating a degree of motion artifact associated with the output signals based on the concentration of the pairs of differential absorption values dA_x and dA_y along a best-fit linear line.

36. The method of claim 32, wherein the processing step further comprises at least one of the following steps:

emphasizing the output signals that are determined to be reliable in the evaluating step;

de-emphasizing the output signals that are determined to be unreliable in the evaluating step; and

eliminating the output signals that are determined to be unreliable in the evaluating step.