Detection of sensor off conditions in a pulse oximeter

• US 6,510,329 B2
• Filed: 01/24/2001
• Issued: 01/21/2003
• Est. Priority Date: 01/24/2001
• Status: Expired due to Term

• Current Assignee: Datex-Ohmeda Incorporated (General Electric Company)
• Original Assignee: Datex-Ohmeda Incorporated (General Electric Company)
• Inventors: Heckel, Donald W.
• Primary Examiner(s): Wolfe, Willis R.
• Assistant Examiner(s): Gimie, Mahmoud
• Application Number:
  US09/769,795
• Publication Number:
  US 20020137995A1
• Time in Patent Office: 727 Days
• Field of Search: 600/323, 600/310, 600/322, 600/336, 600/309, 600/328, 356/41, 356/40, 128/920
• US Class Current: 600/310
• CPC Class Codes: A61B 2560/0276 Determining malfunction
• CPC Class Codes: A61B 5/14551 for measuring blood gases
• CPC Class Codes: A61B 5/6826 Finger
• CPC Class Codes: A61B 5/6838 Clamps or clips
• CPC Class Codes: A61B 5/6843 Monitoring or controlling sensor contact pressure
• CPC Class Codes: A61B 5/7221 Determining signal validity, reliability or quality A61B5/7207 and A61B5/7217 take precedence
• CPC Class Codes: Y10S 128/92 Computer assisted medical diagnostics

Abstract

A system for detecting sensor off conditions in a pulse oximeter is disclosed. In one implementation, a detector off condition is identified based a volatility of a signal quality of a filtered detector signal. The detector signal is first filtered, for example, to reduce the effects of noise or to otherwise reduce or eliminate components extraneous to the signal of interest. The filtered signal is used to perform a volatility measurement. In this regard, a correlation coefficient may be computed relative to two channels of the filtered signal and a correlation coefficient may be monitored over a moving window to calculate variance values for the correlation coefficient. The resulting variance values may be used alone or in combination with other parameter values to identify a sensor off condition.

Claims

1. A method for use in detecting and responding to a sensor off condition in a pulse oximetry system, said pulse oximetry system including a first source for providing a first channel signal associated with a first wavelength and a second source for providing a second channel signal associated with a second wavelength, said method comprising the steps of:

obtaining first and second detector signals associated with said first and second wavelengths, respectively;

monitoring said first and second detector signals over time period including a first time window and a second time window;

first determining a first correlation value relating to a first correlation between said first and second signals for said first time window;

second determining a second correlation value relating to a second correlation between said first and second signals for said second time window;

performing a comparison between said first and said second correlation values;

based on said comparison, determining a change value related to a change in said first correlation value relative to said second correlation value; and

identifying a sensor off condition based on said change value.

2. The method of claim 1, wherein said monitoring step comprises monitoring said first and second signals over a time period of between about 3 seconds and 30 seconds.

3. The method of claim 1, wherein said monitoring step comprises monitoring said first and second signais over a time period of between about 7 seconds and 10 seconds.

4. The method of claim 1, wherein determining a plurality of correlation values comprises for each said correlation value correlating a plurality of individual data points from said first and second signals for a corresponding time window.

5. The method of claim 2, wherein calculating each said correlation value comprises correlating said data points from said corresponding time window wherein said corresponding time window is at least partially included. within said monitored time period.

6. The method of claim 1, wherein said determining a value related to, change step comprises determining a mathematical deviation of at least one said correlation value from a normalized value associated with said plurality of correlation values.

7. The method of claim 6, wherein determining a mathematical deviation of said at least one correlation value comprises determining a standard deviation of said at least one correlation value from a mean correlation value associated with said plurality of correlation values.

8. The method of claim 1, wherein said identifying a sensor off condition comprises comparing at least one of said value related to change to and a normalized value associated with said plurality of correlation values to a predetermined threshold.

9. A method for use in detecting and responding to a sensor off conditions in a pulse oximetry system, said pulse oximetry system including a first source for providing a first channel signal associated with a first wavelength and a second source for providing a second channel signal associated with a second wavelength, comprising the steps of:

defining a measure of signal quality for a detected signal based on a comparative analysis of said first and second channel signals;

monitoring said measure of signal quality over a moving time window of a predetermined temporal length;

first calculating a reference value for said measure of signal quality for said moving time window;

second calculating a value related to variance of said measure of signal quality relative to said reference value; and

identifying a sensor off condition based on said value related to variance.

10. The method of claim 9 wherein said step of defining a measure of signal quality comprises identifying at least one feature within a plethysrrographic signal.

11. The method of claim 10, wherein said step of monitoring said signal quality comprises identifying a correlation between said first and second channels within said plethysmographic signal.

12. The method of claim 11, wherein said step of monitoring comprises determining a plurality of correlation values between said channels for said moving time window.

13. The method of claim 12, wherein said step of first calculating comprises determining said reference value as a mean value for said correlation values for said moving time window.

14. The method of claim 13, wherein said step of second calculating a value related to variance comprises at least one of:

comparison of data points from at least one of said first and second channels with said reference value associated with said moving time window; and

statistical comparison of at least one correlation value to said reference value associated with said moving time window.

15. A method for use in detecting and responding to a sensor off conditions in a pulse oximetry system, said pulse oximetry system including a detector for receiving light signals modulated by tissue of a patient, said method comprising the steps of:

obtaining first and second signals associated with first and second wavelengths of said light signals, respectively, from said detector;

filtering each said first and second signals to obtain a series of corresponding filtered values for each said wavelength, wherein filtering comprises performing a weighted average of at least a portion of each said signal over a moving time window associated with each said filtered value;

utilizing said corresponding filtered values for said first and second signals to determine a correlation value relating to a correlation between said first and second signals; and

using said correlation value to identify a sensor off condition.

16. The method of claim 15, wherein said obtaining step comprises obtaining a plurality of data points for each said wavelength.

17. The method of claim 16, wherein for each said wavelength said step of filtering comprises performing a weighted average of said data points within said moving time window to determine said filtered values.

18. The method of claim 15, wherein said using step comprises using a first correlation values associated with a first moving time window and a second correlation value associated with a second time window to identify said sensor off condition.

19. The method of claim 18, wherein said step of identifying comprises comparing said first correlation value to said second correlation value.

20. The method of claim 19, wherein said step identifying comprises providing an sensor off indication when a change is said first correlation value relative to said second correlation value exceeds a predetermined threshold for a predetermined period.

21. A method for use in detecting and responding to a sensor off conditions in a pulse oximetry system, said method comprising the steps of:

defining a measure of correlation between first and second channels for a detected signal;

monitoring said measure of correlation between said first and second channels over a time window of a predetermined length;

determining a first value related to a magnitude of at least one of said first and second channels for said detected signal relative to said time window;

determining a second value related to a variance of said measure of correlation between said first and second channels relative to a normalized correlation value for said time window; and

using said first value and said second value to identify a sensor off condition.

22. The method of claim 21, wherein said step of defining a measure of correlation comprises identifying at least one feature within a plethysmographic waveform for said first and second channels.

23. The method of claim 21, wherein said step of monitoring comprises determining a plurality of correlation values between said channels over said time window.

24. The method of claim 23, wherein said step of determining a second value related to variance further comprises determining a variance between at least a first correlation value and a reference value for said plurality of correlation values for said time window.

25. The method of claim 24, wherein said step of determining a second value comprises statistical comparison of said plurality of correlation values to a mean value of said plurality of correlation values for said time window.

26. The method of claim 21, wherein said step determining a first values comprises measuring an intensity of said detected signal.

27. The method of claim 21, wherein said first value is utilized to select a threshold for comparison with said second value.