Method and apparatus for diagnosing sleep breathing disorders

• US 5,769,084 A
• Filed: 07/10/1996
• Issued: 06/23/1998
• Est. Priority Date: 07/10/1996
• Status: Expired due to Term

• Current Assignee: The United States of America As Represented By The Secretary of Agriculture
• Original Assignee: the united states of america as represented by the secretary of the navy
• Inventors: Newman, A. Kief, Katz, Richard A., Lawee, Michael S.
• Primary Examiner(s): Getzow, Scott M.
• Application Number:
  US08/687,098
• Time in Patent Office: 713 Days
• Field of Search: 128/700, 128/702, 128/699, 128/716
• US Class Current: 600/513
• CPC Class Codes: A61B 5/113 occurring during breathing
• CPC Class Codes: A61B 5/4818 Sleep apnoea

Abstract

An apparatus and method for identifying the timing of the onset of and duion of an event characteristic of sleep breathing disorder. Chaotic processing techniques analyze data concerning one or more cardio-respiratory functions such as nasal air flow, chest wall effort, oxygen saturation, heart rate and heart activity. Excursions of the resulting signal beyond a threshold provide markers for the timing of such an event that is useful in the diagnosis of obstructed sleep apnea and other respiratory dysfunctions.

Claims

1. A method for generating markers identifying the timing of the onset of an event characteristic of a sleep breathing disorder including the steps of:

monitoring at least one cardio-respiratory function over time;

generating a time series vector representation of each monitored cardio-respiratory function;

generating a signal for each monitored function based upon chaotic processing of the corresponding time series vector representation; and

identifying as a marker each excursion of each signal beyond a corresponding threshold value thereby to indicate the timing of the onset of an event.

2. A method as recited in claim 1 wherein said monitoring includes monitoring first and second cardio-respiratory functions and said signal generating step includes generating a chaotic radius signal and a chaotic differential radius signal based upon chaotic and differential processing of the time series vector representations of the first and second cardio-respiratory functions, respectively.

3. A method as recited in claim 2 wherein the first monitored cardio-respiratory function is heart action as measured by an electrocardiograph and said monitoring, generating and identifying steps act on the signal from the electrocardiograph.

4. A method as recited in claim 2 wherein the second cardio-respiratory function is taken from the group of cardio-respiratory functions consisting of oral-nasal air flow, oxygen saturation, chest wall impedance and heart rate and said monitoring, generating and identifying steps act on at least one of each of the second functions.

5. A method as recited in claim 4 wherein the first monitored cardio-respiratory function is heart action as measured by an electrocardiograph and said monitoring, generating and identifying steps act on at least one of each of the first and second functions.

6. A method as recited in claim 2 wherein said generation of the differential radius signal includes:

generating an embedding delay value based upon the time series vector representation;

comparing the magnitudes of the terms of the vector representation at a given time and at a time delayed by the embedding delay value to obtain a chaotic radius; and

generating in response to each value of the chaotic radius the differential chaotic radius for the given time.

7. A method as recited in claim 6 wherein said step of determining an embedding delay includes:

converting a monitored function into a time series of samples;

generating a vector time delay interval in response to the data in the time series of samples; and

generating the time series vector representation based upon the value of the data in the time series of samples at intervals corresponding to the vector time delay interval.

8. A method as recited in claim 7 wherein said step of generating the vector time delay interval includes the step of obtaining average mutual information from the time series of samples.

9. A method as recited in claim 7 wherein said generation of the embedding delay value includes the step of obtaining global false nearest neighbor information from the time series vector representation.

10. A method as recited in claim 7 wherein said step of generating the differential chaotic radius includes comparing, at times corresponding to each vector time delay interval, the magnitude of the time series vector representation at that time and the magnitude of the time series vector representation at time offset by the embedding delay value.

11. A method as recited in claim 7 wherein:

the sampling frequency at which said time series representation is generated for each monitored function is greater than the greatest frequency of physiologic relevance with respect to the monitored cardio respiratory functions;

the vector time delay is an integer multiple of the sampling interval; and

the embedding function is an integer multiple of the vector time delay.

12. A method as recited in claim 11 wherein the sampling frequency is at least 40 Hz.

13. Apparatus for generating markers identifying the timing of the onset of an event characteristic of a sleep breathing disorder including:

monitoring means for monitoring at least one cardio-respiratory function over time;

first means for generating a time series vector representation of each monitored cardio-respiratory function;

second means for generating a signal for each monitored function based upon chaotic processing of the corresponding time series vector representation; and

marker identifying means for identifying as a marker each excursion of each signal from said second generating means beyond a corresponding threshold value thereby to indicate the timing of the onset of an event.

14. Apparatus as recited in claim 13 wherein said monitoring means includes means for monitoring first and second cardio-respiratory functions and said second generating means includes means for generating a chaotic radius signal and a chaotic differential radius signal based upon chaotic and differential processing of the time series vector representations of the first and second cardio-respiratory functions, respectively.

15. Apparatus as recited in claim 14 wherein said monitoring means includes an electrocardiograph for monitoring, as a first monitored cardio-respiratory function, heart action and said monitoring, first and second generating means and said marker identifying means act on the signal from the electrocardiograph.

16. Apparatus as recited in claim 14 wherein said monitoring means includes at least one of a group of cardio-respiratory function monitors consisting of thermistor means for monitoring oral-nasal air flow, finger pulse oximetry monitoring means for monitoring oxygen saturation, chest wall monitoring means for measuring chest wall impedance and heart rate monitoring means for monitoring heart rate and wherein first and second generating means and said marker identifying means act on signals from at least one of each of said group of monitoring means.

17. Apparatus as recited in claim 16 wherein said monitoring means includes an electrocardiograph for monitoring as a first monitored cardio-respiratory function heart action and said first and second generating means and said marker identifying means act on at least one of each of the first and second functions.

18. Apparatus as recited in claim 14 wherein said second generation means includes:

means for generating an embedding delay value based upon the time series vector representation;

means for comparing the magnitudes of the terms of the vector representation at a given time and at a time delayed by the embedding delay value to obtain a chaotic radius; and

means for generating, in response to each value of the chaotic radius the differential chaotic radius for the given time.

19. Apparatus as recited in claim 18 wherein said embedding delay determining means includes:

means for converting a monitored function into a time series of samples;

means for generating a vector time delay interval in response to the data in the time series of samples; and

means for generating the time series vector representation based upon the value of the data in the time series of samples at intervals corresponding to the vector time delay interval.

20. Apparatus as recited in claim 19 wherein said means for generating the vector time delay interval includes means for obtaining average mutual information from the time series of samples.

21. Apparatus as recited in claim 19 wherein said generation of the embedding delay value includes the step of obtaining global false nearest neighbor information from the time series vector representation.

22. Apparatus as recited in claim 19 wherein said means for generating the differential chaotic radius includes means for comparing, at times corresponding to each vector time delay interval, the magnitude of the time series vector representation at that time and the magnitude of the time series vector representation at time offset by the embedding delay value.

23. Apparatus as recited in claim 19 wherein said first generating means includes means for establishing a sampling frequency at which said time series representation is generated for each monitored function that is greater than the greatest frequency of physiologic relevance with respect to the monitored cardio respiratory functions and wherein the vector time delay is an integer multiple of the sampling interval and the embedding function is an integer multiple of the vector time delay.

24. Apparatus as recited in claim 19 wherein said first generating means includes means for establishing a sampling frequency of at least 40 Hz and wherein the vector time delay is an integer multiple of the sampling interval and the embedding function is an integer multiple of the vector time delay.