System and method for detecting respiratory insufficiency in the breathing of a subject

• US 8,660,971 B2
• Filed: 10/09/2009
• Issued: 02/25/2014
• Est. Priority Date: 10/15/2008
• Status: Active Grant

• Current Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Original Assignee: Koninklijke Philips N.V.
• Inventors: Orr, Joseph Allen
• Primary Examiner(s): Chaki, Kakali
• Assistant Examiner(s): RIOJAS RAMIREZ, MARIO
• Application Number:
  US13/121,002
• Publication Number:
  US 20110178968A1
• Time in Patent Office: 1,600 Days
• Field of Search: None
• US Class Current: 706/20
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

Respiratory insufficiency is detected by classifying preliminary breaths identified through a capnogram as being valid or artifact. Individual breaths are classified as being valid or artifact by determining values of a plurality of breathing parameters for a given breath, inferring a value for a key parameter from the determined values for the plurality of breathing parameters, and comparing the inferred value for the key parameter to a predetermined threshold.

Claims

1. A system configured to detect respiratory insufficiency, the system comprising:

(a) one or more processors configured to obtain information related to the breathing of a subject, and to execute one or more modules, the one or more modules comprising;

(1) a parameter module configured to determine parameters of the breathing of the subject based on the information obtained by the processor, wherein the parameters are determined on a per preliminary breath basis, wherein the preliminary breath basis refers to individual breaths and not to a quality of the breaths, but only a level of analysis in identifying respiration,

(2) a key parameter module configured to infer a value for a key parameter of the breathing of the subject on a per preliminary breath basis from the parameters determined by the parameter module, wherein the key parameter is a parameter not directly derivable from the information obtained by the processor, further wherein the key parameter module includes an artificial neural network that models a relationship between a predetermined set of breathing parameters and the key parameter, wherein responsive to inputs of the breathing parameters for a given preliminary breath, the artificial neural network outputs the value for the key parameter for the given preliminary breath, and

(3) a classification module configured to detect respiratory insufficiency in the breathing of the subject by (i) comparing the inferred value of the key parameter for a given preliminary breath to a predetermined threshold, (ii) classifying, in response to the comparison, individual preliminary breaths as being (ii)(a) valid, productive breaths or (ii)(b) artifact, deficient breaths, respectively, and (iii) responsive to the classification of individual preliminary breaths, determining an overall respiratory insufficiency, wherein determining the overall respiratory insufficiency includes determining a respiratory rate to convey sufficiency of respiration of the subject, wherein (iii)(a) responsive to detection of a valid, productive breath, updating a calculation of respiratory rate to reflect the detected valid, productive breath, and (iii)(b) responsive to detection of an artifact, deficient breath, not updating the calculation of respiratory rate until a next valid, productive breath is detected, wherein the parameters determined by the parameter module include values determined by aggregating predetermined sets of samples of CO₂ levels during a given preliminary breath, wherein (i) a first parameter is determined from a first set that comprises a first number of samples of the level of CO₂ during the given preliminary breath, (ii) another parameter is determined from a next set that comprises a next number of samples of the level of CO₂ during the same given preliminary breath, and (iii) responsive to the given preliminary breath being a short breath not providing enough sample points to provide data from which each of the parameter values can be determined, the parameter module (iii)(a) automatically assigns preset values to those sample points that are not provided because of the short breath and (iii)(b) determines a corresponding parameter with use of the assigned values.

2. The system of claim 1, wherein the classification module detects respiratory insufficiency in the breathing of the subject if the inferred value of the key parameter crosses the threshold.

3. The system of claim 1, wherein the key parameter includes one or more of expired volume, peak respiratory flow, maximum airway pressure, or chest wall movement.

4. The system of claim 1, wherein the predetermined threshold is configurable by a user.

5. The system of claim 1, wherein the parameters determined by the parameter module include one or more of a time ratio between the time of inspiration and the time of expiration, a fraction of inspiration time during which the level of CO₂ is some predetermined amount less than the maximum CO₂ level during a breath, a difference between the maximum partial pressure of CO₂ observed during a breath and the partial pressure of CO₂ observed a predetermined period of time prior to the point in time at which the maximum was observed, a fraction of expiration time during a breath when a capnogram is within 10% of the maximum CO₂ detected during the breath, a standard deviation of capnogram samples taken during inspiration, or a mean of capnogram samples taken during inspiration.

6. The system of claim 1, wherein the information obtained by the processor includes output signals generated by one or more sensors in communication with gas provided to and/or received from the airway of the subject.

7. A method of detecting respiratory insufficiency, the method comprising:

obtaining information related to the breathing of a subject;

determining parameters of the breathing of the subject based on the obtained information, wherein the parameters are determined on a per preliminary breath basis, wherein the preliminary breath basis refers to individual breaths and not to a quality of the breaths, but only a level of analysis in identifying respiration;

inferring a value for a key parameter of the breathing of the subject on a per preliminary breath basis from the determined parameters, wherein the key parameter is a parameter not directly derivable from the obtained information related to the breathing of the subject, further wherein inferring the key parameter includes using an artificial neural network that models a relationship between a predetermined set of breathing parameters and the key parameter, wherein responsive to inputs of the breathing parameters for a given preliminary breath, the artificial neural network outputs the value for the key parameter for the given preliminary breath; and

detecting respiratory insufficiency in the breathing of the subject by (i) comparing the inferred value of the key parameter for a given preliminary breath to a predetermined threshold, (ii) classifying, in response to the comparison, individual preliminary breaths as being (ii)(a) valid, productive breaths or (ii)(b) artifact, deficient breaths, respectively, and (iii) responsive to the classification of individual preliminary breaths, determining an overall respiratory insufficiency, wherein determining the overall respiratory insufficiency includes determining a respiratory rate to convey sufficiency of respiration of the subject, wherein (iii)(a) responsive to detection of a valid, productive breath, updating a calculation of respiratory rate to reflect the detected valid, productive breath, and (iii)(b) responsive to detection of an artifact, deficient breath, not updating the calculation of respiratory rate until a next valid, productive breath is detected, wherein the parameters determined from the obtained information related to the breathing of the subject include values determined by aggregating predetermined sets of samples of CO₂ levels during a given preliminary breath wherein (i) a first parameter is determined from a first set that comprises a first number of samples of the level of CO₂ during the given preliminary breath (ii) another parameter is determined from a next set that comprises a next number of samples of the level of CO₂ during the same given preliminary breath, and (iii) responsive to the given preliminary breath being a short breath not providing enough sample points to provide data from which each of the parameter values can be determined, (iii)(a) automatically assigning preset values to those sample points that are not provided because of the short breath and (iii)(b) determining a corresponding parameter with use of the assigned values.

8. The method of claim 7, wherein respiratory insufficiency in the breathing of the subject is detected if the inferred value of the key parameter crosses the threshold.

9. The method of claim 7, wherein the key parameter includes one or more of expired volume, peak respiratory flow, maximum airway pressure, or chest wall movement.

10. The method of claim 7, wherein the predetermined threshold is configurable by a user.

11. The method of claim 7, wherein the parameters determined from the obtained information related to the breathing of the subject include one or more of a time ratio between the time of inspiration and the time of expiration, a fraction of inspiration time during which the level of CO₂ is some predetermined amount less than the maximum CO₂ level during a breath, a difference between the maximum partial pressure of CO₂ observed during a breath and the partial pressure of CO₂ observed a predetermined period of time prior to the point in time at which the maximum was observed, a fraction of expiration time during a breath when a capnogram is within 10% of the maximum CO₂ detected during the breath, a standard deviation of capnogram samples taken during inspiration, or a mean of capnogram samples taken during inspiration.

12. The method of claim 7, wherein the obtained information related to the breathing of the subject includes output signals generated by one or more sensors in communication with gas provided to and/or received from the airway of the subject.

13. A system configured to detect respiratory insufficiency, the system comprising:

means for obtaining information related to the breathing of a subject;

means for determining parameters of the breathing of the subject based on the obtained information, wherein the parameters are determined on a per preliminary breath basis, wherein the preliminary breath basis refers to individual breaths and not to a quality of the breaths, but only a level of analysis in identifying respiration;

means for inferring a value for a key parameter of the breathing of the subject on a per preliminary breath basis from the determined parameters, wherein the key parameter is a parameter not directly derivable from the obtained information related to the breathing of the subject, further wherein inferring the key parameter includes using an artificial neural network that models a relationship between a predetermined set of breathing parameters and the key parameter, wherein responsive to inputs of the breathing parameters for a given preliminary breath, the artificial neural network outputs the value for the key parameter for the given preliminary breath; and

means for detecting respiratory insufficiency in the breathing of the subject by (i) comparing the inferred value of the key parameter for a given preliminary breath to a predetermined threshold, (ii) classifying, in response to the comparison, individual preliminary breaths as being (ii)(a) valid, productive breaths or (ii)(b) artifact, deficient breaths, respectively, and (iii) responsive to the classification of individual preliminary breaths, determining an overall respiratory insufficiency, wherein determining the overall respiratory insufficiency includes determining a respiratory rate to convey sufficiency of respiration of the subject, wherein (iii)(a) responsive to detection of a valid, productive breath, updating a calculation of respiratory rate to reflect the detected valid, productive breath, and (iii)(b) responsive to detection of an artifact, deficient breath, not updating the calculation of respiratory rate until a next valid, productive breath is detected, wherein the parameters determined from the obtained information related to the breathing of the subject include values determined by aggregating predetermined sets of samples of CO₂ levels during a given preliminary breath wherein (i) a first parameter is determined from a first set that comprises a first number of samples of the level of CO₂ during the given preliminary breath, (ii) another parameter is determined from a next set that comprises a next number of samples of the level of CO₂ during the same given preliminary breath, and (iii) responsive to the given preliminary breath being a short breath not providing enough sample points to provide data from which each of the parameter values can be determined, (iii)(a) automatically assigning preset values to those sample points that are not provided because of the short breath and (iii)(b) determining a corresponding parameter with use of the assigned values.

14. The system of claim 13, wherein respiratory insufficiency in the breathing of the subject is detected if the inferred value of the key parameter crosses the threshold.

15. The system of claim 13, wherein the key parameter includes one or more of expired volume peak respiratory flow, maximum airway pressure, or chest wall movement.

16. The system of claim 13, wherein the parameters determined obtained information related to the breathing of the subject include one or more of a time ratio between the time of inspiration and the time of expiration, a fraction of inspiration time during which the level of CO₂ is some predetermined amount less than the maximum CO₂ level during a breath, a difference between the maximum partial pressure of CO₂ observed during a breath and the partial pressure of CO₂ observed a predetermined period of time prior to the point in time at which the maximum was observed, a fraction of expiration time during a breath when a capnogram is within 10% of the maximum CO₂ detected during the breath, a standard deviation of capnogram samples taken during inspiration, or a mean of capnogram samples taken during inspiration.

17. The system of claim 13, wherein the obtained information related to the breathing of the subject includes output signals generated by one or more sensors in communication with gas provided to and/or received from the airway of the subject.