Automatic detection and reporting of cardiac asystole

• US 6,304,773 B1
• Filed: 05/21/1999
• Issued: 10/16/2001
• Est. Priority Date: 05/21/1998
• Status: Expired due to Term

• Current Assignee: Physio-Control Incorporated (Stryker Corporation)
• Original Assignee: Medtronic Physio-Control Manufacturing Corp. (Medtronic Plc)
• Inventors: Taylor, James W., Stickney, Ronald E.
• Primary Examiner(s): Schaetzle, Kennedy
• Application Number:
  US09/316,468
• Time in Patent Office: 879 Days
• Field of Search: 600/515-518, 607/5, 128/920, 128/923, 128/925
• US Class Current: 600/515
• CPC Class Codes: A61N 1/3904 External heart defibrillators [EHD]
• CPC Class Codes: A61N 1/39044 in combination with cardiopulmonary resuscitation [CPR] therapy

Abstract

A medical device (e.g., an automated external defibrillator) automatically detects and reports cardiac asystole by first obtaining ECG data and calculating one or more ECG measures from the ECG data. The defibrillator evaluates the ECG data by classifying the ECG data into a class indicative of cardiac condition, wherein one class is indicative of cardiac asystole. If the defibrillator classifies the ECG data into the class indicative of cardiac asystole, the defibrillator reports the asystole classification on a display. The defibrillator may classify the ECG data into a rhythm class associated with a cardiac rhythm, such as asystole, and report the rhythm class of the ECG data on the display. The defibrillator may also suggest a procedure to undertake, such as a therapy (e.g., defibrillation for a shockable cardiac rhythm), based on the classification of the ECG data. The defibrillator may reserve reporting the detection of asystole until a predetermined number of ECG classifications over at least a predetermined period of time have resulted in an asystole classification. The defibrillator may also reserve reporting the detection of asystole until an ECG classification resulting in asystole has been performed outside of a predetermined period of time after application of a defibrillation pulse. A statistical binary classification and regression tree may be used to classify the ECG data according to cardiac rhythm. Other signal data, such as impedance or phonocardiographic signal data, may also be obtained and classified with the ECG data.

Claims

1. A method of evaluating ECG data to automatically detect and report cardiac asystole, comprising:

(a) obtaining ECG data;

(b) calculating one or more ECG measures from the ECG data;

(c) classifying the ECG data in multiple classifiers using the one or more ECG measures, wherein each classifier in the multiple classifiers classifies the ECG data into a class indicative of cardiac condition, and wherein one class indicative of cardiac condition is indicative of cardiac asystole; and

(d) reporting an asystole classification if the ECG data is classified in the class indicative of cardiac asystole.

2. The method of claim 1, wherein the one or more ECG measures includes a measure in the group consisting of average amplitude, average frequency, baseline content, R-wave count, and ECG rate.

3. The method of claim 1, wherein one of the multiple classifiers is a statistical binary classification and regression tree.

4. The method of claim 1, wherein a first classifier of the multiple classifiers classifies the ECG data into a rhythm class associated with a cardiac rhythm.

5. The method of claim 4, further comprising reporting the rhythm class into which the ECG data is classified.

6. The method of claim 4, wherein a second classifier of the multiple classifiers classifies the ECG data into a class associated with a shockable or non-shockable cardiac condition.

7. The method of claim 6, wherein the second classifier classifies the ECG data into a class associated with a shockable or non-shockable cardiac condition based on the rhythm class into which the first classifier classifies the ECG data.

8. The method of claim 6, further comprising setting a flag if the first classifier classifies the ECG data of a patient into a rhythm class indicative of cardiac perfusion, and if the second classifier classifies the ECG data of the patient into a class associated with a non-shockable cardiac condition, the flag indicating that on subsequent delivery of a defibrillation pulse to the patient, if the patient has previously received a defibrillation pulse, the subsequent defibrillation pulse should convey an amount of energy equal to the energy conveyed in the previous defibrillation pulse delivered most recently.

9. The method of claim 4, further comprising automatically prompting on a display a procedure to undertake based on the rhythm class into which the ECG data is classified.

10. The method of claim 9, wherein the procedure automatically prompted on the display is a therapy associated with the cardiac rhythm indicated by the rhythm class.

11. The method of claim 9, wherein the procedure is an instruction providing guidance for prioritizing the delivery of a therapy.

12. The method of claim 11, wherein the instruction includes providing CPR for a period of time followed by delivery of a defibrillation pulse if the rhythm class into which the ECG data is classified is associated with medium- to low-amplitude ventricular fibrillation, or providing immediate delivery of a defibrillation pulse if the rhythm class into which the ECG data is classified is associated with high-amplitude ventricular fibrillation.

13. The method of claim 9, further comprising grouping two or more rhythm classes to make a super class having an associated therapy, wherein the procedure automatically prompted on the display is a therapy based on the super class of the rhythm class into which the ECG data is classified.

14. The method of claim 1, wherein obtaining ECG data comprises obtaining a continuous stream of ECG data, and wherein classifying the ECG data is performed on the continuous stream of ECG data at one or more instances of time.

15. The method of claim 1, wherein obtaining ECG data comprises obtaining a segment of ECG data over an interval of time.

16. The method of claim 15, further comprising obtaining multiple segments of ECG data over different intervals of time and classifying each segment of the multiple segments of ECG data, wherein the step of reporting an asystole classification includes:

(a) determining an overall ECG classification based on the classification of each segment of the multiple segments of ECG data; and

(b) reporting the overall ECG classification if the overall ECG classification is indicative of cardiac asystole.

17. The method of claim 16, wherein the multiple segments of ECG data include three segments of ECG data, and the overall ECG classification is based on a consensus of the classification of two of the three segments.

18. The method of claim 1, further comprising recording the classification of the ECG data in a history of ECG classifications, and reporting an asystole classification only if the history of ECG classifications indicates that a predetermined number of ECG classifications over at least a predetermined period of time have resulted in an asystole classification.

19. The method of claim 1, wherein an asystole classification is not reported if a defibrillation pulse has been delivered within a predetermined period of time.

20. The method of claim 1, further comprising automatically charging one or more defibrillation capacitors to deliver a defibrillation pulse if the ECG data is classified in a class associated with a shockable cardiac condition.

21. The method of claim 1, further comprising obtaining additional signal data indicative of cardiac condition and calculating one or more signal measures from the additional signal data, the multiple classifiers using the one or more signal measures and the one or more ECG measures to classify the additional signal data and the ECG data into a class indicative of cardiac condition.

22. The method of claim 21, wherein the additional signal data is patient impedance data.

23. The method of claim 21, wherein the additional signal data is phonocardiographic data.

59. The method of claim 18, wherein the ECG classifications in the predetermined number of ECG classifications are consecutive ECG classifications.

60. The method of claim 59, wherein the predetermined period of time is at least four minutes.

24. A method of evaluating ECG data to automatically detect and report cardiac asystole, comprising:

(a) obtaining ECG data;

(b) calculating one or more ECG measures from the ECG data;

(c) classifying the ECG data into a plurality of classes including shockable, asystole, and non-asystole classes using the one or more ECG measures;

(d) recording the classification of the ECG data in a history of ECG classifications; and

(e) reporting an asystole classification only if the history of ECG classifications indicates that a predetermined number of ECG classifications over at least a predetermined period of time resulted in an asystole classification.

25. The method of claim 24, wherein the one or more ECG measures includes a measure in the group consisting of average amplitude, average frequency, baseline content, R-wave count, and ECG rate.

26. The method of claim 24, wherein a statistical binary classification and regression tree is used to classify the ECG data.

27. The method of claim 24, wherein the ECG data are classified into a rhythm class associated with a cardiac rhythm, and wherein the ECG data are classified into one of the shockable, asystole, or non-asystole classes depending on the rhythm class into which the ECG data is classified.

28. The method of claim 27, further comprising reporting the rhythm class into which the ECG data is classified.

29. The method of claim 27, further comprising automatically prompting on a display a procedure to undertake based on the rhythm class into which the ECG data is classified.

30. The method of claim 29, wherein the procedure automatically prompted on the display is a therapy associated with the cardiac rhythm indicated by the rhythm class.

31. The method of claim 29, wherein the procedure is an instruction providing guidance for prioritizing the delivery of a therapy.

32. The method of claim 31, wherein the instruction includes providing CPR for a period of time followed by delivery of a defibrillation pulse if the rhythm class into which the ECG data is classified is associated with medium- to low-amplitude ventricular fibrillation, or providing immediate delivery of a defibrillation pulse if the rhythm class into which the ECG data is classified is associated with high-amplitude ventricular fibrillation.

33. The method of claim 24, wherein obtaining ECG data comprises obtaining a continuous stream of ECG data, and wherein classifying the ECG data is performed on the continuous stream of ECG data at one or more instances of time.

34. The method of claim 24, wherein obtaining ECG data includes obtaining multiple segments of ECG data over different intervals of time, wherein classifying the ECG data includes classifying each segment of the multiple segments of ECG data, and wherein reporting an asystole classification includes:

(a) determining an overall ECG classification based on the classification of each segment of the multiple segments of ECG data; and

(b) reporting the overall ECG classification if the overall ECG classification is indicative of cardiac asystole.

35. The method of claim 34, wherein the multiple segments of ECG data include three segments of ECG data, and the overall ECG classification is based on a consensus of the classification of two of the three segments.

36. The method of claim 24, wherein an asystole classification is not reported if a defibrillation pulse has been delivered within a predetermined period of time.

37. The method of claim 24, further comprising automatically charging one or more defibrillation capacitors to deliver a defibrillation pulse if the ECG data is classified in the shockable class.

38. The method of claim 24, further comprising obtaining additional signal data indicative of cardiac condition and using the one or more ECG measures and one or more signal measures calculated from the additional signal data to classify the additional signal data and the ECG data.

39. The method of claim 38, wherein the additional signal data is patient impedance data.

40. The method of claim 38, wherein the additional signal data is phonocardiographic data.

61. The method of claim 24, wherein the ECG classifications in the predetermined number of ECG classifications are consecutive ECG classifications.

62. The method of claim 61, wherein the predetermined period of time is at least four minutes.

41. A medical device for automatically detecting and reporting cardiac asystole, comprising:

(a) a plurality of electrodes configured to detect ECG signals;

(b) a converter in signal communication with the plurality of electrodes for converting the ECG signals into ECG data;

(c) a processing unit in signal communication with the converter for receiving and automatically evaluating the ECG data, the processing unit calculating one or more ECG measures from the ECG data and using the one or more ECG measures to classify the ECG data as shockable or non-shockable, and if classified as non-shockable, then classify the ECG data as asystole or non-asystole; and

(d) a display in signal communication with the processing unit, the processing unit reporting a detection of asystole on the display if the ECG data is classified as asystole.

42. The medical device of claim 41, wherein the processing unit classifies the ECG data into a rhythm class associated with a cardiac rhythm, the processing unit further reporting the rhythm class into which the ECG data is classified and prompting a therapy on the display based on the rhythm class into which the ECG data is classified.

43. The medical device of claim 41, further comprising a memory in signal communication with the processing unit, wherein the memory stores a history of ECG classifications and the processing unit records the classification of the ECG data in the history of ECG classifications.

44. The medical device of claim 43, wherein the processing unit reports a detection of asystole only if the history of ECG classifications indicates that a predetermined number of ECG classifications over at least a predetermined period of time resulted in an asystole classification.

45. The medical device of claim 41, further including a defibrillation circuit in signal communication with the processing unit for delivering defibrillation pulses, wherein the processing unit does not report a detection of asystole if the defibrillation circuit has delivered a defibrillation pulse within a predetermined period of time.

46. The medical device of claim 45, wherein the defibrillation circuit includes one or more defibrillation capacitors and the processing unit automatically instructs the defibrillation circuit to charge the one or more defibrillation capacitors to deliver a defibrillation pulse if the processing unit classifies the ECG data as shockable.

47. The medical device of claim 45, wherein the medical device is an automated external defibrillator.

48. The medical device of claim 41, further comprising:

(a) one or more sensors configured to detect a signal that is indicative of cardiac condition other than an ECG signal; and

(b) a converter in signal communication with the one or more sensors for converting the signal indicative of cardiac condition into signal data;

wherein the processing unit receives and evaluates the signal data with the ECG data, the processing unit calculating one or more signal measures from the signal data and using the one or more signal measures and the one or more ECG measures to classify the signal data and ECG data.

49. The medical device of claim 48, wherein the signal indicative of cardiac condition is an impedance signal.

50. The medical device of claim 48, wherein the signal indicative of cardiac condition is a phonocardiographic signal.

63. The medical device of claim 44, wherein the ECG classifications in the predetermined number of ECG classifications are consecutive ECG classifications.

64. The medical device of claim 63, wherein the predetermined period of time is at least four minutes.

51. A medical device for automatically evaluating cardiac condition, comprising:

(a) a plurality of sensors configured to detect one or more signals from a patient indicative of cardiac condition;

(b) a converter in signal communication with the plurality of sensors for converting the one or more signals into signal data;

(c) a processing unit in signal communication with the converter for receiving and automatically evaluating the signal data by classifying the signal data into a class indicative of cardiac condition, the processing unit determining an instruction based on the classification of the signal data for guiding a first-responding caregiver in providing emergency triage and treatment of the patient without the benefit of expert human intervention;

(d) a memory in signal communication with the processing unit for receiving and recording the class into which the signal data is classified by the processing unit, the class being recorded in a history of ECG classifications in the memory; and

(e) a display in signal communication with the processing unit, the processing unit prompting the instruction on the display, and if the history of ECG classifications stored in the memory indicates that a predetermined number of ECG classifications over at least a predetermined period of time resulted in an asystole classification, the processing unit also prompting notification of asystole on the display.

52. The medical device of claim 51, wherein the medical device is an automated external defibrillator and the one or more signals are ECG signals.

53. The medical device of claim 52, wherein the processing unit classifies the signal data into a rhythm class indicative of a cardiac rhythm, and wherein the instruction is tailored according to the cardiac rhythm of the rhythm class into which the signal data is classified.

54. The medical device of claim 53, wherein the processing unit uses a binary classification and regression tree to classify the signal data into a rhythm class.

55. The medical device of claim 52, further comprising one or more defibrillation capacitors, wherein the processing unit automatically charges the one or more defibrillation capacitors for delivery of a defibrillation pulse if the signal data is classified in a class indicative of a shockable cardiac condition.

56. The medical device of claim 51, wherein the processing unit automatically prompts an asystole detection on the display if the signal data is classified into a class indicative of cardiac asystole.

57. The medical device of claim 51, wherein the ECG classifications in the predetermined number of ECG classifications are consecutive ECG classifications.

58. The medical device of claim 57, wherein the predetermined period of time is at least four minutes.