System and method of impedance cardiography and heartbeat determination

• US 5,309,917 A
• Filed: 02/12/1992
• Issued: 05/10/1994
• Est. Priority Date: 09/12/1991
• Status: Expired due to Term

• Current Assignee: Seemedx Incorporated
• Original Assignee: Drexel University
• Inventors: Wang, Xiang, Sun, Hun H.
• Primary Examiner(s): Jaworski, Francis
• Assistant Examiner(s): JASTRZAB, JEFFREY R
• Application Number:
  US07/834,425
• Time in Patent Office: 818 Days
• Field of Search: 128/670, 128/693, 128/696, 128/697, 128/700, 128/702, 128/713, 128/734
• US Class Current: 600/508
• CPC Class Codes: A61B 5/0205 Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
• CPC Class Codes: A61B 5/0295 using plethysmography, i.e. measuring the variations in the volume of a body part as modified by the circulation of blood therethrough, e.g. impedance plethysmography
• CPC Class Codes: A61B 5/0535 Impedance plethysmography for measuring blood flow A61B5/0295
• CPC Class Codes: A61B 5/316 Modalities, i.e. specific diagnostic methods
• CPC Class Codes: A61B 5/346 Analysis of electrocardiograms
• CPC Class Codes: A61B 5/347 Detecting the frequency distribution of signals
• CPC Class Codes: A61B 5/7239 using differentiation including higher order derivatives
• CPC Class Codes: A61B 5/7257 using Fourier transforms

Abstract

Cardiac monitoring is disclosed in which thoracic impedance and EKG signals are gathered and processed for improved resolution and accuracy. EKG signals are adaptively processed by digitizing, filtering, differentiating and raising the resultant differential by a power greater than one to emphasize changes in the slope of the EKG signal. Blocks of the thus processed EKG data are analyzed to identify peak amplitude and to compare spacing between peak amplitude adaptively to more accurately identify R wave peaks. Stroke volume is determined from a thoracic impedance signal and its time derivative. Preferably, a time-frequency distribution is taken of the time derivative thoracic impedance signal after low- and high-pass filtering to identify B and X wave events in the signal which are used to determine ventricular ejection time and dz/dt_min for a determination of heart stroke volume by conventional methods. Alternatively, stroke volume is determined by a new relationship between a product of a pair of impedances simultaneously sensed on opposing sides of a patient'"'"'s heart at the peak of a heartbeat.

Claims

1. A method of processing a time-derivative, thoracic impedance signal generated from a patient to identify events in the time-derivative impedance signal associated with beats of the patient'"'"'s heart, the method comprising the steps of:

generating a time-frequency distribution of data from the time-derivative impedance signal for a selected portion of the time-derivative impedance signal spanning a single heartbeat of the patient; and

identifying a time of occurrence of at least one cardiac event in the selected portion of said time-derivative impedance signal from the time-frequency distribution.

2. The method of claim 1 wherein the identifying step comprises the steps of:

determining a maximum amplitude value of the time-frequency distribution; and

identifying an amplitude value of the time-frequency distribution at least as great as a predetermined fractional value of the maximum amplitude value.

3. The method of claim 2 wherein the amplitude value identifying step comprises the steps of:

initially identifying a time of occurrence of a maximum amplitude value in the selected portion of the time-derivative impedance signal; and

identifying the first amplitude value of the time-frequency distribution at least as great as the predetermined fractional value occurring in the time-frequency distribution most immediately preceding the time of occurrence of the maximum amplitude value of the time-derivative impedance signal.

4. The method of claim 3 further comprising the steps of determining stroke volume of the patient during the single heartbeat from the difference between the time of occurrence of the first amplitude value of the time-frequency distribution and the time of occurrence of the maximum amplitude value of the time-derivative signal.

5. The method of claim 3 further comprising the step of identifying a second amplitude value of the time-frequency distribution at least as great as the predetermined fractional value and following the time of occurrence of the maximum amplitude value of the time-derivative signal by at least a predetermined time period.

6. The method of claim 5 further comprising the step of determining stroke volume of the patient during the single heartbeat from the difference between the time of occurrence of the first amplitude value of the time-frequency distribution and the time of occurrence of the second amplitude value of the time-frequency distribution.

7. The method of claim 2 wherein the amplitude value identifying step comprises the steps of:

initially identifying a time of occurrence of a maximum amplitude value in the selected portion of the time-derivative impedance signal; and

identifying a second amplitude value of the time-frequency distribution at least as great as the predetermined fractional value and following the time of occurrence of the maximum amplitude value of the time-derivative signal by at least a predetermined time period.

8. The method of claim 1 wherein the step of generating the time-frequency distribution comprises generating a Cohen derived time-frequency distribution of the selected portion of the time-derivative signal.

9. The method of claim 8 wherein the step of generating the time-frequency distribution further comprises generating the distribution with a window function.

10. The method of claim 1 further comprising the initial steps of:

differentiating an electrocardiogram signal from the patient with respect to time;

scaling the differentiated electrocardiogram signal nonlinearly to emphasize amplitude peaks in the differentiated electrocardiogram signal;

identifying a maximum amplitude value of the scaled signal occurring in an interval including several consecutive seconds of the scaled electrocardiogram signal;

identifying each amplitude value of the scaled signal in the selected interval at least as great as a predetermined fraction of the maximum amplitude value; and

generating the time-frequency distribution from a portion of the time-derivative signal spanning the time of occurrence of only one of the identified amplitude values of the scaled signal.

11. The method of claim 10 further comprising the step of transmitting a bi-level timing signal spanning the selected interval of the scaled signal and including level transitions corresponding in time to the occurrence of each of the identified amplitude values of the scaled signal during the selected interval.

12. The method of claim 10 further comprising the step of filtering the electrocardiogram signal to remove low frequency components of the signal from patient movement and high frequency components of the signal from sources of external electrical interference before the differentiating step.

13. The method of claim 12 wherein the step of filtering comprises passing data of the electrocardiogram signal through a high-pass filter to eliminate frequency components of about five hertz or less from the signal data.

14. The method of claim 13 wherein the step of filtering further comprises passing the high-pass filtered data through a low-pass filter to at least eliminate components of about fifty hertz or more from the high-pass filtered data.

15. The method of claim 13 wherein the step of passing the high-pass filtered data further comprises passing said filtered data through a low-pass filter to at least eliminate components of about forty hertz or more from the high-pass filtered data.

16. The method of claim 12 wherein the electrocardiogram signal data is digitized and the filtering step comprises passing the digitized electrocardiogram data through an all-integer coefficient filter.

17. The method of claim 10 further comprising the steps of:

generating a coefficient of variation for the amplitude values of the scaled signal identified from the selected interval;

comparing the coefficient of variation to a predetermined value; and

transmitting a bi-level timing signal spanning the selected interval and including level transitions corresponding in time to the time of occurrence of each of the identified amplitude values of the scaled signal during the selected interval only if the coefficient of variation for the interval is at least less than a predetermined fractional value.

18. The method of claim 1 further comprising the step of initially filtering the time-derivative signal to remove low frequency components from patient movements and higher frequency components from sources of electrical interference external to the patient and wherein the generating step comprises generating the time-frequency distribution from the filtered time-derivative signal.

19. The method of claim 18 wherein the filtering step comprises removing frequency components of about 5 hertz or less and about 50 hertz or more from the time-derivative impedance signal.

20. The method of claim 1 wherein the identifying step comprises the steps of:

integrating the time-frequency distribution of data over a range of frequencies associated with cardiac events occurring during the beating of the patient'"'"'s heart;

determining a maximum amplitude value of the integrated time-frequency distribution associated with the selected portion of the time-derivative impedance signal; and

identifying an amplitude value of the integrated time-frequency distribution associated with the selected portion of the time derivative impedance signal, the identified amplitude being at least as great as a predetermined fractional value of the maximum amplitude value.

21. The method of claim 20 wherein the amplitude value identifying step comprises the steps of:

initially identifying a time of occurrence of a maximum amplitude value in the selected portion of the time-derivative impedance signal; and

identifying a first amplitude value of the integrated time-frequency distribution at least as great as the predetermined fractional value, which occurs in the integrated time-frequency distribution most immediately preceding in time the occurrence of the maximum amplitude value identified in the selected portion of the time-derivative impedance signal.

22. The method of claim 21 further comprising the steps of determining stroke volume of the patient during the single heartbeat from the difference between the time of occurrence of the first amplitude value of the integrated time-frequency distribution and the time of occurrence of the maximum amplitude value of the time-derivative signal.

23. The method of claim 22 further comprising the step of identifying a second amplitude value of the integrated time-frequency distribution at least as great as the predetermined fractional value and following the time of occurrence of the maximum amplitude value of the time-derivative signal by at least a predetermined time period.

24. The method of claim 23 further comprising the step of determining stroke volume of the patient during the single heartbeat from the difference between the time of occurrence of the first amplitude value of the integrated time-frequency distribution and the time of occurrence of the second amplitude value of the integrated time-frequency distribution.

25. The method of claim 20 wherein the amplitude value identifying step comprises the steps of:

initially identifying a time of occurrence of a maximum amplitude value in the selected portion of the time-derivative impedance signal; and

identifying a second amplitude value of the integrated time-frequency distribution, the second amplitude value being at least as great as the predetermined fractional value and the second amplitude value following the time of occurrence of the maximum amplitude value in the selected portion of the time-derivative impedance signal by at least a predetermined time period.

26. The method of claim 20 wherein the integrating step comprises the step of combining contributions to the time-frequency distribution of data in a range of about 30 Hz to about 55 Hz.

27. A method of determining stroke volume of a patient'"'"'s heart from thoracic impedance measurements comprising the steps of:

applying a fluctuating excitation current to the patient through a first pair of electrodes located on opposing upper and lower sides of the patient'"'"'s heart;

generating from a second pair of electrodes located on opposing sides of the patients'"'"'s heart between the first pair of electrodes a first signal related to impedance of the patient at a first electrode of the second pair and a second signal related to impedance of the patient at a second electrode of the second pair;

identifying an impedance from each of the first and second signals occurring simultaneously with one heartbeat of the patient; and

multiplying together the identified impedances of each of the first and second signals to estimate a cardiac parameter at least related to stroke volume of the patient'"'"'s heart during the one heartbeat.

28. The method of claim 27 further comprising the step of taking the difference of the two identified impedances and dividing the difference by a product of the two identified impedances to estimate the stroke volume.

29. A method of determining stroke volume of a patient'"'"'s heart from the thoracic impedance measurements comprising the steps of:

applying a fluctuating excitation current to the patient through a first pair of electrodes located on opposing upper and lower sides of the patient'"'"'s heart;

generating from a second pair of electrodes located on opposing sides of the patients'"'"'s heart between the first pair of electrodes, a first signal indicative of absolute impedance of the patient at only the first electrode of the second pair and a second signal indicative of absolute impedance of the patient at only the second electrode of the second pair;

identifying two individual values, one from each of the first and second signals, indicating two absolute impedances occurring at one instant at the first and second electrodes of the second pair during one heartbeat of the patient; and

multiplying together the two identified individual values to generate a product value proportional to each of the two absolute impedances at the one instant to estimate a cardiac parameter at least related to stroke volume of the patient'"'"'s heart during the one heartbeat.

30. In a system for processing a time-derivative thoracic impedance signal generated from a patient to identify events in the time-derivative signal associated with the beats of the patient'"'"'s heart, the improvement comprising:

memory means for at least temporarily storing a time-derivative thoracic impedance signal and processor means coupled with the memory means for generating a time-frequency distribution of data from the time-derivative thoracic impedance signal for a selected portion of the time-derivative thoracic impedance signal spanning a single heartbeat of the patient and for identifying a time of occurrence of at least one cardiac event in the selected portion of the time-derivative thoracic impedance signal from the time-frequency distribution.