Method and apparatus for identifying cardiac risk
A cardiac-based metric is computed based upon characteristics of a subject's cardiac function. In accordance with one or more embodiments, the end of a mechanical systole is identified for each of a plurality of cardiac cycles of a subject, based upon an acoustical vibration associated with closure of an aortic valve during the cardiac cycle. The end of an electrical systole of an electrocardiogram (ECG) signal for each cardiac cycle is also identified. A cardiac-based metric is computed, based upon a time difference between the end of the electrical systole and the end of the mechanical systole, for the respective cardiac cycles.
1. A method for computing a cardiac-based metric for use as a predictor of cardiac risk, the method comprising:
for each of a plurality of cardiac cycles of a subject, identifying the end of a mechanical systole based upon an acoustical vibration associated with closure of an aortic valve during the cardiac cycle; and
computing the cardiac-based metric based upon, for each of the plurality of cardiac cycles, a time difference between the end of the mechanical systole and the end of an electrical systole of an electrocardiogram (ECG) signal for the cardiac cycle.
2. The method of claim 1 , wherein identifying the end of the mechanical systole includes, in a circuit, processing electronic heart sound data for the cardiac cycle to identify the end of the mechanical systole based on characteristics of the heart sound data.
3. The method of claim 2 , further including identifying the end of the electrical systole by, in a circuit, processing electronic ECG signal data for the cardiac cycle to identify the end of the electrical systole based on characteristics of the ECG signal data.
4. The method of claim 1 , wherein identifying the end of the mechanical systole includes identifying the end of a mechanical systole using a heart sound signal obtained from the subject during the cardiac cycle.
5. The method of claim 1 further including identifying the end of the electrical systole by identifying a T-wave offset for the ECG signal.
6. The method of claim 1 , wherein computing the cardiac-based metric includes computing at least one of:
at least one of a mean and median of the values of the time difference for each cardiac cycle;
at least one of variance, standard deviation, and standard error of the time difference for each cardiac cycle; and
a short-term instability metric by computing one of the mean, standard deviation and root mean square of successive differences between beat-to-beat values in a window segment including the plurality of cardiac cycles.
7. The method of claim 1 , wherein computing the cardiac-based metric includes computing
a short-term instability metric by computing one of the mean, standard deviation and root mean square of successive differences between beat-to-beat values in a window segment including the plurality of cardiac cycles, and
a long-term instability metric based on a combination of variance in the beat-to-beat values and the computed short-term instability metric.
8. The method of claim 1 , wherein identifying the end of a mechanical systole includes
detecting the closure of the aortic valve using an acoustical signal containing energy associated with closure of the aortic valve and noise energy,
decomposing the acoustical signal from a first domain into subcomponents of the acoustical signal in a second domain,
identifying at least two of the subcomponents that respectively exhibit an energy level of which at least half is associated with closure of the aortic valve, and
mathematically combining the identified subcomponents to compute a time function that identifiably changes in value upon aortic valve closure.
9. The method of claim 1 , wherein the ECG signal includes an ECG component originating from heart tissue of the subject and a noise component, further including identifying the end of the electrical systole by:
decomposing the ECG signal from a first domain into subcomponents of the ECG signal in a second domain,
identifying the location of a QRS complex of the cardiac cycle based upon a spatial distribution of the subcomponents,
identifying a first time window in the cardiac cycle that includes the QRS complex,
identifying at least one additional time window in the cardiac cycle that does not include the QRS complex,
for each time window, identify subcomponents having more energy corresponding to the ECG component than noise energy, and
constructing a denoised ECG in the first domain by combining the identified subcomponents, and using the denoised ECG to identify the end of the electrical systole.
10. The method of claim 1 , further including identifying the end of the electrical systole by:
decomposing the ECG signal from a first domain into subcomponents of the ECG signal in a second domain,
identifying the location of a QRS complex of the cardiac cycle based upon a spatial distribution of the subcomponents,
establishing a T-wave offset search window relative to the location of the QRS complex,
identifying at least two subcomponents having an energy value that is predominantly energy of a T-wave of the cardiac cycle,
mathematically combining the at least two identified subcomponents to compute an emphasis signal having an identifiable inflection corresponding to a location of the T-wave offset,
identifying the T-wave offset location based upon a characteristic of the emphasis signal, and
using the T-wave offset location to identify the end of the electrical systole.
11. An apparatus for computing a cardiac-based metric for use as a predictor of cardiac risk comprising:
a first module configured and arranged to, for each of a plurality of cardiac cycles of a subject, identify the end of a mechanical systole based upon an acoustical vibration associated with closure of an aortic valve during the cardiac cycle; and
a second module configured and arranged to compute a cardiac-based metric based upon, for each of the plurality of cardiac cycles, a time difference between the end of the mechanical systole and the end of an electrical systole of an electrocardiogram (ECG) signal for the cardiac cycle.
12. The apparatus of claim 11 , further including
a sensor configured and arranged to convert mechanical vibrations generated by heart sounds of the subject into an electrical heart sound signal, the first module being configured and arranged to receive the heart sound signal and to identify the end of the mechanical systole based upon the acoustical vibration using an acoustical vibration indicated via the received heart sound signal, and
at least two electrodes configured and arranged to sense the ECG signal from the subject's heart, the second module being configured and arranged to receive the ECG signal.
13. The apparatus of claim 11 , further including a third module configured and arranged to identify the end of the electrical systole of the electrocardiogram (ECG) signal for the cardiac cycle.
14. The apparatus of claim 13 , wherein the third module is configured and arranged to identify the end of the electrical systole by identifying a T-wave offset for the ECG signal.
15. The apparatus of claim 13 , wherein the ECG signal includes an ECG component originating from heart tissue of the subject and a noise component, wherein the third module is configured and arranged to identify the end of the electrical systole by:
decomposing the ECG signal from a first domain into subcomponents of the ECG signal in a second domain,
identifying the location of a QRS complex of the cardiac cycle based upon a spatial distribution of the subcomponents,
identifying a first time window in the cardiac cycle that includes the QRS complex,
identifying at least one additional time window in the cardiac cycle that does not include the QRS complex,
for each time window, identify subcomponents having more energy corresponding to the ECG component than noise energy,
constructing a denoised ECG in the first domain by combining the identified subcomponents, and
identifying a T-wave offset in the denoised ECG as the end of the electrical systole.
16. The apparatus of claim 13 , wherein the third module is configured and arranged to identify the end of the electrical systole by:
decomposing the ECG signal from a first domain into subcomponents of the ECG signal in a second domain,
identifying the location of a QRS complex of the cardiac cycle based upon a spatial distribution of the subcomponents,
establishing a T-wave offset search window relative to the location of the QRS complex,
identifying at least two subcomponents having an energy value that is predominantly energy of a T-wave of the cardiac cycle,
mathematically combining the at least two identified subcomponents to compute an emphasis signal having an identifiable inflection corresponding to a location of the T-wave offset,
identifying the T-wave offset location based upon a characteristic of said emphasis signal, and
using the T-wave offset location to identify the end of the electrical systole.
17. The apparatus of claim 11 , wherein the second module is configured and arranged to compute the cardiac-based metric by computing at least one of:
at least one of a mean, a median, variance, standard deviation, and standard error of the values of the time difference for the cardiac cycles; and
a short-term instability metric by computing at least one of the mean, standard deviation and root mean square of successive differences between beat-to-beat values in a window segment including the plurality of cardiac cycles, the beat-to-beat values corresponding to heartbeats that define a successive start of the cardiac cycles.
18. The apparatus of claim 17 , wherein the second module is configured and arranged to compute the cardiac-based metric by
computing a short-term instability metric based on at least one of the mean, standard deviation and root mean square of successive differences between beat-to-beat values in a window segment including the plurality of cardiac cycles, the beat-to-beat values corresponding to heartbeats that define a successive start of the cardiac cycles, and
computing a long-term instability metric by computing a variance in the beat-to-beat values, multiplying the computed variance by two and subtracting the computed short term instability metric.
19. The apparatus of claim 11 , wherein the first module is configured and arranged to identify the end of a mechanical systole by
detecting the closure of the aortic valve using an acoustical signal containing energy associated with closure of the aortic valve and noise energy,
decomposing the acoustical signal from a first domain into subcomponents of the acoustical signal in a second domain,
identifying at least two of the subcomponents exhibiting an energy level of which at least half is associated with closure of the aortic valve, and
mathematically combining the identified subcomponents to compute a time function that identifiably changes in value upon aortic valve closure.
20. A method comprising:
for each of a plurality of cardiac cycles of a subject, identifying the end of a mechanical systole based upon an acoustical vibration associated with closure of an aortic valve during the cardiac cycle; and
computing a cardiac-based metric indicative of arrhythmic risk by:
computing beat-to-beat values of a subject's cardiac function based on, for each of the plurality of cardiac cycles, a time difference between the end of the mechanical systole and the end of an electrical systole of an electrocardiogram (ECG) signal for the cardiac cycle,
computing a time series for each of successive lags of the beat-to-beat values by applying a low-pass filter and decimator to the beat-to-beat values, and
evaluating the arrhythmic risk based upon a trend of the variability of the time series.