Method and apparatus for identifying cardiac risk

• US 9,008,762 B2
• Filed: 06/16/2014
• Issued: 04/14/2015
• Est. Priority Date: 11/03/2009
• Status: Active Grant

• Current Assignee: VivaQuant LLC
• Original Assignee: VivaQuant LLC
• Inventors: Brockway, Marina, Brockway, Brian
• Primary Examiner(s): Gedeon, Brian T
• Application Number:
  US14/305,927
• Publication Number:
  US 20140296726A1
• Time in Patent Office: 302 Days
• Field of Search: None
• US Class Current: 600/515
• CPC Class Codes: A61B 5/0006 ECG or EEG signals
• CPC Class Codes: A61B 5/0245 by using sensing means generating electric signals, i.e. ECG signals
• CPC Class Codes: A61B 5/259 using conductive adhesive means, e.g. gels
• CPC Class Codes: A61B 5/316 Modalities, i.e. specific diagnostic methods
• CPC Class Codes: A61B 5/349 Detecting specific parameters of the electrocardiograph cycle
• CPC Class Codes: A61B 5/7253 characterised by using transforms
• CPC Class Codes: A61B 7/00 Instruments for auscultation
• CPC Class Codes: G06F 17/14 Fourier, Walsh or analogous domain transformations , e.g. Laplace, Hilbert, Karhunen-Loeve, transforms for correlation function computation G06F17/156; spectrum analysers G01R23/16
• CPC Class Codes: G06F 2218/04 Denoising
• CPC Class Codes: G06F 2218/10 by analysing the shape of a waveform, e.g. extracting parameters relating to peaks
• CPC Class Codes: G16H 50/30 for calculating health indices; for individual health risk assessment
• CPC Class Codes: H03H 17/0248 Filters characterised by a particular frequency response or filtering method

Abstract

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.

Claims

1. A method comprising:

identifying a plurality of cardiac cycles in an electrical signal representative of an electrocardiogram (ECG) from a subject;

for each of the plurality of cardiac cycles,

identifying T-wave offset of the ECG, and

identifying a segment of an acoustical vibration representative of heart sounds from the subject, based upon a T-wave offset time of a corresponding ECG synchronized with the heart sounds;

constructing an array of the identified segments from each of the plurality of cycles;

computing heart sound and noise components of the acoustical vibration using blind source separation; and

detecting the presence of a heart sound based upon energy in the heart sound components and the noise components.

2. The method of claim 1, wherein detecting the presence of a heart sound includes detecting the presence of a S3 heart sound based upon characteristics of the energy in the heart sound and noise components being indicative of S3 heart sounds.

3. The method of claim 1, wherein computing heart sound and noise components of the acoustical vibration using blind source separation includes using at least one of:

principal component analysis;

eigenvalue decomposition; and

independent component analysis.

4. The method of claim 1, further including applying a time-frequency decomposition to one of the array of identified segments and the acoustical vibration, prior to computing the heart sound and noise components.

5. The method of claim 4, wherein applying a time-frequency decomposition includes using at least one of:

a wavelet related transform;

a Gabor transform;

a Fourier transform;

a discrete cosine transform; and

a filter bank.

6. The method of claim 1 wherein computing heart sound and noise components of the acoustical vibration using blind source separation includes preforming principal component analysis followed by independent component analysis.

7. The method of claim 1 wherein identifying the segment of the acoustical vibration based upon the T-wave offset time includes beginning each identified segment at T-wave offset.

8. The method of claim 1, further including identifying the approximate location of an S2 heart sound for at least one of the identified segments of the acoustical vibration and blanking at least a portion of said S2 heart sound, wherein detecting the presence of a heart sound includes detecting the presence of an S3 heart sound, based on the blanking of the at least a portion of the S2 heart sound.

9. The method of claim 1, wherein identifying the segment includes identifying a segment that terminates about 100 msec prior to a Q-wave onset of a subsequent cardiac cycle.

10. The method of claim 1, wherein identifying the segment includes detecting one of P-wave onset or P-wave peak of a subsequent cardiac cycle, and terminating the segment at the detected one of the P-wave onset or the P-wave peak.

11. The method of claim 1, further including computing the energy of the heart sounds as the root mean square of the computed heart sound components.

12. An apparatus comprising:

an input circuit;

a computer circuit configured and arranged with the input circuit to;

receive an electrical signal representative of an electrocardiogram (ECG) from a subject;

identify a plurality of cardiac cycles in the electrical signal;

for each of the plurality of cardiac cycles,

identify T-wave offset of the ECG, and

identify a segment of an acoustical vibration representative of heart sounds from the subject, based upon a T-wave offset time of a corresponding ECG synchronized with the heart sounds;

construct an array of the identified segments from each of the plurality of cycles;

compute heart sound and noise components of the acoustical vibration using blind source separation; and

detect the presence of a heart sound based upon energy in the heart sound components and the noise components.

13. The apparatus of claim 12, wherein the computer circuit is configured and arranged to detect the presence of a heart sound by detecting the presence of an S3 heart sound based upon characteristics of the energy in the heart sound and noise components being indicative of S3 heart sounds.

14. The apparatus of claim 12, wherein the computer circuit is configured and arranged to apply a time-frequency decomposition to one of the array of identified segments and the acoustical vibration, prior to computing the heart sound and noise components.

15. The apparatus of claim 12 wherein the computer circuit is configured and arranged to compute heart sound and noise components of the acoustical vibration using blind source separation by preforming principal component analysis followed by independent component analysis.

16. The apparatus of claim 12 wherein the computer circuit is configured and arranged to identify the segment of the acoustical vibration based upon the T-wave offset time by beginning each identified segment at T-wave offset.

17. The apparatus of claim 12, wherein the computer circuit is configured and arranged to:

identify the approximate location of an S2 heart sound for at least one of the identified segments of the acoustical vibration, and blank at least a portion of said S2 heart sound, and

detect the presence of a heart sound by detecting the presence of an S3 heart sound, based on the blanking of the at least a portion of the S2 heart sound.

18. The apparatus of claim 12, wherein the computer circuit is configured and arranged to identify the segment by identifying a segment that terminates about 100 msec prior to the Q-wave onset of the subsequent cardiac cycle.

19. The apparatus of claim 12, wherein the computer circuit is configured and arranged to identify the segment by detecting one of P-wave onset or P-wave peak of a subsequent cardiac cycle, and terminating the segment at the detected one of the P-wave onset or the P-wave peak.

20. An apparatus comprising:

means for identifying a plurality of cardiac cycles in an electrical signal representative of an electrocardiogram (ECG) from a subject;

means for, for each of the plurality of cardiac cycles,

identifying T-wave offset of the ECG, and

identifying a segment of an acoustical vibration representative of heart sounds from the subject, based upon a T-wave offset time of a corresponding ECG synchronized with the heart sounds;

means for constructing an array of the identified segments from each of the plurality of cycles;

means for computing heart sound and noise components of the acoustical vibration using blind source separation; and

means for detecting the presence of a heart sound based upon energy in the heart sound components and the noise components.