Electrocardiogram signal detection

Patent metadata

Field	Value
Publication number	US-12318209-B2
Application number	US-202117342387-A
Country	US
Kind code	B2
Filing date	Jun 8, 2021
Priority date	Nov 8, 2012
Publication date	Jun 3, 2025
Grant date	Jun 3, 2025

Abstract

Apparatuses and methods for extracting, de-noising, and analyzing electrocardiogram signals. Any of the apparatuses described herein may be implemented as a (or as part of a) computerized system. For example, described herein are apparatuses and methods of using them or performing the methods, for extracting and/or de-noising ECG signals from a starting signal. Also described herein are apparatuses and methods for analyzing an ECG signal, for example, to generate one or more indicators or markers of cardiac fitness, including in particular indicators of atrial fibrillation. Described herein are apparatuses and method for determining if a patient is experiencing a cardiac event, such as an arrhythmia.

First claim

What is claimed is:

1. A method comprising:
identifying a set of sub-regions within an electrocardiogram (ECG) signal;
cross-correlating each sub-region of the set of sub-regions with each other sub-region in the set of sub-regions;
determining, by a processor, two or more sub-regions of the set of sub-regions that are correlated based on the cross-correlating, the two or more sub-regions that are correlated corresponding to sub-regions that are characteristic of the ECG signal; and
differentially processing the two or more sub-regions of the ECG signal to generate a de-noised ECG signal.

2. The method of claim 1 , wherein differentially processing the two or more sub-regions comprises:
removing the two or more sub-regions from the ECG signal;
filtering the two or more sub-regions using a first filtering regime;
filtering the remaining sub-regions of the ECG signal using a second filtering regime; and
adding the filtered two or more sub-regions to the filtered remaining sub-regions of the ECG signal to generate the de-noised ECG signal.

3. The method of claim 1 , wherein cross-correlating a first sub-region of the set of sub-regions with a second sub-region of the set of sub-regions comprises:
comparing the first and second sub-regions and determining a location of maximum correlation between the first and second sub-regions, wherein the first and second sub-regions are correlated if the maximum correlation between the first and second sub-regions exceeds a correlation threshold.

4. The method of claim 3 , wherein determining the two or more sub-regions comprises:
identifying from among the set of sub-regions, a longest sequence of consecutive, correlated sub-regions;
identifying each sub-region outside of the longest sequence that correlates with a threshold number of sub-regions from the longest sequence, wherein the two or more sub-regions correspond to the longest sequence of sub-regions and each sub-region outside of the longest sequence that is correlated with the threshold number of sub-regions from the longest sequence.

5. The method of claim 1 , further comprising:
generating a correlation matrix indicating a level of correlation of each sub-region of the set of sub-regions with each other sub-region in the set of sub-regions.

6. The method of claim 2 , wherein the first filtering regime comprises Principal Component Analysis (PCA), and the second filtering regime comprises polynomial filtering.

7. The method of claim 1 , wherein the set of sub-regions may comprise one of a set of P-waves, a set of R-waves, a set of Q-waves, a set of S-waves, a set of T-waves, a set of QRS regions, a set of P-R intervals, and a set of R-R intervals.

8. The method of claim 1 , further comprising pre-filtering the ECG signal before identifying the set of sub-regions in the ECG signal.

9. The method of claim 8 , wherein pre-filtering the ECG signal comprises performing wavelet filtering on the ECG signal identifying the set of sub-regions.

10. The method of claim 1 , further comprising determining if the de-noised ECG signal is indicative of a heart condition.

11. An apparatus comprising:
a set of electrodes to measure an electrocardiogram (ECG) signal;
a memory; and
a processor operatively coupled to the memory, the processor to:
identify a set of sub-regions within the ECG signal;
cross-correlate each sub-region of the set of sub-regions with each other sub-region in the set of sub-regions;
determine two or more sub-regions of the set of sub-regions that are correlated based on the cross-correlating, the two or more sub-regions that are correlated corresponding to sub-region that are characteristic of the ECG signal; and
differentially process the two or more sub-regions of the ECG signal to generate a de-noised ECG signal.

12. The apparatus of claim 11 , wherein to differentially process the two or more sub-regions, the processor is to:
remove the two or more sub-regions from the ECG signal;
filter the two or more sub-regions using a first filtering regime;
filter the remaining sub-regions of the ECG signal using a second filtering regime; and
add the filtered two or more sub-regions to the filtered remaining sub-regions of the ECG signal to generate the de-noised ECG signal.

13. The apparatus of claim 11 , wherein to cross-correlate a first sub-region of the set of sub-regions with a second sub-region of the set of sub-regions, the processor is to:
compare the first and second sub-regions and determining a location of maximum correlation between the first and second sub-regions, wherein the first and second sub-regions are correlated if the maximum correlation between the first and second sub-regions exceeds a correlation threshold.

14. The apparatus of claim 13 , wherein to determine the two or more sub-regions, the processor is to:
identify from among the set of sub-regions, a longest sequence of consecutive, correlated sub-regions;
identify each sub-region outside of the longest sequence that correlates with a threshold number of sub-regions from the longest sequence, wherein the two or more sub-regions correspond to the longest sequence of sub-regions and each sub-region outside of the longest sequence that is correlated with the threshold number of sub-regions from the longest sequence.

15. The apparatus of claim 11 , wherein the processor is further to:
generate a correlation matrix indicating a level of correlation of each sub-region of the set of sub-regions with each other sub-region in the set of sub-regions.

16. The apparatus of claim 12 , wherein the first filtering regime comprises Principal Component Analysis (PCA), and the second filtering regime comprises polynomial filtering.

17. The apparatus of claim 11 , wherein the set of sub-regions may comprise one of a set of P-waves, a set of R-waves, a set of Q-waves, a set of S-waves, a set of T-waves, a set of QRS regions, a set of P-R intervals, and a set of R-R intervals.

18. The apparatus of claim 11 , wherein the processor is further to pre-filter the ECG signal before identifying the set of sub-regions in the ECG signal.

19. The apparatus of claim 18 , wherein to pre-filter the ECG signal, the processor is to perform wavelet filtering on the ECG signal identifying the set of sub-regions.

20. The apparatus of claim 11 , wherein the processor is further to determine if the de-noised ECG signal is indicative of a heart condition.

Assignees

• Alivecor Inc

Inventors

• Galloway Conner Daniel Cross
• Albert David E

Classifications

• A61B5/358

  Detecting ST segments · CPC title

• A61B5/353

  Detecting P-waves · CPC title

• A61B5/36

  Detecting PQ interval, PR interval or QT interval · CPC title

• A61B5/02438

  with portable devices, e.g. worn by the patient · CPC title

• A61B5/0006

  ECG or EEG signals · CPC title