Devices and Methods for Real-Time Denoising of Electrocardiograms

• US 20150018702A1
• Filed: 07/10/2014
• Published: 01/15/2015
• Est. Priority Date: 07/10/2013
• Status: Active Grant

• Current Assignee: AliveCor Inc.
• Original Assignee: AliveCor Inc.
• Inventors: ALBERT, David E., GALLOWAY, Conner Daniel Cross, VALYS, Alexander Vainius, HUGHES, Nicholas Peter
• Granted Patent: US 9,247,911 B2
• Time in Patent Office: Days
• US Class Current: 600/523
• CPC Class Codes: A61B 5/25 Bioelectric electrodes therefor
• CPC Class Codes: A61B 5/316 Modalities, i.e. specific diagnostic methods
• CPC Class Codes: A61B 5/339 Displays specially adapted therefor
• CPC Class Codes: A61B 5/349 Detecting specific parameters of the electrocardiograph cycle
• CPC Class Codes: A61B 5/7203 for noise prevention, reduction or removal
• CPC Class Codes: A61B 5/7235 Details of waveform analysis A61B5/349 takes precedence

Abstract

Apparatuses and methods (including methods of using such apparatuses) for de-noising electrocardiograms (ECGs) by manually or automatically adjusting the amount of filtering of an ECG signal. For example, real-time ECG signals may be filtered by combining in a weighted fashion an unfiltered portion of an ECG (or a filtered portion of the same ECG) with the same portion of the ECG that has been filtered. The weighting may be adjusted manually and/or automatically. Also described herein are methods for real-time filtering of ECG signals using a combination of filtering techniques including filtering to correct baseline wander, Savitzky-Golay denoising, and threshold smoothing. Multiple filtering techniques may be combined in a weighed manner to provide signal de-noising.

Claims

1. A method of processing an electrocardiogram (ECG) signal of a user to provide improved readability of the ECG signal for a medical professional in diagnosing the ECG signal, the method comprising:

receiving an ECG signal; and

filtering the ECG signal as the ECG signal is received by applying a first filtering stage, a second filtering stage, and a third filtering stage to the ECG signal, wherein the first, second, and third filtering stages are different from one another.

2. The method of claim 1, further comprising displaying the filtered ECG signal in real-time as the ECG signal is received and filtered, wherein the displayed filtered ECG signal has improved readability such that a medical professional can better diagnose the filtered ECG signal.

3. The method of claim 1, wherein the first, second, and third filtering stages are performed in sequence.

4. The method of claim 1, wherein the first, second, and third filtering stages are performed in parallel.

5. The method of claim 1, wherein applying the first filtering stage comprises removing baseline wander of the ECG signal.

6. The method of claim 5, wherein removing the baseline wander comprises removing a moving average of the ECG signal from a portion of the ECG signal.

7. The method of claim 1, wherein applying the second filtering stage comprises removing high-frequency noise from the ECG signal.

8. The method of claim 7, wherein removing the high-frequency noise comprises applying a Savitzky-Golay de-noising filter to smooth the ECG signal.

9. The method of claim 8, wherein applying the Savitzky-Golay de-noising filter comprises applying a high order Savitzky-Golay filter to a QRS segment of the ECG signal and applying a low order Savitzky-Golay filter to a non-QRS segment of the ECG signal.

10. The method of claim 1, wherein applying the third filtering stage comprises removing low-amplitude, high frequency noise from the ECG signal.

11. The method of claim 10, wherein removing the low-amplitude, high frequency noise from the ECG signal comprises applying threshold fit smoothing to the ECG signal.

12. The method of claim 1, further comprising pre-processing the received ECG signal before filtering the ECG signal.

13. The method of claim 12, wherein pre-processing the received ECG signal comprises applying user-selected mains filter.

14. The method of claim 1, further comprising detecting an amount of noise of the received ECG signal.

15. The method of claim 14, further comprising varying an amount of filtering applied to the received ECG signal in response to the amount of noise detected.

16. The method of claim 14, wherein detecting the amount of noise comprises training a module of a computing device with a noise model.

17. The method of claim 1, further comprising varying an amount of filtering applied to the received ECG signal in response to user input.

18. A method of processing an electrocardiogram (ECG) signal of a user to provide improved readability of the ECG signal for a medical professional in diagnosing the ECG signal, the method comprising:

providing an ECG signal to a computing device;

filtering the provided ECG signal;

displaying the filtered ECG signal on a display of the computing device; and

varying an amount of filtering applied to the displayed ECG signal in response to user input, wherein the amount of filtering is varied in real-time as the ECG signal is displayed, and wherein the displayed filtered ECG signal has improved readability such that a medical professional can better diagnose the filtered ECG signals.

19. The method of claim 18, wherein providing the ECG signal comprises receiving, with the computing device, the ECG signal as the ECG signal is measured from an ECG sensor coupled to the computing device.

20. The method of claim 18, wherein providing the ECG signal comprises receiving, with the computing device, the ECG signal as the ECG signal is measured from an onboard sensor of the computing device.

21. The method of claim 18, wherein the ECG signal is displayed in real-time as the ECG signal is provided and filtered.

22. The method of claim 18, wherein filtering the displayed ECG signal comprises applying a first filtering stage, a second filtering stage, and a third filtering stage to the provided ECG signal, wherein the first, second, and third filtering stage are different from one another.

23. The method of claim 18, wherein the computing device comprises one or more of a smartphone, a tablet computer, a laptop computer, a personal computer, a personal digital assistant, or a wearable computer.

24. A method of processing an electrocardiogram (ECG) signal of a user, the method comprising:

receiving an ECG signal;

pre-processing the received ECG signal; and

filtering the pre-processed ECG signal by (i) removing baseline wander of the ECG signal, (ii) removing high frequency noise from the ECG signal, and (iii) removing low-amplitude, high frequency noise from the ECG signal, wherein steps (i) to (iii) are performed sequentially.