System and method for biventricular pacemaker pulse detection in surface ECG

• US 10,085,661 B2
• Filed: 04/16/2014
• Issued: 10/02/2018
• Est. Priority Date: 04/16/2013
• Status: Active Grant

• Current Assignee: Koninklijke Philips N.V.
• Original Assignee: Koninklijke Philips N.V.
• Inventors: Chien, Cheng-hao
• Primary Examiner(s): Wu, Eugene T
• Application Number:
  US14/784,594
• Publication Number:
  US 20160058317A1
• Time in Patent Office: 1,630 Days
• CPC Class Codes: A61B 5/316 Modalities, i.e. specific diagnostic methods
• CPC Class Codes: A61B 5/318 Heart-related electrical modalities, e.g. electrocardiography [ECG]
• CPC Class Codes: A61B 5/341 Vectorcardiography [VCG]
• CPC Class Codes: A61B 5/349 Detecting specific parameters of the electrocardiograph cycle
• CPC Class Codes: A61B 5/7217 of noise originating from a therapeutic or surgical apparatus, e.g. from a pacemaker
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

A system for locating the existence of a biventricular pace pulse for ECG, which includes a non-biventricular pulse detector for finding the onsets of ventricular pulses, a lead-wise biventricular pulse detector for checking whether there are two separated ventricular pulses in one heartbeat, and a vector-based biventricular pulse detector for determining the existence of the biventricular pace pulse if the lead-wise biventricular pulse detector does not find that there are two separated ventricular pulses in one heartbeat.

Claims

1. A method for locating an existence of a biventricular pace pulse for electrocardiograph (ECG), comprising:

transforming ECG data to a 3-dimensional VectorCardioGraph (VCG);

calculating VCG distance features comprising spatial distance and spatial velocity;

locating a first ventricular pulse;

calculating VCG angle features including angles between vectors and the first ventricular pulse;

determining the existence of the biventricular pace pulse by classifying vectors based on spatial distance, spatial velocity, VCG angle features and temporal distances of vectors to the first ventricular pulse, wherein;

an impulse response of a low-pass filter is rejected from being classified as the biventricular pace pulse based on the spatial distances and the temporal distances of the vectors to the first ventricular pulse;

a recharging wave is rejected from being classified as the biventricular pace pulse based on the angles between the vectors and the first ventricular pulse and temporal distances of the vectors to the first ventricular pulse; and

the biventricular pace pulse is determined to exist if an eligible pulse vector exists after performing said rejections.

2. The method of claim 1, wherein the first ventricular pulse is located by determining a vector having a largest spatial distance out of all vectors that have a spatial velocity that exceeds a spatial velocity threshold.

3. The method of claim 1, wherein the ECG data includes two adjacent ventricular pulses per heartbeat.

4. The method of claim 1, wherein the ECG data includes two partially overlapping ventricular pulses per heartbeat.

5. The method of claim 1, further comprising displaying single beat ECG waveforms of the ECG data.

10. The detector of claim 1, wherein the processor is configured to identify the first ventricular pulse by determining a vector having a largest spatial distance out of all vectors that have a spatial velocity that exceeds a spatial velocity threshold.

6. A system for locating an existence of a biventricular pace pulse for electrocardiograph (ECG), comprising:

a non-biventricular pulse detector configured to find onsets of one or more ventricular pulses;

a lead-wise biventricular pulse detector for determining whether there are two separated ventricular pulses in one heart beat; and

a vector-based biventricular pulse detector configured to determine the existence of the biventricular pace pulse if the lead-wise biventricular pulse detector does not determine that there are two separated ventricular pulses in one heartbeat,

wherein the vector-based biventricular pulse detector is configured to;

calculate VectorCardioGraph (VCG) distance features comprising spatial distance and spatial velocity,

find a first ventricular pulse,

calculate VCG angle features including angles between vectors and the first ventricular pulse, and

determine the existence of the biventricular pace pulse by classifying vectors based on spatial distance, spatial velocity, VCG angle features and temporal distances of vectors to the first ventricular pulse, wherein the vector-based biventricular pulse detector is configured to;

reject an impulse response of a low-pass filter from being classified as the biventricular pace pulse based on the spatial distances and the temporal distances of the vectors to the first ventricular pulse;

reject a recharging wave from being classified as the biventricular pace pulse based on the angles between the vectors and the first ventricular pulse and temporal distances of the vectors to the first ventricular pulse; and

determine that the biventricular pace pulse exists if an eligible pulse vector exists after performing said rejections.

7. The system of claim 6, wherein the vector-based biventricular pulse detector is configured to identify the first ventricular pulse by determining a vector having a largest spatial distance out of all vectors that have a spatial velocity that exceeds a spatial velocity threshold.

8. The system of claim 6, wherein the system has a sampling rate that is less than 5000 samples per second.

9. A vector-based biventricular pace pulse detector, comprising:

a processor configured to;

calculate VectorCardioGraph (VCG) distance features comprising spatial distance and spatial velocity,

find a first ventricular pulse,

calculate VCG angle features including angles between vectors and the first ventricular pulse, and

determine an existence of a biventricular pace pulse by classifying vectors based on spatial distance, spatial velocity, VCG angle features and temporal distances of vectors to the first ventricular pulse, wherein the processor is configured to;

reject an impulse response of a low-pass filter from being classified as the biventricular pace pulse based on the spatial distances and the temporal distances of the vectors to the first ventricular pulse;

reject a recharging wave from being classified as the biventricular pace pulse based on the angles between the vectors and the first ventricular pulse and temporal distances of the vectors to the first ventricular pulse; and

determine that the biventricular pace pulse exists if an eligible pulse vector exists after performing said rejections.

11. The detector of claim 9, further comprising an input configured to receive electrocardiograph (ECG) data.

12. The detector of claim 11, wherein the processor is further configured to transform the ECG data to a 3-dimensional VCG.

13. The detector of claim 9, wherein the detector has a sampling rate that is less than 5000 samples per second.