Dual channel P-wave detection in surface electrocardiographs

• US 4,793,361 A
• Filed: 03/13/1987
• Issued: 12/27/1988
• Est. Priority Date: 03/13/1987
• Status: Expired due to Term

• Current Assignee: Cardiac Pacemakers Incorporated 4100 North Hamline AVE St Paul MN
• Original Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Inventors: DuFault, Robert A.
• Primary Examiner(s): Jaworski, Francis J.
• Assistant Examiner(s): MANUEL, GEORGE C
• Application Number:
  US07/025,731
• Time in Patent Office: 655 Days
• Field of Search: 128/696, 128/704, 128/705, 128/419 PG
• US Class Current: 600/509
• CPC Class Codes: A61B 5/30 Input circuits therefor
• CPC Class Codes: A61B 5/349 Detecting specific parameters of the electrocardiograph cycle

Abstract

A method and apparatus for reliably detecting the P-wave component in a surface electrocardiograph. An adaptive Least Means Square filter is implemented in a programmed computer and the computer receives as its inputs a digitized version of the ECG waves picked up from two discrete locations on a subject'"'"'s chest wall. The LMS adaptive filter functions to accommodate the energy component of the QRS complex thereby leaving the P-wave as the paramount signal component in the filter'"'"'s "error" output.

Claims

1. A system for detecting P-waves in ECG signals obtained from chest contacting surface electrodes, comprising:

(a) first and second surface electrodes individually coupled to first and second differential amplifying means for obtaining time varying ECG signals from two separate locations on the chest wall of a patient, the output of first differential amplifying means comprising a "reference" waveform;

(b) adaptive filter means having means for receiving an "input" signal and a "desired" signal and having "error" signal feedback means and "estimate" output means;

(c) means coupling the output of said first differential amplifying means to said means for receiving a "desired" signal;

(d) means coupling the output of said second differential amplifying means to said means for receiving an "input" signal;

(e) summing means connected to receive said "desired" signal and coupled to said "estimate" output means of said adaptive filter means for producing an "error" signal proportional to the difference between said "desired" signal and said "estimate" output means of said adaptive filter means; and

(f) means coupling said "error" signal feedback means to said adaptive filter means for varying the characteristics of said adaptive filter means whereby the energy content of said "error" signal due to the QRS content of said ECG signals is minimized while the energy content due to the P-wave is enhanced.

2. The system as in claim 1 wherein said adaptive filter means is a least means square filter.

3. The system as in claim 2 wherein said adaptive filter means includes a plurality of variable weighting elements which are updated in accordance with the expression:

space="preserve" listing-type="equation">W.sub.k+1 =W.sub.k +2μ

ε

.sub.k X.sub.K

where k is a time index;

ε

_k is said error signal at time k;

X_k is said output signal vector from said second differential amplifier means at time k;

μ

is an adaptation time constant;

W_k is the weight vector value at time k; and

W_k+1 is the weight vector value at a time later than k following the preceding updating sequence.

4. A method of enhancing the P-wave component of the ECG PQRST complex comprising the steps of:

(a) attaching at least two electrodes to a patient'"'"'s skin surface at predetermined locations;

(b) processing the electrical signals arising from said electrodes by filtering and amplification;

(c) applying the processed electrical signals to an analog-to-digital converting means for digitizing said electrical signals at a predetermined sampling rate;

(d) introducing the digitized electrical signals into a digital computer preprogrammed to execute the LMS algorithm, such that the QRS component of the ECG PQRST complex forming the "error" output of said LMS algorithm is substantially reduced while the P-wave component thereof remains substantially unattenuated; and

(e) comparing said P-wave component in said "error" output to a predetermined reference level.

5. The method as in claim 4 and further including the step of filtering said "error" output of the LMS algorithm to suppress relatively high frequency components therefrom prior to said comparing step.