Low distortion ECG filter

• US 20030083583A1
• Filed: 12/09/2002
• Published: 05/01/2003
• Est. Priority Date: 07/20/2000
• Status: Active Grant

• Current Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Original Assignee: Cardiac Pacemakers Incorporated (Boston Scientific Corporation)
• Inventors: Kovtun, Vladimir V., Dodson, Randall, Bange, Joseph E.
• Granted Patent: US 6,785,573 B2
• Time in Patent Office: Days
• US Class Current: 600/509
• CPC Class Codes: A61B 5/30 Input circuits therefor
• CPC Class Codes: A61B 5/308 for electrocardiography [ECG]
• CPC Class Codes: A61B 5/725 using specific filters therefor, e.g. Kalman or adaptive filters A61B5/316 take precedence
• CPC Class Codes: Y10S 128/901 Suppression of noise in electric signal

Abstract

A low pass filter especially adapted for use with ECG equipment designed to negate any noise energy that may be superimposed upon the ECG signal itself. A plurality of low-pass FIR filter stages are connected in tandem between a signal input point and a signal output point where each of the stages has a distinct cut-off frequency characteristic. Means are provided for selectively turning ones of the plurality of stages on or off.

Claims

1. Signal processing apparatus comprising:

(a) a plurality of concatenated digital filter stages connected between a signal source and a signal output point, each of said filter stages including

(i) a time delay buffer operative to receive at an input digital data representing sampled values of a possibly noisy analog input signal at a predetermined sampling rate and for at least temporarily storing said sampled values for predetermined time intervals, said time delay buffer providing an output;

(ii) means for deriving from the output of the time delay buffer an average-over-time noise signal component falling within a predetermined frequency band, and

(iii) means for selectively subtracting the average-over-time noise signal component from the output of the time delay buffer to remove the noise component from the sampled values stored in the time delay buffer.

2. The signal processing apparatus of claim 1 wherein the means for deriving the average over-time noise signal component comprises:

(a) a high-pass filter connected to receive the output of the time delay buffer; and

(b) a dynamic noise level calculator operative to receive filtered data from the high-pass filter and averaging the filtered data over time.

3. The signal processing apparatus of claim 2 wherein the high-pass filter is a finite impulse response filter.

4. The signal processing apparatus of claim 3 wherein the high-pass filter has a predetermined number of taps corresponding to the delay time of the time delay buffer.

5. The signal processing apparatus of claim 2 and further including wherein means for setting the filtered data from the high-pass filter between upper and lower limits.

6. The signal processing apparatus of claim 1 and further including a smoothing filter connected to said signal output point of the plurality of digital filter stages.

7. The signal processing apparatus of claim 1 wherein each of said plurality of digital filter stages has a different cut-off frequency value.

8. The signal processing apparatus of claim 1 and further including a control for selected ones of the plurality of concatenated digital filter stages that selectively disables at least one of the plurality of stages.

9. The signal processing apparatus of claim 8 wherein the control is applied to the means for subtracting.

10. The signal processing apparatus of claim 1 wherein the plurality of digital filter stages are each implemented in software executable by a microcomputer.

11. The signal processing apparatus of claim 1 wherein the plurality of digital filter stages are each implemented in hardware components.

12. The apparatus of claim 1 wherein the signal source comprises digitized ECG waveforms.

13. A method of processing ECG signals comprising the steps of:

(a) applying amplified analog waveforms representative of ecg vectors to an analog-to-digital converter;

(b) obtaining digital values from the analog-to-digital converter representative of the amplitude of the ecg vector signals at discrete sampling times;

(c) storing the digital values in a multi-stage time delay buffer;

(d) computing from the stored digital values a scaled noise signal falling within a first predetermined frequency band;

(e) dynamically determining upper and lower limits for the scaled noise signal computed in step (d); and

(f) subtracting the signal computed in step (d) from a data sample stored in a predetermined stage of the time delay buffer to yield an output signal devoid of noise falling within said first predetermined frequency band.

14. The method of claim 13 and further including the steps of applying said output signal to a further time delay buffer and repeating steps (d), (e) and (f) on the stored digital values in the further time delay buffer, step (d) involves a second predetermined frequency band different from said first predetermined frequency band.