Browse Prior Art Database

Selective Filter for Electrocardiograph

IP.com Disclosure Number: IPCOM000061814D
Original Publication Date: 1986-Sep-01
Included in the Prior Art Database: 2005-Mar-09
Document File: 2 page(s) / 47K

Publishing Venue

IBM

Related People

Bonner, RE: AUTHOR

Abstract

This method proposes a means to detect and correct muscle tremor without distorting the electrocardiogram (ECG) signal. It proposes a non- linear filter designed to take advantage of the characteristics of an ECG. In the operation of ECGs the frequency spectra of the signal and noise have a large overlap. This is generally limited to specific events in the time domain manifestation of the signal. These include QRS complexes and sharp P and T waves. When a filter is designed to remove the effects of muscle tremors in the ECG, the objective is to avoid filtering the foregoing effects or to at least minimize filtering them. Generally the remainder of the ECG, the baseline, may be safely filtered since the signal-noise spectral overlap is almost zero in this area.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 71% of the total text.

Page 1 of 2

Selective Filter for Electrocardiograph

This method proposes a means to detect and correct muscle tremor without distorting the electrocardiogram (ECG) signal. It proposes a non- linear filter designed to take advantage of the characteristics of an ECG. In the operation of ECGs the frequency spectra of the signal and noise have a large overlap. This is generally limited to specific events in the time domain manifestation of the signal. These include QRS complexes and sharp P and T waves. When a filter is designed to remove the effects of muscle tremors in the ECG, the objective is to avoid filtering the foregoing effects or to at least minimize filtering them. Generally the remainder of the ECG, the baseline, may be safely filtered since the signal-noise spectral overlap is almost zero in this area. QRS complexes can usually be detected since this signal is generally much larger than the noise generated by muscle tremor. This detection is not too difficult when multiple recorded ECG leads are used. The T wave usually has minimal spectral overlap with the noise so it does not cause a particular problem. The P wave is the most difficult to record since it is small and often has enough high frequency to be affected by a muscle tremor filter. The basic problem is to develop a means of detecting a P wave (or any sharp wave) in the presence of moderate muscle tremor. The effects of the overlap are simulated in Fig. 1 which shows a pure muscle tremor in Fig. 1A and a P...