Method And Apparatus For Producing Sample Electrocardiograms

Abstract

An electronic instrument for medical use measures the time intervals between QRS complex portions of the electrical signal wave representing heartbeats. The intervals are quantized in digital form. The occurence of each subinterval is stored. Every incoming time interval is compared with the stored values. The instrument, upon receiving and comparing a novel interval, for example, either too short or too long, will activate an electrocardiograph recorder. The instrument produces a relatively small number of electrocardiographic (ECG) records, in the form of paper strips, but those strips are likely to contain a history of arrhythmic heart activity.

Description

BACKGROUND OF THE INVENTION

This invention relates to the instrument, analysis and processing by means of an electronic instrument of heart activity, and more particularly to the field of electrocardiography.

DESCRIPTION OF THE PRIOR ART

It is possible to detect and amplify voltages associated with the contraction of heart muscles by means of electrodes attached to a patient. The voltages are amplified and the voltage waveforms displayed by a pen marking device on a strip of paper (electrocardiograph).

It is considered good medical practice to continuously monitor a patient's electrocardiographic waveform (ECG) for a minimum of several days following a heart attack. Continuous monitoring is also used during operations and during periods of intensive care. It is believed that continuous monitoring for as little as 30 minutes gives considerably more information on arrhythmias than the standard ECG which may take less than 1 minute. The monitoring of ECG is particularly useful in persons who have had coronaries and are on antirhythmic drug therapy. Long term records (hours or days) of the ECG have shown value in predicting coronary prone persons, and in spotting brief periods of arrhythmias, which sometimes are potentially fatal, and sometimes are responsible for such symptoms as dizziness or light-headedness. Long term ECG records are sometimes recorded on audio magnetic tape for subsequent analysis, usually at higher than normal "playback" speeds.

At the standard ECG speed of 25mm/sec, a continuous record would use 90 meters of paper/hour. This is clearly impractical. As an alternative, an oscilloscope display is used in operating rooms, coronary care and intensive care units. An oscilloscope display is evanescent and does not furnish a document for more leisurely study or as a patient record. For such "permanent" records, such as ECG paper strips of magnetic tapes, it is desirable to obtain from the vast amount of possible records only a few samples. These samples should include representations of "critical portions" such as the occurrence of a premature beat, a missed beat (heart block), a run of tachycardia, etc.

The selection of critical portion samples is presently accomplished by having a specially trained nurse, who more or less continually observes the oscilloscope display, turn on the ECG recorder when a section of possible interest occurs in the displayed waveform. The ECG signal to the recorder is delayed from 2 to 20 seconds, with respect to the oscilloscope display, so that the event can be recorded after it has occurred. When the long term monitoring is by means of tape recording, the analyst reviewing the tape usually searches for and writes out, using a pen recorder and paper strip, those samples he deems important to include in the summary record.

The desirability of automatically monitoring and printing out appropriate sample records is generally recognized. The continual 24-hour observation of an oscilloscope by highly trained persons is expensive and limits such monitoring to the very high risk patients. Visual monitoring is also usually less than perfect due to lapses of attention and fatigue effect. This is particularly true with the common practice of having one observer simultaneously monitor four to eight oscilloscope displays. When reviewing tape recorded data during high speed playback (usually 60 times normal speed) only very competent persons are able to spot short intervals of waveforms representing heart arrhythmia, which intervals may be displayed for only a fraction of a second.

An automatic sample read-out system should not give an overabundance of samples of the same thing or run continuously. A number of circuits have been proposed to detect ectopic (premature) heart beats based on beat-to-beat (R-R interval) timing or which operate on waveform morphology. An ectopic beat detector can signal a recorder to turn on and if the recorder is fed with the delayed ECG waveform the record of the ectopic beat will be automatically captured. However, that system has the disadvantage that in some persons ectopics occur frequently -- six to 15 occurrences a minute is not uncommon, so that the recorder would run almost continuously.

OBJECTIVES OF THE INVENTION

The objectives of the present invention in which the beat-to-beat intervals of the patient are compared to a set of such stored intervals so that only novel intervals will operate the recorder for a short fixed time, are:

a. to provide an instrument and method which will detect arrhythmic heart beats in a patient and which will initiate the ECG recording of the patient upon such detection;

b. to provide such an instrument and method that will not initiate an overabundance of ECG recordings on frequent premature heartbeats;

c. to provide such an instrument and method which does not need constant attention by trained or other personnel but will operate entirely automatically;

d. to provide such an instrument and method which will provide a set of samples initiated by arrhythmic heart activity, which samples are likely to provide information of interest to medical personnel;

e. to provide such an instrument and method which provides a permanent record, which may be retained as part of the patient's record, and which is relatively small in size and yet likely to be a history of those portions of heart activity of particular interest.

SUMMARY OF THE INVENTION

The instrument of the present invention works with the conventional electrodes to detect heart activity. The voltages are amplified and the intervals between successive beats that have occurred area measured, that is, the R-R interval is measured. The set of intervals that have occurred is stored in a memory unit. The value of each incoming interval is compared to the stored set.

When an interval occurs which is not yet included in the stored set, for example, too long an interval due to a missed heartbeat, the instrument initiates the operation of a conventional ECG recorder which then runs for a predetermined time period. The ECG to the recorder is appropriately delayed so that the inscribed record includes the novel R-R interval. The new R-R interval value is then added to the stored set so that future occurences of the value will not initiate the recorder operation.

BRIEF DESCRIPTION OF THE DRAWINGS

FIGS. 1a-1e and FIGS. 4a and 4b are ECG waveforms showing the intervals between heart beats, and the range of intervals resulting from those waveforms;

FIGS. 2a and 2c are block diagrams of the circuit of the instrument of the present invention;

FIG. 2b is a timing diagram illustrating the comparison function which takes place to either activate or not activate the ECG recorder; and

FIGS. 3a-3e depict the state of the storage device; and

FIGS. 5a and 5b are histograms of heart activity.

DESCRIPTION OF THE PREFERRED EMBODIMENT

In conventional medical terminology, the first small deflection of the waveform associated with the depolarization of the auricles of the heart, is commonly referred to as the "P wave." This deflection is followed by a complex of deflections, associated with the depolarization of the ventricles, commonly referred to as the "QRS" complex. The complex of deflections is followed by a longer deflection, associated with the depolarization of the ventricles, commonly referred to as the "T wave." One set of each of the P, QRS and T deflections occurs with each heart beat. The time period between QRS complexes or between R deflections is denoted the "R--R interval" and is shown in the figures as a line with arrows at both its ends. This interval is generally not less than .2 seconds and seldom exceeds 2.0 seconds.

Non-normal rhythms of the heart are generally referred to as arrhythmias. For the normal person the R-R intervals vary somewhat about an average value. An absolutely constant heart rate (a fixed R--R interval) is not typical of a normal heart, and in this sense is an arrhythmia; a very high heart rate (a short R--R interval) and a very low heart rate (a long R--R interval) are also usually referred to as arrhythmias. Detecting the presence and the type of an arrhythmia is of great value in the diagnosis and management of cardiac patients.

Of particular interest in cardiac diagnosis and management, and of significant clinical importance, is arrhythmia that is due to "premature beats," "ecoptic beats" or "premature ventricular contractions" ("PVCs"). R--R intervals are so labeled and the beats are unlabeled. The R--R interval from the QRS complex of the normal beat to the QRS complex of the premature beat is less than the R--R interval between normal beats; this interval is the "coupling time." For single PVCs the R--R interval between it and the following normal beat is usually longer than the R--R interval between normal beats. This interval is called the compensatory pause.

As shown in FIG. 1a, the R--R intervals are taken between the peaks of the QRS complex. The instrument then places that measurement in digital form, using a scale of a second divided into 25 parts (subintervals), the scale being shown in FIG. 1b. The R--R intervals will be from .2 to 2 seconds long, i.e., 5 to 50 subintervals.

In FIG. 1c a normal R--R interval, i.e., a heartbeat of a normal patient at the highest rate, is designated A and the R--R interval at the lowest normal rate is designated B. The range of intervals, on the scale, is A-B.

In contrast, as shown in FIG. 1d, in the samples from a "PVC" subject, i.e., one having premature beats, the normal beat interval is "A," the shortest coupling time is C, the longest coupling time is D and the interval for the compensatory pause is E. FIG. 1e shows the chart with samples from a patient having a heart with AV blockage (the entire QRS complex is missing). The sample produced will show the block. As shown, the missed beat interval F falls outside the range of normal intervals A.

The block diagram of the circuit of the present invention of FIG. 2 shows electrodes 10, 11 and 12 which are adapted to be connected to a patient. The electrodes 10, 11 and 12 are connected to differential amplifier 13. The output of amplifier 13 is to a delay device 14 and to a QRS detector 15, for example, a peak detecting circuit. The QRS detector produces a sharp trigger pulse at each QRS peak and screens out other portions of the ECG waveform such as P waves, T waves, muscle artifacts, electrode movement artifacts, etc. The delay device may be a magnetic recording device or an electronic delay line.

The QRS detector 15 is connected to an interval timer 16 which measures the intervals between the output pulses of the QRS detector. The measurement is in .04 second subintervals. The interval timer 16 is connected to a set of memory cells 17, which may be a set of flipflop circuits connected in a register. Each memory cell is either in a state 0 or a state 1. The record samples are obtained over a period, for example, a 1-hour period. At the beginning of each period all memory cells are reset to state 0. The memory cells are combined with the R--R interval measuring circuit such that the occurrence of a given value of R--R interval will set the corresponding memory cell to state 1. Once a cell is 1 it remains 1 until reset (new period). All further occurrences of the same R--R interval will thus have no effect.

The memory cells (storage device) establish a set of intervals which have already occurred. The inhibitor 18 is connected to QRS detector 15 and to the storage device 17. Each trigger from the QRS detector 15 tries to turn the recorder 19 on. As shown in the comparison function diagram of FIG. 2b, if that particular subinterval has occurred previously in the period (i.e., memory cell is 1) it is inhibited and has no effect. However, if that subinterval has not occurred before in the period, for example, because it is either too long or too short, it triggers a conventional monostable circuit whose output turns on the ECG recorder 19 for a period of several (preferably 2.6) seconds. Preferably the ECG recorder 19 is a conventional single pen moving paper strip graph recorder. The stylus drive 20 of the recorder 19 is operated from the delay device 14. The preferred speed of the ECG record is 25mm/sec so that each .04 subinterval of the interval timer 16 corresponds to 1 mm of paper movement of the ECG recorder.

The operation of the circuits of FIGS. 2a and 2c is shown in connection with FIGS. 3a-3e. In FIG. 3a all the memory cells are at zero at the beginning of a data period.

Assuming the first few beats have "normal" R--R intervals some of the boxes (corresponding to memory cells) in the normal range become 1, see FIG. 3b. FIGS. 3c and 3d show the result of a premature beat, the short R--R interval from the premature to the preceding normal beat (coupling time) causing an "early" memory cell to go to 1. The following long R--R interval (compensatory pause) between premature and succeeding normal beats causes a "late" memory cell to go to 1. By the end of the data period all memory cells corresponding to the R--R intervals that have occurred are 1, see FIG. 3e.

Each trigger from the QRS detector 15 tries to turn the recorder 19 on. If the memory cell for the R--R interval just measured (i.e., interval from the trigger to its predecessor) is not yet 1, it succeeds. If the memory cell is 1, it is inhibited. Thus the recorder will only be turned on when a novel R--R interval appears. The maximum number of records cannot exceed the number of quantisized R--R intervals, about 45, and one sample record will be produced for each R--R interval which does occur.

The signal to the recorder stylus 20 is the output of the ECG amplifier 13 delayed in time by delay device 14. Preferably the delay time is 2 seconds, although longer times are equally feasible. A suitable period for the recorder to remain on is 2.6 seconds. As shown in FIG. 4a the last beat of the pair whose R--R interval caused the sample will be .6 seconds or 15 mm from the end of the record (the extra .6 second insuring that the T wave will be recorded).

The number of samples cannot exceed the number of R--R intervals that actually occurred and will often be less since a "new" R--R interval can occur while the recorder is still running due to a previous "new" R--R interval. The preferred mode of operation of the recorder is to have it run for, say, 2.6 seconds as measured from the last of the "turn on" signals. This will result in some records lasting over 2.6 seconds, see FIG. 4b , so that the record will contain all the beat pairs giving rise to new R--R intervals.

The sampling technique of the present invention is most valuable when combined with an R--R interval histogram. The height of the bars in the histogram, tells the number of occurrences of a given R--R interval, and the sample record, described above, tells what type of beats caused this bar. Use of a proper mounting form can make it easy to reconstruct which sample goes with which bar. FIG. 5a shows an example of an R--R interval histogram which is characteristic of isolated PVCs. The center clusters are the R--R intervals between normal beats, and the early and late clusters are due to the premature beats. Adding the bar heights in either gives the total number of PVCs that occurred. The ECG samples would include specimens of normal beats and PVCs.

Much of the circuitry shown in FIG. 2a is common with that needed to produce the histogram. The memory cells previously described would be the least significant digit controlling the bar height. A simple point of view is that a sample is taken when, and only when, a bar starts, i.e., goes from a count of 0 to a count of 1. FIG. 5b depicts some variations on the sampling scheme. One could take a sample only when the count reaches, say, 5. This might be advantageous to avoid records due to a few artifacts just starting a bar. If desired, it would also be easy to arrange for multiple samples on a bar, say at counts 1, 5, 50, 100 and 200. FIG. 5b shows a switching arrangement that permits this type of flexibility.

Claims

I claim:

1. The method of automatically recording arrhythmic heartbeats in a patient, comprising attaching electrodes to the patient to a monitor his heartbeat, amplifying the electrical wave signals received on the electrodes, measuring the time intervals between similar portions of the said electrical wave signals, converting said timed intervals into quantified interval and, by the employment of an electronic circuit, storing the set of said quantified intervals, comparing subsequent intervals with said set of stored intervals and commencing the recording of an electrocardiogram and storing of said subsequent interval upon the comparison of an interval which does not match any of said set of stored intervals.

2. The method of claim 1 wherein the said similar portions of the signals are the peaks of the QRS complex.

3. The method of claim 1 and including the step of operating the recorder after the said commencement for a short predetermined time period.

4. An instrument for the detection of arrhythmic heartbeats, including electrodes adapted to be connected to a patient, an amplifier connected to said electrodes, a detector of a selected portion of the wave signal and a signal delay device both connected to said amplifier, an interval timer to time the interval between heartbeats connected to said detector, a memory device connected to said timer to store a set of said intervals, each one of which is different from the others of the set, a comparison device connected with said detector and to said storage device to compare intervals with said set of stored intervals and to produce an operating pulse and to store said composed interval only upon a finding of non-identity, an ECG recorder connected to said delay device and connected in series with said comparison device, and a period timer connected to said recorder, whereby upon receipt of said operating pulse said recorder records the signals from said delay device for the period set by said period timer.

5. An instrument as in claim 4 wherein said detector is a detector of the peaks of the QRS complex and produces a trigger pulse only at each said peak.