Surgery – Diagnostic testing – Cardiovascular
Reexamination Certificate
2000-09-29
2004-05-11
Layno, Carl (Department: 3762)
Surgery
Diagnostic testing
Cardiovascular
C600S516000
Reexamination Certificate
active
06735466
ABSTRACT:
TECHNICAL FIELD
The invention is directed to techniques for measuring alternans in an electrocardiogram (ECG) waveform.
BACKGROUND
Alternans, a subtle beat-to-beat change in the repeating pattern of an (ECG) waveform can be indicative of electrical instability of the heart and increased susceptibility to sudden cardiac death. Alternans results in an ABABAB . . . pattern of variation of waveform shape between successive beats in an ECG waveform. The level of variation has been found to be a useful characterization of an individual's cardiac electrical stability, with increasing variation being indicative of decreasing stability.
Referring to
FIG. 1
, an ECG waveform for a single beat is typically referred to as a PQRST complex. Briefly, the P wave appears at initiation of the beat and corresponds to activity in the atria, while the QRST complex follows the P wave and corresponds to ventricular activity. The QRS component represents the electrical activation of the ventricles, while the T wave represents their electrical recovery. The ST segment is a relatively quiescent period. It has been found that the T wave is the best interval of the ECG complex for detecting alternans. That is, a level of variation in the T waves of alternating beats is the best indicator of the electrical stability of the ventricles—the heart's main pumping chambers.
While an ECG waveform typically has a QRS amplitude measured in millivolts, an alternans pattern of variation with an amplitude on the order of a microvolt may be clinically significant. Accordingly, the alternans pattern may be too small to be detected by visual inspection of the electrocardiogram and often must be detected and quantified electronically. Such electronic detection and quantification of the alternans pattern is further complicated by the presence of noise in the ECG waveforms, as the noise may result in beat-to-beat variations that have a larger magnitude than the alternans pattern of variation.
The noise in an ECG signal can be classified into three categories: baseline noise generated in the electrode, physiologic noise, and external electrical noise. The baseline noise is low frequency noise that appears as an undulating baseline upon which the ECG rides. Baseline noise is attributable to motion and deformation of the electrode, and results from low frequency events such as patient respiration and patient motion. As a result, the magnitude of baseline noise tends to increase with exercise. Typically, the frequency content of baseline noise is below 2 Hz.
Physiologic noise results from physiologic processes, such as skeletal muscle activity, that interfere with the ECG signal. The electrical activity of the skeletal muscles creates potentials that are additive with respect to the potentials created by the heart. The frequency content of the skeletal muscle signals is comparable to the frequency content of the QRS complex, and is typically greater than 10 Hz. When measuring T-wave alternans, additional physiologic noise may result from changes in the position of the heart due to respiration or from changes in the projection of the electrical potential from the heart to the skin surface due to thoracic conductivity changes arising from the inflation and deflation of the lungs with respiration.
External electrical noise results, for example, from ambient electromagnetic activity in the room, electrode cable motion, and variations in amplifiers or other components of the ECG circuitry. External electrical noise may be eliminated or reduced through the use of high quality components and through the reduction of ambient electromagnetic activity by, for example, deactivating high power equipment.
Noise in the ECG waveform also can mimic the presence of alternans where none exists. For example, if a patient is breathing at one half or one third of the heart rate, the respiration may introduce a harmonic signal having the ABABAB . . . pattern of alternans. Motion that repeats with some periodicity, such as that resulting from exercise, can create electrode noise with a similar pattern.
One alternans measurement approach that attempts to address the effects of noise is referred to as the spectral method for measuring T-wave alternans. This method is described in detail in U.S. Pat. No. 4,802,491, which is incorporated by reference. In summary, this method involves concurrently analyzing 128 beats of a continuous stream of ECG signals. The spectral method uses measurements from time synchronized points of consecutive T waves. A time series is created by measuring, for each of the 128 beats, the T-wave level at a fixed point relating to the QRS complex. This process is repeated to create a time series for each point in the T wave. A frequency spectrum is then generated for each time series, and the spectra are averaged to form a composite T-wave alternans spectrum. Since the T-waves are sampled once per beat for each time series, the spectral value at the Nyquist frequency, i.e. 0.5 cycle per beat, indicates the level of beat-to-beat alternation in the T-wave waveform.
The alternans power is statistically compared to the noise power to discriminate the beat-to-beat T-wave variation due to abnormal electrical activity of the heart from the random variation due to background noise. The alternans power is calculated by subtracting the mean power in a reference band used to estimate the background noise level (for example, the frequency band of 0.44-0.49 cycle per beat) from the power at the Nyquist frequency (0.50 cycle per beat). Alternans is considered to be significant if the alternans is at least three times the standard deviation of the noise in the noise reference band.
SUMMARY
The invention provides improved techniques for measuring T-wave alternans.
The spectral method for T-wave alternans measurement relies on two assumptions. The first assumption is that the physiological T-wave alternans is phase-locked with the T-wave and therefore can be sampled at a constant phase in every beat. This assumption is valid due to the physiological factors associated with the generation of T-wave alternans.
The second assumption is that the noise within the noise reference band is white. To satisfy this condition, colored noise (e.g., motion artifact due to exercise) must be avoided within the noise reference band. During an ergometer exercise test, the pedaling rate can be adjusted to ⅓ or ⅔ of the heart rate using auditory and visual cues to move most of the colored noise away from the noise reference band and the alternans frequency. Nonetheless, colored noise often can fall within the noise reference noise band and at the alternans frequency during ergometer exercise due to the exercise, other motion artifact, respiration, and other sources. Furthermore, during treadmill exercise, it is difficult to control the stepping rate and it is common for the exercise-induced motion artifact to create colored noise that falls within the noise reference band and at the alternans frequency.
In one general aspect, measuring alternans in a physiological signal includes processing the physiological signal to create a processed signal having an asymmetric spectrum (i.e., a spectrum that is asymmetric around DC), and processing the processed signal to measure alternans in the physiologic signal.
Processing the physiological signal to create a processed signal may include creating the processed signal as an analytical signal, or as an approximation of an analytical signal. Creating an analytical signal may include generating a frequency domain representation of the physiological signal, modifying the frequency domain representation to remove components corresponding to negative frequencies, and generating the analytical signal as a time domain representation of the modified frequency domain representation.
The physiological signal may be an electrocardiogram. The electrocardiogram may be recorded from a subject during exercise, such as exercise using an ergometer or a treadmill.
Processing the processed signal may include sampling th
Cambridge Heart, Inc.
Fish & Richardson P.C.
Layno Carl
LandOfFree
Analytical signal method for analysis of T-wave alternans does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Analytical signal method for analysis of T-wave alternans, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Analytical signal method for analysis of T-wave alternans will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3235420