Shimon Abboud

Simulation of high-resolution QRS complex using a ventricular model with a fractal conduction system. Effects of ischemia on high-frequency QRS potentials.

Recent studies have analyzed the high-fidelity surface electrocardiographic signal, and efforts have been made to increase the diagnostic sensitivity of the electrocardiogram by observing its high-frequency components. It was found that the high-frequency (150-250-Hz) electrocardiogram appears to detect evidence of transient ischemia with greater sensitivity than visual inspection of the surface electrocardiogram. A finite-element three-dimensional model of the ventricles with a self-similar (fractal) conduction system has been introduced as a bridge to the understanding of electrocardiographic phenomena related to high-frequency potentials. The model was activated, and the dipole potential generated by adjacent activated and resting cells was calculated to obtain a high-resolution QRS complex. Normal and ischemic activation processes was stimulated by regional reduction in conduction velocity. It was found that although the resulted low-frequency QRS complex was not significantly altered from normal conditions, the high-frequency components exhibited morphological changes similar to the ones observed during animal experiments and human studies. Based on the results obtained from the model, it can be concluded that these morphological changes can be attributed to a slowing of conduction velocity in the region of ischemia and that the model is adequate for meeting the challenges imposed by the requirements of high-frequency methods applied in clinical cardiology.

Effects of coronary occlusion on high-frequency content of the epicardial electrogram and body surface electrocardiogram.

The very high-frequency content (150 to 250 Hz) of epicardial electrogram waveforms was studied in 19 anesthetized dogs subjected to occlusion of left anterior descending coronary artery. Computer techniques of digital averaging and digital band-pass filtering were applied. Signals were obtained from epicardial electrodes placed in the ischemic left ventricular region and on the noninjured right ventricular surface, and from the body surface electrocardiogram. All recordings were made simultaneously before, during, and after coronary occlusion and subjected to the same analysis. The waveforms obtained from the ischemic left ventricular region showed a considerable decrease in high-frequency content, while those obtained from the noninjured right ventricular surface remained unchanged. The results correlated with the appearance of a zone of reduced amplitude in the body surface high-frequency QRS complex. Therefore, this macroscopic phenomenon measured noninvasively from the body surface is explained by local reduction of high-frequency activity in the ischemic region of the myocardium.

High frequency electrocardiography using an advanced method of signal averaging for non-invasive detection of coronary artery disease in patients with normal conventional electrocardiogram

The ability to detect coronary artery disease (CAD) in patients with normal, conventional electrocardiograms (ECG) was studied using high frequency electrocardiography and an advanced method of signal averaging in three groups of subjects. Group A consisted of ten healthy subjects under the age of 30; Group B of 15 patients with chest pains and normal coronary arteries; and Group C of 20 patients with chest pains and CAD. Four hundred ECG waveforms from leads V3, V4 and V5 were recorded, and the recorded analog data were digitized. The digitized waveforms were averaged using a cross-correlation function which operates in the frequency domain (fast-Fourier transform algorithm). The signals were filtered with a digital band pass filter with 150 and 250 Hz low and high cut-off frequencies, respectively. Zones of reduced amplitude in the envelope of the filtered QRS complex of at least two precordial leads were found more significantly in patients with CAD (75%) than in patients without CAD (12% for Groups A and B, p less than 0.00003). However, no significant difference was found between the voltage over the high frequency QRS complex and that of its envelope between the three groups. It is concluded that the morphology of the high frequency QRS complex as determined by this advanced method of analysis of the signal averaged ECG may be a useful indicator for the presence of CAD in patients with normal conventional ECGs.

Non-invasive recording of late ventricular activity using an advanced method in patients with a damaged mass of ventricular tissue.

Late potentials occurring after the QRS complex were detected on the body surface using an advanced signal averaging technique. ECG waveforms were recorded from patients with and without recurrent ventricular tachycardia who had a damaged mass of ventricular tissue (left ventricular aneurysm or right ventricular dysplasia). The recorded analog data were digitized by a 10 bit A/D converter with a sampling rate of 1280 Hz onto digital magnetic tape. The digitized waveforms were averaged with a CDC-6600 computer using an advanced algorithm which employs a cross correlation function to extract fiducial synchronizing marks for the signal averaging. Waveforms were filtered with a digital bandpass filter with 30 Hz and 250 Hz low and high cut-off frequencies, respectively. Late potentials were detected in seven out of ten patients with left ventricular aneurysm (or right ventricular dysplasia) and recurrent ventricular tachycardia and in four out of five patients with left ventricular aneurysm (or right ventricular dysplasia) but without recurrent ventricular tachycardia. The delayed depolarizations, which were recorded, had an amplitude of 3.0-27.0 microV and extended a mean of 90 msec beyond the termination of the QRS complex. In three patients with left ventricular aneurysm the delayed waveforms were abolished by aneurysmectomy. In nine control subjects no late potentials were detected. We conclude that late potentials which represent late depolarization of a damaged mass of ventricular tissue can be detected in patients with and without recurrent ventricular tachycardia.

The use of cross-correlation function for the alignment of ECG waveforms and rejection of extrasystoles.

The cross-correlation function for alignment of ECG waveforms and rejection of artifacts was studied. ECG waveforms were recorded and digitized (10-bit resolution) at 1.28 KHz and were processed with a CDC 6600 computer. The cross-correlation function was calculated using the cross spectrum and the fast Fourier transform algorithm. The maximum value of the cross-correlation function and the time location of that value were searched for (a) the similarity between the waveforms (for elimination of extrasystoles), and (b) measuring the relative time delay (for the waveforms alignment in the averaging process) between the ECG waveforms. Prior to the correlation procedure each of the ECG waveforms was filtered through a nonrecursive digital bandpass filter. Experiments with various filters indicated that when low-frequency cutoff at 30 Hz and high-frequency cutoff at 250 Hz were used the maximum value of the cross-correlation function was higher than 0.9 for most of our recorded waveforms and more accurate results were obtained.

QUANTIFICATION OF THE FETAL ELECTROCARDIOGRAM USING AVERAGING TECHNIQUE

A signal analysis procedure is described for obtaining time intervals parameters of the fetal electrocardiogram as recorded from the maternal abdomen. Applying averaging to the fetal electrocardiogram quantification of the PR interval, QRS duration and QT interval were measured. This technique which includes the subtraction of an averaged maternal ECG waveform using cross-correlation function and fast Fourier transform algorithm, enables the detection of all the fetal QRS complexes in spite of their coincidence with the maternal ECGs. Results that were obtained from 21 pregnant women at the gestational age of 32-41 weeks and an example of a recording with fetal premature ventricular contractions are presented. This method shows an important improvement with respect to detection of fetal heart rate and detection of arrhythmia disturbances in the fetal ECG. The averaging procedure can be used to evaluate long-lived alterations in the fetal ECG.

Effect of thrombolysis on the evolution of late potentials within 10 days of infarction.

Patients with late potentials in the signal averaged electrocardiogram are more at risk of lethal arrhythmias in the period after acute myocardial infarction. To test the effects of thrombolysis on the incidence and evolution of late potentials, 158 consecutive patients were prospectively studied during the first 10 days after acute myocardial infarction. The study population consisted of two groups: 93 control patients treated conservatively and 65 patients treated with intravenous thrombolysis. Recordings of signal averaged electrocardiogram were obtained within two days and 7-10 days after infarction. The incidence of late potentials in the first two days after infarction was not significantly different in the thrombolytic and control groups (14% v 11.8%). By 7-10 days the incidence of late potentials among patients who underwent thrombolysis remained unchanged (14%); however, it increased significantly in the control group (11.8% to 22.5%). Thus thrombolysis seems to reduce the evolution of late potentials within 10 days of infarction. Because the risk of fatal arrhythmias is higher in patients with late potentials this study may partly explain the reduced mortality after thrombolysis.

Numerical solution of the potential due to dipole sources in volume conductors with arbitrary geometry and conductivity

The integral conservation equation for biological volume conductors with general geometry and arbitrary distribution of electrical conductivity is solved using a finite volume method. An effective conductivity was defined for the boundaries between regions with abrupt change of the conductivity to allow the simultaneous solution of the entire domain although the derivatives are not continuous. The geometrical singularities arising from the spherical topology of the coordinate system are removed using the conservation law. The resulting finite volume solution method is efficient both in central processing unit (CPU) time and memory requirements, allowing the solution of the volume conductor equation using a large number of mesh points (of the order of 10/sup 5/) even on small workstations (like SGI Indigo). It results in very accurate solutions, as several comparisons with analytical solutions of head models reveal. The proposed finite volume method is an attractive alternative to the finite element and boundary element methods that are more common in bioelectric applications.

Subtle alterations in the high-frequency QRS potentials during myocardial ischemia in dogs

The high-frequency components in the QRS complex were examined before and during the occlusion of the left anterior descending coronary artery. ECG waveforms from different stages of the ischemia state were averaged using cross-correlation technique and fast Fourier transform algorithm. The average waveforms were the filtered with nonrecursive digital filters (150-250 Hz) to obtain the high-frequency QRS complex. We tested our nonrecursive digital filter for ringing and the trigger jitter during the averaging process. We demonstrated that the high-frequency signal left after filtering is the result of a subtle low-amplitude constituent of the QRS complex and that the morphological change during ischemia (as zone of reduced amplitude) is a result of changes in these components of the original ECG signal.

Correlation between skull thickness asymmetry and scalp potential estimated by a numerical model of the head

The contribution of asymmetric skull thickness to the scalp potential amplitude was investigated numerically. The model consisted of four conductive layers representing the scalp, the skull, the cerebrospinal fluid, and the cortex with a current dipole in the occipital region. The potential created by the dipole was calculated assuming quasistatic formulation and linear media. The governing equation was discretized by the finite volume method to ensure the conservation of fluxes even in regions with abrupt changes of the conductivity. The large set of the algebraic equations for the electric potential was solved iteratively by the successive overrelaxation method. The model confirmed previous experimental studies suggesting that the potential amplitude is 60% smaller on the side with the thicker bone if the asymmetry of the skull thickness exceeds 40%. The model developed suggests that skull thickness asymmetry can create nonnegligible asymmetries in the potential measured on the scalp above homotopic points of the two hemispheres.<<ETX>>