A.-Robert LeBlanc

An Integrated System for Intraoperative Cardiac Activation Mapping

This paper describes an electrophysiological data acquisition and processing system which is programmed to rapidly generate cardiac activation maps for experimental studies and antiarrhythmia surgery. The basic system consists of a PDP-11/23+ minicomputer (DEC) with 1.5 Mbit memory and a 10 Mbit hard disk, a 64-channel data acquisition unit controlled by a specially designed interface card, and a modified video terminal. The data acquisition unit includes 64 instrumentation amplifiers with programmable gain and bandwidth. Signals are sampled and digitized at a maximum rate of 1000 samples/s/channel (10 bits) and transferred to the interface card by an optically isolated data bus. The operator controls the system by pointing to function boxes and signals appearing on the graphic terminal with a light pen. The software is based on a general-purpose data acquisition program with a command language interpreter. This program includes a setup section to define the systems parameters (gains, bandwidths, sampling rate) and an acquisition section to initiate data recording into a ring buffer, display the signals simultaneously on the screen, select heartbeats, and store or retrieve data on disk. The data processing procedures (in this case, mapping) can be easily interchanged to accommodate future signal processing needs. Current mapping procedures include an automatic detection section and an editor to manually define the local activation times on any electrogram. Another section displays the activation sequence as a map of isochronal lines. Typical processing time from the selection of a heartbeat to the visualization of the corresponding isochrone map is 2 min.

A new set of fast algorithms for mathematical morphology I: idempotent geodesic transforms

Abstract This paper generalizes the geodesic transforms used in mathematical morphology for binary image processing by introducing the concept of morphological operations constrained by an initial set and a transformable set. These new operations allow the efficient extension of geodesic constraints to grayscale image processing as they can be implemented with simple algorithms. They also make possible the application of new constraints, leading to new transforms with new properties. Two particular cases of these new transforms are then studied thoroughly: erosion and dilation operations constrained by an initial set and a transformable set that are iterated until idempotence is reached, or idempotent geodesic transforms (IGT). A sequential local transform (SLT) algorithm is presented that greatly speeds up the computation of IGT; the conditions for convergence for the algorithm and the relation between speed of convergence and the complexity of the initial set and the transformable set are investigated.

Evaluation of a subject-specific transfer-function-based nonlinear QT interval rate-correction method.

The QT interval in the electrocardiogram (ECG) is a measure of total duration of depolarization and repolarization. Correction for heart rate is necessary to provide a single intrinsic physiological value that can be compared between subjects and within the same subject under different conditions. Standard formulas for the corrected QT (QTc) do not fully reproduce the complexity of the dependence in the preceding interbeat intervals (RR) and inter-subject variability. In this paper, a subject-specific, nonlinear, transfer function-based correction method is formulated to compute the QTc from Holter ECG recordings. The model includes five parameters: three describing the static QT-RR relationship and two representing memory/hysteresis effects that intervene in the calculation of effective RR values. The parameter identification procedure is designed to minimize QTc fluctuations and enforce zero correlation between QTc and effective RR. Weighted regression is used to better handle unbalanced or skewed RR distributions. The proposed optimization approach provides a general mathematical framework for further extensions of the model. Validation, robustness evaluation and comparison with existing QT correction formulas is performed on ECG signals recorded during sinus rhythm, atrial pacing, tilt-table tests, stress tests and atrial flutter (29 subjects in total). The resulting average modeling error on the QTc is 4.9 ± 1.1 ms with a sampling interval of 2 ms, which outperforms correction formulas currently used. The results demonstrate the benefits of subject-specific rate correction and hysteresis reduction.

A new set of fast algorithms for mathematical morphology II: identification of topographic features on grayscale images

Abstract New algorithms for morphological image processing are presented. These algorithms are based on recursive idempotent geodesic transforms using a pixel-dependent structuring element. Such transforms are used as building blocks to construct new, fast algorithms for the location of topographic features on grayscale images. This approach is first used for the location of regional extrema and then for watershed and water parting determination. The new algorithms are compared to previously known ones in terms of speed and results quality. The same basic transforms are also used to build evaluation tools for measuring the relative importance of watersheds and water partings. The usefulness of the new approach is illustrated by application examples on a biomedical image.

The role of extracellular potassium transport in computer models of the ischemic zone

Ischemic heart disease is associated with large mortality and morbidity. Understanding of the relations between coronary artery occlusion, geometry of the ischemic region, physiology of ischemia, and the resulting changes in electrocardiogram (ECG) leads and catheter signals is important to support diagnosis and treatment. Computer models play an important role in understanding ischemia, by linking experimental to clinical results. In this paper we argue that the observed transport of extracellular potassium should be represented in such models. We used a diffusion equation to describe the transport mechanism. This model reproduced the measured spatial distribution of potassium, and its temporal development. We discuss the role of potassium transport next to other aspects of ischemia: the mechanism of changes in action potential and ECG, cellular coupling, anisotropic bidomain tissue conductivity, and the geometry of the ischemic zone.

Continuous ST-segment monitoring during coronary angioplasty using orthogonal ECG leads

In order to characterize ST-segment shifts during transient coronary artery occlusion, 24 patients with single-vessel disease were continuously monitored during percutaneous transluminal coronary angioplasty by use of a computerized orthogonal lead system. Changes of ST-segment (J + 60 ms) in leads X, Y, and Z and of the ST vector magnitude were analyzed by using 20 microV as a threshold for significant ST-segment shift. The sensitivity and magnitude of this shift were compared among the left anterior descending, right coronary, and circumflex artery groups (11, 8, and 5 patients, respectively) during balloon inflation. Significant ST-segment shifts were seen in 22 patients (92%) in ST-VM, Y, and Z leads and all patients in lead X (100%). There was no significant difference in sensitivity of either the ST vector magnitude or the most sensitive lead for occlusion detection among the three groups. There was a significantly greater magnitude of ST shift during left anterior descending artery occlusion than during right coronary artery and circumflex artery occlusions in ST-VM. Analysis of the direction of ST shifts in the X, Y, and Z leads showed a characteristic pattern, which could distinguish among the three coronary groups in 21 patients (88%). The presence of collaterals was significantly associated with ST-segment depression in leads oriented toward ischemia (3 of 6 patients) as compared with ST-segment elevation in the absence of collaterals (all of 15 patients), P > .01. It is concluded that ST-segment shift in the orthogonal leads is a reliable marker for myocardial ischemia. It is equally sensitive to occlusion of each of the three major coronary arteries and can thus identify the occluded coronary. An ST-segment depression instead of an elevation was related to the presence of collaterals, which may reflect a lesser degree of ischemia.

Extraction and Analysis of

Analysis of T waves in the ECG is an essential clinical tool for diagnosis, monitoring, and follow-up of patients with heart dysfunction. During atrial flutter, this analysis has been so far limited by the perturbation of flutter waves superimposed over the T wave. This paper presents a method based on missing data interpolation for eliminating flutter waves from the ECG during atrial flutter. To cope with the correlation between atrial and ventricular electrical activations, the CLEAN deconvolution algorithm was applied to reconstruct the spectrum of the atrial component of the ECG from signal segments corresponding to TQ intervals. The locations of these TQ intervals, where the atrial contribution is presumably dominant, were identified iteratively. The algorithm yields the extracted atrial and ventricular contributions to the ECG. Standard T-wave morphology parameters (T-wave amplitude, T peak-T end duration, QT interval) were measured. This technique was validated using synthetic signals, compared to average beat subtraction in a patient with a pacemaker, and tested on pseudo-orthogonal ECGs from patients in atrial flutter. Results demonstrated improvements in accuracy and robustness of T-wave analysis as compared to current clinical practice.

\hbox{T}

AIMS To quantify the sensitivity of QT heart-rate correction methods for detecting drug-induced QTc changes in thorough QT studies. METHODS Twenty-four-hour Holter ECGs were analyzed in 66 normal subjects during placebo and moxifloxacin delivery (single oral dose). QT and RR time series were extracted. Three QTc computation methods were used: (1) Fridericias formula, (2) Fridericias formula with hysteresis reduction, and (3) a subject-specific approach with transfer function-based hysteresis reduction and three-parameter non-linear fitting of the QT-RR relation. QTc distributions after placebo and moxifloxacin delivery were compared in sliding time windows using receiver operating characteristic (ROC) curves. The area under the ROC curve (AUC) served as a measure to quantify the ability of each method to detect moxifloxacin-induced QTc prolongation. RESULTS Moxifloxacin prolonged the QTc by 10.6 ± 6.6 ms at peak effect. The AUC was significantly larger after hysteresis reduction (0.87 ± 0.13 vs. 0.82 ± 0.12, p<0.01) at peak effect, indicating a better discriminating capability. Subject-specific correction further increased the AUC to 0.91 ± 0.11 (p<0.01 vs. Fridericia with hysteresis reduction). The performance of the subject-specific approach was the consequence of a substantially lower intra-subject QTc standard deviation (5.7 ± 1.1 ms vs. 8.8 ± 1.2 ms for Fridericia). CONCLUSION The ROC curve provides a tool for quantitative comparison of QT heart rate correction methods in the context of detecting drug-induced QTc prolongation. Results support a broader use of subject-specific QT correction.

Waves in Electrocardiograms During Atrial Flutter

A method is described for graphically presenting interval data, such as neural interspike intervals or electrocardiographic R-R intervals, in a form that facilitates the identification of nonstationaries. The method is essentially a plot of isoprobability contours of the cumulative interval histogram, as functions of time. A sequential algorithm is used for updating the contour-line positions. This display is used in an interactive system for visually identifying nonstationarities, and for subsequently comparing selected segments of the data quantitatively using the Kolmogorov-Smirnov test.