Ventricular tachychardia (VT) is one of the mechanisms that induce sudden cardiac death. A way to cure VT is to ablate its exit site using catheter ablation. But it can be difficult, and in some cases impossible, to localize this exit site accurately enough for ablation.
We propose a process to facilitate catheter ablation using an automatic method to detect the exit site to guide the cardiologist during the intervention.
The proposed process uses the QRS complex integral of the standard 12-lead ECG, which is treated as coordinates of the exit site in ECG space and projected on the 3D space using linear transformation. It is based on an iterative process: 4 pacing sites are used to initialize a database which is used to identify a first target site, where a new pacing is applied. We repeat this process until there is no large movement between two successive estimated points. This method is fast enough to be used in a catheter ablation procedure.
We tested this process with 12 arrhythmia exit sites, randomly selected in a realistic ventricular model such that 2 were subendocardial, 2 intramural, and 2 subepicardial in each ventricle. With 12 pacing sites, 6 targets had been approximated to less than 1mm. Five more were within 5mm distance, and one was at 10mm distance.
We conclude that the proposed method does converge to the exit site, often within the radius of an ablation lesion. This is important in situations where further confirmation cannot be obtained, for example if the VT cannot be induced. However, in some cases the convergence was erratic, making convergence hard to establish, and required more than 10 pacing sites. The success of this simple test shows, nevertheless, that it would be worthwile to investigate improved algorithms.