Analysis of a High-resolution technique for Estimation of Conduction Velocity Vectors for Closely Spaced Electrodes Using a 2D Cardiac Tissue Model.

Because of the 1KHz sampling rates, standard clinical systems face a challenge of resolving local activation time (LAT) differences (less than 1ms) for closely spaced electrodes. Recently, Gaeta et al designed a novel technique that calculates LAT differences by transformation of bipolar electrogram (EGM) amplitude known as the DELTA (Determination of EGM Latencies by Transformation of Amplitude) method with a resolution less than 1 millisecond. This thesis evaluates the conduction velocity (CV) vectors measured using the DELTA method against CV vectors obtained using the activation times identified from the underlying transmembrane voltage (VMAT) at high temporal resolution and from the unipolar electrograms (UAT) at clinical temporal resolution. A 2D model of both normal and fibrotic cardiac tissue was developed to compute simulated extracellular EGMs and transmembrane potentials. LAT differences were estimated using the DELTA, UAT and VMAT methods for different stimulation sites. The triangulation technique was used to convert these LAT differences to vector maps and the magnitude and angle generated by both methods were compared. From the analysis, it was determined that the DELTA method gave estimates of the conduction velocity vectors with closely spaced electrodes (1mm,1.4mm and 2mm) that were nearly as accurate as the high resolution VMAT method (error less than 5%) in both normal and low-level fibrotic tissue (below level 3). For more complex wavefronts, defined here as wavefronts generated by using more than one source of simulation, the recorded errors with DELTA were slightly higher (5-10%) and followed no specific trend. In all cases, the DELTA method was more accurate in estimating CV magnitude and angle than the standard UAT method using electrodes separated by 2mm or less.