Activation times (AT) describe the sequence of cardiac depolarization and represent one of the most important parameters for analysis of cardiac electrical activity. However, estimation of ATs can be challenging due to multiple sources of noise such as fractionation or baseline wander. If ATs are estimated from signals reconstructed using electrocardiographic imaging (ECGI), additional problems can arise from over-smoothing or due to ambiguities in the inverse problem. Often, resulting AT maps show falsely homogeneous regions or artificial lines of block. As ATs are not only important clinically, but are also commonly used for evaluation of ECGI methods, it is important to understand where these errors come from. We present results from a community effort to compare methods for AT estimation on a common dataset of simulated ventricular pacings. ECGI reconstructions were performed using three different surface source models: transmembrane voltages, epi-endo potentials and pericardial potentials, all using 2nd-order Tikhonov and 6 different regularization parameters. ATs were then estimated by the community participants and compared to the ground truth. While the pacing site had the largest effect on AT correlation coefficients (CC larger for lateral than for septal pacings), there were also differences between methods and source models that were poorly reflected in CCs. Results indicate that artificial lines of block are most severe for purely temporal methods. Compared to the other source models, ATs estimated from transmembrane voltages are more precise and less prone to artifacts.
Determination of activation times (ATs) using noninvasive electrocardiographic imaging (ECGI) is a promising technique for future diagnosis in cardiology. However, recent studies showed artificial lines of block (ALBs) in AT maps, estimated from reconstructed source signals. Although a variety of different source models and estimation methods are used, few attempts have been made to compare these. For this reason, a systematic compari- son was performed using three different source models (surface transmembrane voltages (TMVs), extracellular potentials (EPs) on an epi-endocardial surface, and EPs on a pericar- dial surface). Four different estimation methods were compared (deflection-based temporal (DefB-T), deflection-based spatiotemporal (DefB-St), correlation-based temporal (CorrB-T), and correlation-based spatiotemporal (CorrB-St)). Four physiological cases with different pacings and 10 pathological cases with elongated scars and patches were taken into account. Monodomain simulations were performed and resulting body surface potentials (BSPs) were calculated and corrupted with an additive white Gaussian noise (AWGN) to obtain a signal to noise ratio (SNR) of 20 dB. Reconstructions were performed using second-order Tikhonov regularization with 5 different degrees of smoothing and the L-curve-method to analyze the influence of the regularization. Subsequent AT estimation showed that TMVs performed better than EPs and showed fewer ALBs. Spatiotemporal approaches showed fewer ALBs than purely temporal ones. Deflection-based (DefB) methods could depict scars, but showed many ALBs. Correlation- based (CorrB) methods performed better than DefB methods and did not show ALBs for TMVs, but overblurred scars for pathological cases. A modification of the ￼method could be developed which resulted in the reproduction of scars without showing ALBs. In addition, it was shown that spatial oversmoothing in the reconstruction leads to character- istic ALBs and that this special kind of ALBs can be recreated by spatially smoothing true source signals. However, this does not explain all occurences of ALBs.