A wavelet-based activation detector for bipolar electrogram analysis during atrial fibrillation

It has been shown that computation of atrial fibrillation (AF) electrogram (EGM) indices based on activation times is limited by the accuracy of the activation detector. In this work, a wavelet-based detector is proposed as a method to reliably extract activation time locations from the wavelet decomposition of non-linearly pre-processed bipolar EGM signal. A more classical amplitude adaptive threshold-based detector was also implemented for comparison purposes. Evaluation and validation was made by means of two scenarios due to the lack of standard databases: First, a simulation study where four real EGM signals, selected for its high SNR, were contaminated with noise at different SNR levels and detection performance was evaluated. Second, the inverse of the median activation cycle length (ACL) obtained from both detectors was compared with the spectral dominant frequency considered as gold standard. The proposed detector is more accurate and reliable than the threshold-based approach in the presence of noise, allowing a more reliable computation of activation-time-based AF clinical indices.