Sequential local activation time (LAT) mapping of intracardiac electrograms' activations requires a stable reference signal to align recording phases.
Objective: This work's purpose is to develop an LAT mapping approach that does not rely on a time-alignment reference (TAR).
Methods: To create an LAT map in absence of TAR (TARLess), the coordinates and LATs of recording electrodes are collected sequentially; a bank of candidate functions (CFs) is constructed that contains constant binary level CFs and non-linear functions of recording points' coordinates. Finally, a subset of CFs is linearly combined to create an activation time function with output matching electrodes' LATs. Synthetic and clinical data were deployed to validate TARLess. A simple two-dimensional computer model was used to create 30 different wavefront collision scenarios in a region with spatial conduction heterogeneities. Furthermore, sequential recordings were collected from seven atrial fibrillation patients during stimulation from one or two sites, after sinus rhythm was achieved post catheter ablation.
Results: We showed that TARLess maps are similar to the one that uses TAR; for the 20 clinical maps, the mean absolute difference between measured LAT with the TAR and TARLess approach was 5.2 ±2.0 milliseconds.
Conclusion: We developed a novel method to create an LAT map of sequential recordings without using any TAR and showed that it can create accurate maps even during the collision of multiple wavefronts.
Significance: TARLess mapping does not require a reference catheter which could lead to reduction in ablation procedure duration, cost, and potential complications.