Sequential mapping of cardiac arrhythmia without fiducial time reference
View Patent ↗A method for generating a propagation and velocity maps for cardiac wavefront propagation including cardiac arrhythmia, sinus rhythm, and paced rhythm. Activation time information is generated in the absence of any time alignment reference, wherein an estimated activation time is a weighted summation of potentially nonlinear and nonorthogonal candidate functions (CFs) selected from a bank of CFs. Time alignments between sequential recordings may be done by including binary level functions among selected CFs. Embodiments are applicable to single catheter mapping and sequential mapping, and are robust as confirmed by the ability to generate propagation maps and conduction velocity in the presence of multiple colliding wavefronts. The propagation and conduction velocity maps may be used for one or more of diagnosing cardiac arrhythmia, localizing cardiac arrhythmia, guiding catheter ablation therapy of cardiac arrhythmia, and guiding cardiac pacing therapy.
1. A method for generating a map for cardiac wavefront propagation and for treating atrial arrhythmia, comprising:
obtaining a set of local activation times (LATs) of electrograms and location coordinates of electrodes of a catheter;
using a processor to execute code stored on non-transitory storage media compatible with the processor, the code containing instructions to direct the processor to generate an activation time function (ATF) by:
using two or more candidate functions (CFs) selected from a bank of CFs and estimating weighting and unknown parameters of the CFs to match available LATs of the electrodes;
determining the ATF as a combination of the selected CFs that matches the LATs;
using the ATF to generate at least one map selected from a propagation map, an isochronal map, and a conduction velocity map of the cardiac wavefront propagation; and
treating atrial arrhythmia in a subject by performing ablation therapy guided by the at least one map.
2. The method of claim 1 , wherein the generating the activation time function and the at least one map does not require using a reference catheter for time alignment.
3. The method of claim 1 , wherein the ATF is a weighted summation of nonlinear, nonorthogonal, candidate functions (CFs) of coordinates of the electrodes of the catheter.
4. The method of claim 1 , comprising using a method selected from exhaustive search, FOS, LASSO, and OLS to expand the activation time function as a linear combination of CFs.
5. The method of claim 1 , comprising using a computationally efficient selection method (CESM) for selecting CFs; and
modifying the ATF by selecting additional CFs.
6. The method of claim 5 , wherein the at least one map is a low resolution map created by using a small number of CFs in the ATF;
extracting time alignment shifts using the low resolution map; and
generating a final propagation map by including a larger number of CFs in the ATF.
7. The method of claim 1 , comprising using the activation times to estimate the conduction velocity.
8. The method of claim 1 , comprising using an absolute error of the LATs and an activation map at different electrode placements to create an error map;
wherein the error map shows reliability of the generated at least one map and identifies sites with complex signals and/or sites with low density of collected electrogram signals.
9. The method of claim 1 , wherein the set electrograms is obtained simultaneously from a single catheter placement.
10. The method of claim 1 , wherein the set of electrograms is obtained from two or more catheter placements during sequential recording, with unknown time shifts between different recording placements.
11. The method of claim 10 , comprising including a predefined binary level set of CFs in the output of the selected CFs;
obtaining a constant value time shift of the activation times of the recordings; and
identifying sequential time alignment references (TARs);
wherein phase differences between sequential recording at different sites are compensated, and the sequential recording is transformed to continuous recording.
12. The method of claim 11 , comprising using the selected CFs with a method selected from FOS, LASSO, OLS, and CESM to identify the sequential TARs.
13. The method of claim 1 , wherein the atrial arrhythmia comprises at least one of atrial flutter, atrial fibrillation, and atrial tachycardia.
14. The method of claim 1 , wherein the atrial arrhythmia comprises atrial fibrillation.
15. The method of claim 1 , wherein a CESM is used to output the at least one map; wherein the CESM comprises selecting all or a portion of the CFs at once.
16. The method of claim 15 , wherein the CESM comprises selecting radial basis CFs with centres uniformly distributed on a mapping region.
17. The method of claim 1 , comprising dividing available measurement points with known LATs into clusters and selecting CFs to generate an ATF for each cluster; and
generating the at least one map using an ATF that uses the selected CFs of each cluster; or
generating the at least one map using an ATF that uses the selected CFs of each cluster and one or more additional CFs selected from previously non-selected CFs.
18. The method of claim 1 , comprising producing a reliability map based on an error between measured LATs and LATs obtained with the determined ATF;
wherein the reliability map shows one or more of: sites where the determined ATF properly describes the wavefront propagation; sites where the determined ATF is unable to predict activation time and complexity of the wavefront propagation.