Identification of recurring patterns in fractionated atrial electrograms using new transform coefficients

Area of Science:

• Cardiovascular Electrophysiology
• Signal Processing in Medicine

Background:

• Complex fractionated atrial electrograms (CFAE) analysis helps distinguish paroxysmal from persistent atrial fibrillation (AF).
• Identifying CFAE patterns can guide radiofrequency catheter ablation to arrhythmia drivers.
• This study introduces a novel technique for detecting and classifying recurrent CFAE patterns.

Purpose of the Study:

• To develop and validate a robust method for detecting and classifying recurrent patterns in CFAE.
• To assess the technique's efficacy in differentiating AF types and guiding ablation.
• To enable real-time mapping of the AF substrate.

Main Methods:

• CFAE recordings from pulmonary veins and left atrium in 20 AF patients (paroxysmal and persistent) were analyzed.
• A new transform derived from ensemble averaging was used to construct basis vectors.
• Recurrent patterns (A and B) were detected and classified using spectral signatures and Euclidean distance thresholds in the presence of interference and noise.

Main Results:

• The technique achieved 96.2% sensitivity and 98.0% specificity in detecting patterns with interference.
• In the presence of interference plus noise, sensitivity was 89.1% and specificity was 97.0%.
• The method demonstrated robust detection and classification of CFAE patterns under noisy conditions.

Conclusions:

• Transform coefficients from ensemble averages effectively quantify synchronized patterns in AF data.
• The technique allows for unbiased, automatic detection of recurrent CFAE patterns amidst interference.
• This method can be implemented in real-time for AF substrate mapping during catheter ablation.