Non-Invasive Characterization of Atrial Flutter Mechanisms Using Recurrence Quantification Analysis on the ECG: A...

Area of Science:

• Computational cardiology
• Biomedical signal processing
• Cardiac electrophysiology

Background:

• Atrial flutter (AFl) is a common arrhythmia with diverse underlying electrophysiological mechanisms.
• Non-invasive discrimination of these mechanisms is crucial for guiding targeted ablation therapies.
• Current methods often require invasive procedures to identify the specific AFl drivers.

Purpose of the Study:

• To implement Recurrence Quantification Analysis (RQA) on 12-lead ECG signals within a computational framework.
• To assess RQA's capability in discriminating between different electrophysiological mechanisms sustaining AFl.
• To evaluate the potential of non-invasive RQA for AFl mechanism identification.

Main Methods:

• Generated 20 distinct AFl mechanisms across 8 atrial models, simulating ECG signals using 8 torso models.
• Applied Principal Component Analysis (PCA) to the 12-lead ECG signals.
• Extracted six RQA-based features from significant PCA scores using individual component RQA and spatial reduced RQA approaches.

Main Results:

• RQA-based features demonstrated significant sensitivity to the dynamic structures of different AFl mechanisms.
• A hit rate of 67.7% was achieved in discriminating the 20 simulated AFl mechanisms.
• RQA features from a clinical ECG sample showed strong agreement with computational framework results.

Conclusions:

• RQA is an effective non-invasive method for distinguishing AFl electrophysiological mechanisms using computational ECG analysis.
• Proof-of-concept using clinical 12-lead ECG validates the utility of the simulation and RQA methods.
• This approach can optimize ablation strategies, reducing time and resources for invasive cardiac mapping.