Fractionated electrograms and rotors detection in chronic atrial fibrillation using model-based clustering

Area of Science:

• Cardiology
• Biomedical Engineering
• Computational Biology

Background:

• Atrial fibrillation (AF) ablation therapy requires precise identification of sustaining substrates.
• Current methods like complex fractionated atrial electrograms (CFAE) detection are inadequate for chronic AF.
• Rotors are hypothesized as key fibrillatory substrates, but their identification and associated CFAE patterns remain challenging.

Purpose of the Study:

• To develop a non-supervised method for identifying fibrillatory substrates in chronic AF.
• To associate specific CFAE patterns with rotor activity.
• To improve the localization of rotors for enhanced ablation therapy.

Main Methods:

• Extraction of two features based on local activation wave detection.
• Inclusion of one feature derived from non-linear dynamics analysis.
• Application of Gaussian mixture model-based clustering for CFAE pattern discrimination and visualization on electroanatomic maps.

Main Results:

• The proposed method successfully differentiates various levels of CFAE fractionation.
• Evidence suggests that clustering can effectively pinpoint the location of rotor cores.
• The approach demonstrated efficacy on both real-world and simulated chronic AF data.

Conclusions:

• The developed non-supervised method can discriminate CFAE patterns associated with fibrillatory substrates.
• Clustering analysis shows promise in locating rotors, a critical substrate in chronic AF.
• This approach offers potential to enhance catheter ablation strategies by improving substrate identification.