Atrial fibrosis identification with unipolar electrogram eigenvalue distribution analysis in multi-electrode arrays.

Area of Science:

• Cardiology
• Biomedical Engineering
• Medical Imaging

Background:

• Atrial fibrosis is a key factor in atrial fibrillation (AF) initiation and progression.
• Current methods using bipolar electrogram (b-EGM) voltage amplitude ( < 0.5 mV) for fibrosis identification have limitations.
• These limitations include disregarding spatiotemporal signal information and sensitivity to catheter orientation.

Purpose of the Study:

• To introduce and validate a new waveform dispersion measure, the dominant-to-remaining eigenvalue dominance ratio (EIGDR), for detecting atrial fibrosis.
• To hypothesize that EIGDR, derived from unipolar electrograms (u-EGMs) in neighbor electrode cliques, correlates with fibrosis presence.
• To compare the EIGDR method's performance against traditional b-EGM voltage mapping.

Main Methods:

• A simulated 2D cardiac tissue model with a fibrosis patch was used for validation.
• EIGDR maps were computed from original and time-aligned u-EGMs ([Formula: see text] and [Formula: see text]).
• The gain in eigenvalue concentration from alignment was mapped ([Formula: see text]), and performance was evaluated under noise and variable electrode-tissue distance.

Main Results:

• The time-aligned EIGDR map ([Formula: see text]) achieved the highest detection accuracy at 94%.
• Traditional b-EGM voltage maps showed 86% detection accuracy.
• The EIGDR strategy demonstrated effectiveness in real u-EGMs from 3D electroanatomical maps, supporting its potential as a fibrosis marker.

Conclusions:

• The proposed EIGDR method, particularly using time-aligned u-EGMs, significantly improves atrial fibrosis detection accuracy compared to conventional b-EGM voltage mapping.
• EIGDR shows promise as a reliable marker for atrial fibrosis.
• Further clinical studies are encouraged to confirm the translation of this technique into clinical practice.