Latent Representations of Intracardiac Electrograms for Atrial Fibrillation Driver Detection

Atrial Fibrillation (AF) is the most prevalent sustained arrhythmia, yet current ablation therapies, including pulmonary vein isolation, are frequently ineffective in persistent AF due to the involvement of non-pulmonary vein drivers. This study proposes a deep learning framework using convolutional autoencoders for unsupervised feature extraction from unipolar and bipolar intracavitary electrograms (EGMs) recorded during AF in ablation studies. These latent representations of atrial electrical activity enable the characterization and automation of EGM analysis, facilitating the detection of AF drivers. The database consisted of 11,404 acquisitions recorded from 291 patients, containing 228,080 unipolar EGMs and 171,060 bipolar EGMs. The au-toencoders successfully learned latent representations with low reconstruction loss, preserving the morphological features. The extracted embeddings allowed downstream classifiers to detect rotational and focal activity with moderate performance (AUC 0.73-0.76) This work highlights the potential of unsupervised learning to uncover physiologically meaningful features from intracardiac signals. Introduction Atrial Fibrillation (AF) is the most common sustained cardiac arrhythmia in adults, affecting an estimated 59 million people around the world in 2019 [1]. It is defined as a supraventricular tachyarrhythmia characterized by disorganized electrical activity of the atrium and ineffective atrial contraction [2]. As life expectancy increases worldwide, the prevalence of AF is expected to rise accordingly [3]. Although some patients may be asymptomatic, many experience symptoms such as palpitations, fatigue, and dyspnea.