Traditional AF classification (e.g., paroxysmal, persistent, permanent) relies largely on patient-reported symptoms and intermittent electrocardiographic monitoring, which cannot continuously or objectively reflect disease progression, nor effectively inform optimal intervention timing. Although "AF burden" (i.e., duration of AF episodes) has emerged as a potential marker, its association with clinical outcomes remains inconsistent due to limitations in monitoring methods and its one-dimensional nature. Based on our previous work, investigators developed a five-dimensional AF progression model using photoplethysmography (PPG) signals collected from wearable devices. This model quantifies AF progression across five domains: episode frequency, duration, temporal aggregation, circadian rhythm, and tachycardia burden, enabling continuous and multidimensional assessment. Prior validation has demonstrated high agreement with 24-hour Holter monitoring and effective identification of high-risk patients.
The BEAT-AF trial is designed to evaluate the clinical utility of this model in a real-world setting. Specifically, investigators will investigate whether early intervention (e.g., optimization of medical therapy or consideration of catheter ablation) in patients with elevated five-dimensional AF burden (≥4.59%) can reduce symptoms, rhythm/rate-related abnormalities, and AF-related adverse events (such as stroke and heart failure). This study is expected to provide new evidence for dynamic monitoring of AF progression, optimal timing of intervention, and personalized management strategies, ultimately improving patient outcomes.