Functional And STructural Assesment of the Heart by Artificial Intelligence-enabled Electrocardiogram for the Management of Atrial Fibrillation

The objective of this study is to evaluate whether an AI-ECG based screening strategy for detecting cardiac functional and structural abnormalities preserves clinical effectiveness and safety, compared with a conventional strategy of routine echocardiography in patients with AF, thereby demonstrating the non-inferiority of AI-ECG guided care.

Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia, with its prevalence having more than doubled over the past decade. AF is associated with an increased risk of stroke, heart failure, and mortality, thereby imposing a substantial burden on both patients and healthcare systems. Accordingly, contemporary clinical guidelines emphasize accurate diagnosis and early, integrated management of AF. In this context, transthoracic echocardiography has become a standard diagnostic tool for the assessment of structural heart disease and cardiac function. Despite being non-invasive and relatively low-cost, echocardiography is subject to several system-level limitations in routine clinical practice, including dependence on specialized equipment and trained personnel, scheduling delays, and inefficiencies related to repeated examinations. These constraints may create bottlenecks in the timely initiation and optimization of AF management. In real-world practice, a considerable proportion of patients with AF undergo echocardiography primarily to confirm the absence of significant structural heart disease or impaired function. A uniform strategy of performing echocardiography in all patients with AF may not be optimal from the perspectives of patient convenience and healthcare resource utilization. Moreover, depending on healthcare system capacity, access to echocardiography may delay the timely selection of optimal AF management. Conversely, selectively performing echocardiography in patients with a higher likelihood of structural or functional cardiac abnormalities may allow for a more efficient, timely, and targeted diagnostic approach. Artificial intelligence-enabled electrocardiography (AI-ECG) offers several practical advantages, including very short acquisition time, patients' convenience, substantially lower cost, and feasibility for repeated assessments during follow-up. AI-ECG may enable sensitive detection of changes in a patient's cardiac status over time. Positioning AI-ECG as an initial screening tool to identify patients with suspected structural or functional heart disease could facilitate a "screening-confirmation" diagnostic pathway, in which echocardiography is reserved for patients with abnormal or suspicious findings on AI-ECG. Such an approach has the potential to streamline initial and follow-up evaluations while maintaining patient safety.