Electrocardiography Meets Artificial Intelligence: Shaping the Future of Heart Failure Prediction A Comprehensive Review

Abstract Heart failure (HF) remains a leading cause of morbidity and mortality worldwide,and early detection and risk stratification are critical for improving patientoutcomes. Recent advances in artificial intelligence (AI) and deep learning (DL)have transformed electrocardiogram (ECG) analysis, enabling more accurate predictionof HF and its complications. This review synthesizes current literatureon AI- and DL-based ECG models for HF detection, prognosis, and treatment response,highlighting progress from traditional feature-based methods to end-toenddeep learning architectures, multimodal integration, and explainable AI approaches.Despite these advances, significant gaps remain, including underrepresentationof HFpEF and pre-clinical HF, limited modeling of comorbidities, scarceprospective validation, and insufficient attention to pediatric and special populations.We discuss these unmet needs and future research directions, emphasizingthe potential of AI-enabled ECG analysis to transform HF diagnosis, monitoring,and personalized care.