Cardiol J. 2026;33:e00226032. doi: 10.5603/cj.111624.
ABSTRACT
Heart failure (HF) remains a leading cause of morbidity, mortality, and healthcare utilization worldwide, despite the availability of effective evidence-based therapies. The principal challenge is no longer the absence of treatment options but the limited capacity of traditional care models to deliver guidelinedirected medical therapy (GDMT) consistently and at scale. The COVID-19 pandemic exposed the fragility of hospital-centered HF care, highlighting the need for more resilient, patient-centered management strategies. Remote monitoring (RM) has been proposed as a solution, yet its clinical impact has been inconsistent due to fragmented data streams, declining patient adherence, and heavy reliance on continuous human oversight. Artificial intelligence (AI) offers an opportunity to address these limitations by integrating multidimensional clinical data, enabling earlier detection of deterioration, supporting adherence, and prioritizing clinically meaningful interventions. Emerging evidence suggests that AI-assisted workflows can accelerate GDMT optimization and improve surrogate and clinical outcomes when implemented within supervised care pathways. This has led to the concept of next-generation remote monitoring (NGRM), in which AI analyzes longitudinal physiological and behavioral signals to generate context-aware alerts and actionable recommendations while reducing clinical workload. Successful implementation, however, requires rigorous validation, clear governance, integration with clinical workflows, and safeguards for safety, equity, and accountability. When embedded within structured HF care pathways, AI-enabled monitoring may help bridge the persistent gap between evidence and real-world implementation.
PMID:41871039 | DOI:10.5603/cj.111624