Integrating Artificial Intelligence with Coronary Plaque Biomechanics for Early Prediction of Heart Failure

Heart failure is a leading cause of hospitalization and long-term disability, with many individuals progressing from subclinical disease to overt symptoms before opportunities for effective intervention arise. Coronary atherosclerosis is central to this progression as it drives ischemia, microvascular dysfunction, inflammation, and chronic myocardial remodeling. These processes are not adequately reflected by conventional measures such as stenosis severity. Recent advances in coronary imaging and computational biomechanics have made it possible to characterize plaque morphology, composition, and the underlying mechanical and hemodynamic environment in much greater detail. Simultaneously, artificial intelligence enables the extraction of prognostic signatures from high-dimensional imaging and clinical data, supporting earlier and more individualized risk assessment.This Research Topic brings together multidisciplinary work at the intersection of artificial intelligence, coronary plaque biomechanics, and heart failure prediction. The goal is to highlight both innovative methodologies and clinically oriented studies that link plaque features and hemodynamic forces to myocardial remodeling, functional decline, and diverse heart failure phenotypes. We welcome contributions spanning imaging-driven discovery, hybrid models that combine physics-based and machine learning methods, multi-modality risk prediction, and real-world clinical validation. We particularly encourage predictive modeling that integrates clinical variables with imaging-derived metrics, such as characteristics from coronary computed tomography angiography, hemodynamic indices, and measures of myocardial structure or function, to improve risk stratification and enable personalized prevention and treatment strategies. Work that addresses interpretability, uncertainty quantification, robustness across imaging platforms and sites, and equitable model performance across populations is especially valued. Our aim is to accelerate reproducible research that advances beyond retrospective association toward actionable clinical decision support for prevention, monitoring, and therapy selection.To foster further advances in the field, we invite original research, methods, datasets, and focused reviews on topics including but not limited to the following:• AI and deep learning for plaque detection, characterization, segmentation, and prognostic modeling from coronary computed tomography angiography, intravascular ultrasound, optical coherence tomography, angiography, and multimodal imaging;• Biomechanical and hemodynamic modeling, including wall shear stress, plaque stress and strain, fluid structure interaction, and patient-specific simulation;• Hybrid and physics-informed machine learning, digital twins, explainable AI, uncertainty quantification, and pipeline benchmarking;• Prediction of incident heart failure and adverse cardiac remodeling, including heart failure with preserved or reduced ejection fraction, with longitudinal outcomes and external validation;• Predictive models that integrate clinical information such as data from electronic health records, demographics, comorbidities, therapies, and laboratory results with imaging data such as coronary and cardiac imaging features, including fusion strategies and deployable clinical risk scores;• Multimodal integration of imaging with biomarkers, wearable devices, and omics data, including federated or privacy-preserving learning;• Clinical translation, including workflow integration, prospective study design, regulatory and reporting considerations, and establishing links to therapy response. Topic editors: Diaa Hakim, ADRIANO CAIXETA Source: https://www.frontiersin.org/research-topics/77450/integrating-artificial-intelligence-with-coronary-plaque-biomechanics-for-early-prediction-of-heart-failure