AI-Driven Computational Physiology in Cardiometabolic Syndrome: From Real-World Data to Mechanistic Understanding

Cardiometabolic syndrome arises from intricate interactions among metabolic, cardiovascular, behavioral, and environmental factors. The convergence of artificial intelligence (AI), computational physiology, and real-world data now enables a deeper, systems-level understanding of these interactions. By linking data-driven discovery with mechanistic insight, across animal and human studies, this Research Topic aims to illuminate the pathways driving disease onset, progression, and treatment response in cardiometabolic conditions.This Research Topic aims to bridge statistical and AI-driven discovery with mechanistic modeling to explain and predict cardiometabolic dynamics, while connecting preclinical and clinical evidence to accelerate translation from bench to bedside. We seek to integrate multimodal data (including EHRs, imaging, omics, biosignals, wearables, and social determinants of health) into coherent models that support precision risk stratification and therapy optimization. The Topic further promotes robust, reproducible methods for causal and mechanistic inference, digital twins, and in-silico experimentation in cardiometabolic research, and it fosters interdisciplinary collaboration among computational scientists, physiologists, and clinicians.We welcome contributions that develop or apply AI and computational/electronic physiology methods to advance mechanistic understanding and translational impact in cardiometabolic syndrome, including but not limited to:• AI-enabled multimodal modeling of cardiovascular–metabolic interactions using ECG/PPG, imaging, omics, sensor signals, and clinical text.• Computational and electronic physiology models simulating cardiac, vascular, and metabolic dynamics, including closed-loop and multi-organ systems.• Animal-to-human translational frameworks validating AI and physiological models across species, including model reduction/augmentation strategies and transfer learning.• Integration of social and environmental determinants of health (SDOH), lifestyle, and exposome data into mechanistic and predictive models.• Causal discovery and mechanistic inference for individualized treatment response, adverse event prediction, and risk stratification.• Digital twins and in-silico trials for disease progression modeling, therapy selection, and dose/target optimization.• Data harmonization, model validation, uncertainty quantification, interpretability, bias mitigation, and external generalization across populations and healthcare systems.• Real-world evidence pipelines linking EHR, registries, multi-omics, imaging, and wearable data with experimental physiology for hypothesis generation and testing.• Benchmark datasets, standards, and open-source tools that enhance reproducibility and clinical translation.We invite contributions that push the boundaries of modeling and methodology while keeping clinical relevance at the forefront, so that precision cardiometabolic care becomes both practical and impactful. Topic editors: Jianwei Zheng, Cyril S Rakovski, Islam Abudayyeh, Long-Sheng Song Source: https://www.frontiersin.org/research-topics/75525/ai-driven-computational-physiology-in-cardiometabolic-syndrome-from-real-world-data-to-mechanistic-understanding