From Wearable Sensing to Digital Twins: A Critical Narrative Review and Design Framework for Artificial Intelligence-Enabled Early Detection, Telemonitoring, and Risk Stratification in Pulmonary Hypertension

Emin Taner Elmas *

Vocational School of Higher Education for Technical Sciences, Division of Motor Vehicles and Transportation Technologies, Department of Automotive Technology, Iğdır University, Iğdır, Turkey and Graduate School of Natural and Applied Sciences, Major Science Department of Bioengineering and Bio-Sciences, Iğdır University, Iğdır, Turkey.

Feyyaz Akçin

Vocational School of Technical Sciences, Graduate of Computer Programming, Iğdır University, Iğdır, Turkey and Department of Software Engineering, Faculty of Engineering, Iğdır University, Iğdır-76000, Türkiye.

*Author to whom correspondence should be addressed.


Abstract

Pulmonary hypertension remains a condition in which the interval between symptom onset and confirmed diagnosis routinely exceeds one to two years, a delay associated with worse survival once idiopathic or connective-tissue-associated disease is finally confirmed. Over the past decade, four largely separate literatures have developed that each promise to shorten this interval or refine the care that follows it: wearable physiological sensing, artificial-intelligence-assisted interpretation of routine cardiac investigations, remote and implantable telemonitoring, and increasingly granular multiparametric risk stratification. A parallel and still largely aspirational literature proposes that these strands could ultimately converge into patient-specific computational digital twins capable of simulating disease trajectory and treatment response. This critical narrative review draws these four strands together for pulmonary hypertension specifically, rather than treating them as isolated sub-fields, and asks what a coherent translational pathway from wearable sensing to digital-twin-supported care would need to contain. Evidence was drawn from peer-reviewed studies, systematic reviews, and guideline documents identified through PubMed/MEDLINE and citation-network searching. Artificial-intelligence-assisted detection now performs well within selected, largely single-centre cohorts across echocardiography, electrocardiography, and chest radiography; hemodynamic telemonitoring has accumulated robust randomised-trial evidence in chronic heart failure but only feasibility-level evidence within pulmonary arterial hypertension itself; risk-stratification tools have become more granular without yet incorporating continuously collected physiological data; and digital-twin research in pulmonary hypertension itself remains almost entirely absent despite a considerably more developed literature in adjacent cardiovascular disease. Group 1 pulmonary arterial hypertension dominates the evidence base disproportionately relative to the more prevalent left-heart- and lung-disease-associated forms of the condition. Building on this synthesis, a tiered design framework is proposed linking opportunistic population-level screening, confirmatory non-invasive work-up, longitudinal telemonitoring, dynamic risk stratification, and prospective digital-twin simulation, together with the validation, equity, and governance work that each tier requires before clinical translation can be regarded as established rather than promising.

Keywords: Pulmonary hypertension, wearable devices, artificial intelligence, telemedicine, risk assessment, digital twin, remote monitoring, machine learning


How to Cite

Elmas, Emin Taner, and Feyyaz Akçin. 2026. “From Wearable Sensing to Digital Twins: A Critical Narrative Review and Design Framework for Artificial Intelligence-Enabled Early Detection, Telemonitoring, and Risk Stratification in Pulmonary Hypertension”. Cardiology and Angiology: An International Journal 15 (3):110-35. https://doi.org/10.9734/ca/2026/v15i3558.

Downloads

Download data is not yet available.