Generation Cardiovascular Risk Assessment Through Wearable Digital Biomarkers and Explainable Artificial Intelligence: Toward Continuous, Personalized, and Predictive Cardiovascular Medicine

cardiovascular disease, digital biomarkers, wearable biosensors, explainable artificial, intelligence, machine learning, cardiovascular risk prediction, predictive analytics, digital health, continuous physiological monitoring, smart wearable devices, precision cardiology, remote patient monitoring, electrocardiographic wearables, photoplethysmography, personalized medicine.

Nasirova Aziza Akbarovna.1
1
Samarkand State Medical University, PhD,
Associate Professor, Department of Internal
Medicine,
Faculty
of
Pediatrics,
Samarkand,Uzbekistan https://orcid.org/00000001-8795-2634 [email protected]
Khilola Shirinova 2
2
PhD, Department of Therapeutic stomatology,
Bukhara State Medical Institute named after Abu
Ali ibn Sino, Bukhara, Republic of Uzbekistan;
E-mail:
[email protected]
https://orcid.org/0000-0002-1167-9261

Received-19-05-2026
Revised-22-06-2026
Accepted-26-06-2026

European Journal of Prosthodontics and Restorative Dentistry (2026) 34(4S), 171–188

Generation Cardiovascular
Risk Assessment Through
Wearable Digital Biomarkers
and Explainable Artificial
Intelligence:
Toward
Continuous,
Personalized,
and
Predictive
Cardiovascular Medicine

Cardiovascular disease (CVD) remains the leading cause of mortality worldwide despite major advances in preventive cardiology, pharmacotherapy, and interventional management. Conventional cardiovascular risk stratification models, including population-based scoring systems derived from static demographic and laboratory variables, have substantially improved preventive care but remain limited in their ability to capture dynamic physiological variability, subclinical disease progression, and individualized risk trajectories. These limitations are particularly relevant in contemporary populations characterized by multimorbidity, heterogeneous lifestyles, and rapidly fluctuating cardiometabolic states. Recent developments in wearable sensor technologies have introduced a paradigm shift in cardiovascular monitoring by enabling continuous, noninvasive acquisition of high-resolution physiological data in real-world settings. Smartwatches, adhesive biosensors, photoplethysmography-enabled devices, electrocardiographic wearables, and multimodal physiological platforms now permit longitudinal assessment of heart rhythm dynamics, autonomic function, vascular stiffness, sleep architecture, physical activity, respiratory variability, and hemodynamic responses. The integration of these continuously acquired signals has accelerated the emergence of digital biomarkers capable of reflecting early pathophysiological alterations preceding overt cardiovascular events. Simultaneously, advances in machine learning and predictive analytics have enabled the extraction of clinically meaningful patterns from complex wearable-derived datasets that exceed the interpretive capacity of conventional statistical approaches. In particular, explainable artificial intelligence frameworks have gained increasing attention for their potential to enhance transparency, clinician trust, and regulatory acceptability while maintaining high predictive performance. These models may facilitate earlier identification of atrial fibrillation, heart failure decompensation, ischemic risk, sudden cardiac death susceptibility, and cardiometabolic deterioration through adaptive and individualized prediction strategies. This review critically examines the evolving role of wearable digital biomarkers and explainable machine learning models in next-generation cardiovascular risk assessment. Emphasis is placed on physiological signal interpretation, multimodal data integration, algorithmic •••••••••••••••••••••••••••••••• ejprd.org- Published by Riset Publishing Services LLC.

EJPRD

Copyright © 2026 by Riset Publishing Services LLC