AI-Powered Cardiovascular Risk Prediction: Integrating Wearable Biosensors and Machine Learning Algorithms

This study examines the factors that improve cardiovascular risk prediction by integrating wearable biosensors with AI-driven analytics. Moving beyond static demographic indicators, the research frames prediction accuracy within Socio-Technical Systems (STS) Theory (Salwei & Carayon, 2022) and the Biopsychosocial (BPS) Model (Adler, 2009), viewing it as an outcome of alignment between technological innovation and behavioral context. Survey data were collected from 650 professionals across Vietnam, Singapore, and the United Kingdom, including cardiologists, biomedical engineers, healthcare managers, and behavioral scientists, yielding 385 valid responses. Exploratory factor analysis confirmed construct validity, and regression results showed that HRV monitoring (β = 0.753) and AI analytics (β = 0.665) significantly enhanced prediction accuracy. Moreover, lifestyle factors specifically sleep and physical activity moderated the HRV–prediction link (β = 0.501), strengthening predictive performance when healthy routines were present. The findings advance theory by establishing predictive accuracy as a socio-technical and biopsychosocial phenomenon, highlighting lifestyle as a critical moderator rather than a confounding variable. Practically, the results underscore the importance of integrating biosensors, AI, and lifestyle tracking into digital health ecosystems for personalized, clinically meaningful assessments. Limitations include the cross-sectional design, reliance on self-reported perceptions, and absence of longitudinal patient outcomes. Future work should employ prospective, multi-center trials and advanced HRV metrics to enhance generalizability.