Adaptive Autoregressive Moving Average (ARMA) Parametric Modeling for Tracking Long-Term Cardiovascular Risk Progression in Chronic Sleep Apnea Patients

• Abiodun Okunola

  Ladoke Akintola University Technology

  Author

Chronic sleep apnea affects approximately 25% of middle-aged adults and more than doubles cardiovascular disease (CVD) mortality through mechanisms of sympathetic activation, oxidative stress, and metabolic derangements . Despite this well-established association, current clinical risk stratification relies almost solely on the Apnea-Hypopnea Index (AHI), overlooking the heterogeneity in multimorbidity and therapeutic response that characterizes this patient population. This study addresses the critical gap in longitudinal cardiovascular risk prediction by developing an adaptive Autoregressive Moving Average (ARMA) parametric modeling framework that captures the temporal dynamics of CVD progression in chronic sleep apnea patients. Leveraging longitudinal data from the Wisconsin Sleep Cohort Study comprising 1,123 participants tracked over several decades , the proposed methodology integrates parametric estimation of respiratory signals  with dynamic time-series modeling to track individual patient trajectories. The adaptive ARMA model demonstrated superior predictive performance with an aggregate accuracy of 89.4%, significantly outperforming static risk assessment methods. The framework successfully identified two distinct phenotypic clusters corresponding to cardiovascular stability and elevated cardiovascular risk profiles, with marked differences in disease progression rates between clusters. This research provides a replicable, computationally efficient framework for personalized cardiovascular risk surveillance in sleep apnea patients, with significant implications for proactive clinical intervention and healthcare resource allocation.

• PDF

Copyright (c) 2026 Abiodun Okunola (Author)

This work is licensed under a Creative Commons Attribution 4.0 International License.