Optimizing CPAP Adherence and Quality of Life in Obstructive Sleep Apnea Patients using 3DoF-KF Hybrid Model Predictive Control

Obstructive sleep apnea (OSA) is a widespread sleep disorder that significantly impacts public health. Continuous positive airway pressure (CPAP) therapy is the gold standard for treating OSA; however, adherence to treatment remains a major challenge, and many patients discontinue CPAP use shortly after initiation. To address this challenge, this article presents a three-degree-of-freedom Kalman filter-based hybrid model predictive control (3DoFKF HMPC) framework aimed at improving CPAP adherence and quality of life among patients with OSA. A hypothetical dynamic model (developed with OSA domain experts) and patterned after the SleepWell24 intervention is presented; this model serves as both the simulation and internal model for the 3DoF-KF HMPC, which provides “ambitious yet attainable” CPAP goals while coordinating actions to mitigate reported symptoms. By integrating behavioral strategies with hybrid model predictive control techniques, the framework personalizes CPAP recommendations, aiming to enhance participant engagement and adherence. The results presented in this article show promise for the effectiveness of a future “just-in-time” control engineering-based adaptive intervention with the goal of optimizing CPAP use and improving healthrelated outcomes for newly diagnosed OSA patients.