A comprehensive mHealth solution for OSA patient management

Abstract

Obstructive Sleep Apnea (OSA) is a potentially fatal sleep disease that causes a variety of pathologies. Positive Airway Pressure (PAP) therapy is the preferred treatment for OSA. Nonetheless, when compared to the other 17 therapies, the PAP therapy has one of the lowest levels of compliance. We present 2 novel mHealth solutions for improving this adherence level in CPAP therapy. The first application, Sleep.py Experts, provides a variety of decision support tools to enable tailored patient management for PAP treatment. The end-users for Sleep.py Experts are the multiple stakeholders that are involved in the PAP therapy like the pulmonologist, The second application, Sleep.py, delivers motivational, educational, diet, physical activity and stress management interventions directly to the patient suffering from OSA. We include an innovative serious game to increase the patient’s compliance with the application. Sleep.py provides the most comprehensive collection of personalized interventions for PAP therapy. Another core feature of these applications is to collect feedback from either the experts or the patients, In this research, we also present a framework for creating this feedback and propagating them across various machine learning models in a continuous learning process. This paradigm is adaptable enough to be used in different therapies or fields. To summarise, we offer a clear roadmap for implementing a continuous learning method based on expert-validated outcomes.