An Internet of Medical Things System to Increase Continuous Positive Airway Pressure Usage in Patients with Sleep-Disordered Breathing

• Published in: SN computer science 2021-04, Vol.2 (2), p.117, Article 117
• Main Authors: Croatti, Angelo, Montagna, Sara, Lombardi, Carolina, Parati, Gianfranco, Pengo, Martino F., Silvani, Alessandro
• Format: Article
• Language: English
• Subjects: Airway management; Algorithms; Applications programs; Artificial intelligence; Blood pressure; Chatbots; Computer Imaging; Computer Science; Computer Systems Organization and Communication Networks; Continuous positive airway pressure; Data Structures and Information Theory; Embedded systems; Information Systems and Communication Service; Internet of medical things; Low cost; Mobile computing; Open source software; Original Research; Patients; Pattern Recognition and Graphics; Sensors; Sleep apnea; Sleep disorders; Sleepiness; Smartphones; Software Engineering/Programming and Operating Systems; Subsystems; Vision; Wrist
• ISSN: 2662-995X; 2661-8907
• DOI: 10.1007/s42979-021-00522-x

Obstructive sleep apnea (OSA) is a highly prevalent sleep disorder associated with increased daytime sleepiness and cardiovascular risk. Continuous positive airway pressure (CPAP), requiring a pressure-generating device connected via tubing to a mask during sleep, is an effective treatment. However, patients’ adherence to CPAP is often suboptimal. Behavioral interventions are effective in improving adherence to CPAP. We aimed to provide proof of principle for the operation of a low-cost, self-standing, internet-based system to measure and promote adherence to CPAP. The system is composed of triaxial acceleration sensors attached to the CPAP mask and to the wrist, able to record CPAP usage information, and a mobile app that collects such information and, thorough a chatbot, feeds back to the patient to improve adherence to treatment. The mask subsystem identifies time periods when the mask is put on based on relatively high values of the ratio between acceleration spectral power at frequencies < 0.35 Hz vs. 0.35–2 Hz over 1-min windows. Accuracy in identification may be increased taking account of the surges in the standard deviation of wrist accelerations over 1-min windows that accompany putting on and taking off the mask. The whole system can represent a unique tool capable of monitoring and improving patients’ adherence to CPAP treatment. Its main strength lies in its simplicity, low cost, and independence from the specific CPAP device and mask employed.