Microsystems for obstructive sleep apnea treatment
Stavros Niarchos Foundation
Misleep developed an automated comfortable and compact system for the non-invasive treatment of obstructive sleep apnea (OSA) through the electrical stimulation (ES) of the genioglossus muscle. A prototype system has been developed which includes an epidermal flexible resistive strain sensor to be placed over the patient’s chest to monitor breath during sleep, a program that analyzes the signal of the sensor using signal processing and machine learning (ML) algorithms, a commercial transcutaneous electrical nerve stimulation (TENS) device that generates electric pulses, a control circuit based on Arduino–nano intervened between the TENS device and the electrode pads for controlling the electrical stimulation (ES) at the patient’s submental region and a Python program running on a PC that controls the whole system. Before the evaluation of the whole system, the algorithm of the system was evaluated using commercial sensors and signals from open and anonymized databases of polysomnography (PSG) studies and home sleep tests. It was concluded that using the developed algorithm a single sensor that monitors breath effort can be used for reliable and real time recognition of obstructive sleep apnea (OSA) events. Real life testing of the system was conducted by attaching the epidermal flexible strain sensor on five different regions of the human body.