Sleep Apnea Detection System Using Machine Learning on Resource-Constrained Devices

Abstract

Sleep Apnea is a condition in which a person has pauses in breathing or very low breathing episodes during sleep. It is a condition that could prove life-threatening if not monitored and treated. A medical diagnosis of Sleep Apnea involves overnight recording of body signals, monitoring by a medical professional, use of hospital based equipment and data analysis for detection of anomalies. During the past decade, the measurement and analysis of human body signals using machine learning techniques on embedded devices have started to transform healthcare applications. The use of cost effective micro-controllers can ensure that health monitoring is available and accessible to all. In this paper, we show that machine learning models deployed on microcontrollers can successfully analyze ECG signals in real-time for Sleep Apnea detection. We have created TinyML models using TensorFlow Lite which we have deployed on cost effective and resource constrained devices like the Raspberry Pi Pico and ESP32. Our setup has given results comparable to more advanced and expensive devices for the detection of Sleep Apnea using ECG signals.