Simple and Autonomous Sleep Signal Processing System for the Detection of Obstructive Sleep Apneas

Abstract

Obstructive sleep apnea (OSA) is a common sleep disorder characterized by repetitive upper airway obstruction, intermittent hypoxemia, and recurrent awakenings during sleep. The most used treatment for this syndrome is a device that generates a positive airway pressure—Continuous Positive Airway Pressure (CPAP), but it works continuously, whether or not there is apnea. An alternative consists on systems that detect apnea episodes and produce a stimulus that eliminates them. Article focuses on the development of a simple and autonomous processing system for the detection of obstructive sleep apneas, using polysomnography (PSG) signals: electroencephalography (EEG), electromyography (EMG), respiratory effort (RE), respiratory flow (RF), and oxygen saturation (SO2). The system is evaluated using, as a gold standard, 20 PSG tests labeled by sleep experts and it performs two analyses. A first analysis detects awake/sleep stages and is based on the accumulated amplitude in a channel-dependent frequency range, according to the criteria of the American Academy of Sleep Medicine (AASM). The second analysis detects hypopneas and apneas, based on analysis of the breathing cycle and oxygen saturation. The results show a good estimation of sleep events, where for 75% of the cases of patients analyzed it is possible to determine the awake/asleep states with an effectiveness of >92% and apneas and hypopneas with an effectiveness of >55%, through a simple processing system that could be implemented in an electronic device to be used in possible OSA treatments.

Keywords: apnea; detection of apneas; electrical stimulation; hypopnea; obstructive sleep apnea; polysomnography; signal processing; sleep signal processing system.

Figure 1

Real signals and FFT of the occipital EEG channel O1 in awake and sleep epoch. (a) Asleep original signal, (b) Frequency analysis of sleep signal, (c) Awake original signal, (d) Frequency analysis of awake signal.

Figure 2

Accumulated signal of frequencies and states given by expert and algorithm. (a) Cumulative frequencies of the FFT signal of one of the EEG channels analyzed, (b) Binary signal of the awake/sleep stages given in annotations by the sleep expert, and (c) Binary signal of the generated awake/sleep stages by the signal processing algorithm.

Figure 3

ROC curve EEG signals. The ROC curves of the 6 EEG channels taken for signal analysis are shown. All the channels show AUC greater than 0.8.

Figure 4

False positive EEG signals. The X-axis shows the thresholds evaluated in the detection of the awake/sleep stages in one of the subjects. The Y-axis shows the number of epochs with false positives.

Figure 5

Relationship between ROC curve and 10% false positives. (a) Linear relationship given in all files and EEG channels of the 20 subjects and (b) Relationship given in all files and capture of chin EMGs in the 20 subjects.

Figure 6

Success rate boxplot for awake/sleep detection on the evaluated subjects. (a) Case (iii): all EEG signals + EMG signals. Maximum, 97.04%; minimum, 70.57%; median, 88.93%; 25th percentile = 82.50%. (b) Case (iv): occipital EEG signals + EMG signals. Maximum, 99.91%; minimum, 80.84%; median, 95.38%; 25th percentile, 91.92%.

Figure 7

Boxplots of the apnea detection algorithm. (a) False positives of apneas, (b) False positives of hypopneas, and (c) Success rates in apneas + hypopneas.