Rapid eye movement detection method based on chest and abdominal breathing signals during sleep

Abstract

The present application proposes a rapid eye movement (REM) detection method based on chest and abdominal breathing signals during sleep, comprising: calculating the correlation between the chest breathing signal and the abdominal breathing signal of a subject; calculating the correlation coefficient between the chest breathing signal and the abdominal breathing signal in different windows by adding a first sliding window to the chest breathing signal and the abdominal breathing signal respectively, to form a correlation coefficient sequence; adding a second sliding window to the correlation coefficient sequence to obtain a mean sequence and a standard deviation sequence, wherein the mean represents the amplitude and the standard deviation represents the fluctuation; obtaining a difference sequence by the difference between the mean sequence and the standard deviation sequence; setting a threshold value of 0.5 for the difference sequence, and determining the sleep stage as REM when the value of the difference sequence is less than the threshold value.

Description

Technical Field

[0001] The present application relates to sleep staging technology, and in particular to a rapid eye movement (REM) detection method based on chest and abdominal breathing signals during sleep. Background Art

[0002] Sleep is an essential physiological process in human life. Rapid Eye Movement (REM), a specialized stage of sleep, possesses unique physiological and psychological characteristics that significantly impact human physical and mental health and cognitive function. In-depth research on REM is crucial for understanding sleep mechanisms, diagnosing and treating sleep disorders, and cognitive neuroscience.

[0003] For the detection of REM periods, polysomnography (PSG) is currently one of the most commonly used and accurate methods for marking REM periods. It places multiple electrodes on the head, face, body and other parts of the sleeper, and simultaneously records multiple physiological signals such as electroencephalogram (EEG), electrooculogram (EOG), electromyography, electrocardiogram, and respiration. According to the standards of the American Academy of Sleep Medicine (AASM), the marking of REM periods usually requires the following conditions to be met at the same time: EEG shows low-voltage, mixed-frequency brain electrical activity, EOG detects rapid eye movements, and mandibular electromyography activity is at a low level, indicating muscle relaxation. However, PSG testing needs to be performed in a professional sleep laboratory, and the subjects need to sleep in an unfamiliar environment, which may affect the sleep quality and the accuracy of the test results; the equipment is complex and cumbersome to operate, requiring professional technicians to operate and analyze the data; the test cost is high and takes a long time, and generally requires the subjects to be monitored overnight in the laboratory.

[0004] In addition to PSG devices, some studies have utilized wearable devices and EEG monitoring devices to detect REM sleep. Wearable devices use sensors such as accelerometers and gyroscopes to monitor changes in movement and posture. Combined with data such as heart rate variability from heart rate sensors, algorithms are used to infer sleep stages, including REM sleep. This method is easy to implement, allowing subjects to monitor their sleep in a familiar environment without disrupting their daily lives. It can also monitor sleep data for multiple days, providing long-term sleep information. However, the algorithms in this method are primarily based on big data models and statistical laws, which have limited adaptability to individual differences. Furthermore, EEG monitoring devices analyze sleep states by collecting scalp EEG signals. During REM sleep, EEG activity exhibits specific frequency and waveform changes, such as low-amplitude and mixed-frequency EEG signals. These characteristics are then used to identify REM sleep. Compared to wearable devices, EEG monitoring devices offer higher accuracy in detecting sleep stages, particularly for REM sleep, providing a better reference for determining REM sleep. They can monitor EEG signal changes in real time, providing direct evidence for studying brain activity during sleep. However, the EEG signals during rapid eye movement and those during wakefulness have similarities in frequency characteristics and EEG activities in brain areas related to eye movements.

[0005] If there were a more convenient and simple method for accurately detecting and identifying the REM phase, it would be of great significance for research and clinical applications in the fields of sleep health and neuroscience. Summary of the Invention

[0006] This application aims to propose a rapid eye movement detection method.

[0007] The rapid eye movement (REM) detection method based on chest and abdominal breathing signals during sleep of the present application includes:

[0008] Calculating the correlation between the chest respiration signal and the abdominal respiration signal of the subject; calculating the correlation coefficient between the chest respiration signal and the abdominal respiration signal in different windows by adding a first sliding window to the chest respiration signal and the abdominal respiration signal respectively, to form a correlation coefficient sequence;

[0009] The second sliding window is added to the correlation coefficient sequence to obtain the mean sequence and standard deviation sequence, where the mean represents the amplitude and the standard deviation represents the fluctuation; the difference sequence is obtained by the difference between the mean sequence and the standard deviation sequence;

[0010] The threshold of the difference sequence is set to 0.5. When the value of the difference sequence is less than the threshold, the sleep stage is determined to be rapid eye movement.

[0011] Preferably, the correlation coefficient is a Pearson correlation coefficient or a Spearman correlation coefficient.

[0012] Preferably, the window size of the first sliding window is 10 seconds, and the step size is 1 second.

[0013] Preferably, the window size of the second sliding window is 100 seconds, and the step size is 1 second.

[0014] The present application determines the REM period by utilizing the correlation between the chest respiratory signal and the abdominal respiratory signal and the REM period, and can significantly distinguish the REM period from other sleep stages. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 These are the chest and abdominal respiratory signals corresponding to different sleep stages and their short-term correlation coefficients. DETAILED DESCRIPTION

[0016] Figure 1 In the figure, a is the chest respiratory signal during sleep, b is the abdominal respiratory signal during sleep, a and b are signals collected synchronously during sleep, and c is the sliding window correlation coefficient sequence of the two respiratory signals a and b. Specifically, by adding sliding windows to the chest respiratory signal and the abdominal respiratory signal respectively (the window size is 10 seconds, and the sliding window step is 1 second), the correlation coefficient between the two respiratory signals in different windows is calculated to form a correlation coefficient sequence, i.e., the waveform data in c; d and e are the mean time series and standard deviation time series of the sliding window of the correlation coefficient sequence c (the window size is 100 seconds, and the sliding window step is 1 second), respectively, f is the difference between the mean time series and the standard deviation time series, g is the manually labeled sleep stage, which is the clinical gold standard for sleep staging, and the red part is the manually labeled REM stage.

[0017] observe Figure 1 Figures c, d, and e show that during REM sleep, the correlation coefficients of the chest and abdominal respiratory signals tend to decrease overall, with large fluctuations. Based on this, we calculate the mean d (indicating a small amplitude) and standard deviation e (indicating a large fluctuation) of the correlation coefficients corresponding to the REM sleep stage and then take the difference between them to detect REM sleep. Based on this, we set a threshold of 0.5 for the difference sequence f. When the value of the sequence in f is less than the threshold, the sleep stage is considered to be REM sleep.

[0018] These findings indicate that the correlation between chest and abdominal respiratory signals during REM sleep differs significantly from that observed in other sleep stages. This correlation can be expressed using correlation coefficients (Pearson and Spearman) and cross-correlations. Quantitative assessment of correlation can be performed using one or more of the following methods: mean, standard deviation, short-term zero-crossing rate, short-term relative energy, median, and quartiles.

[0019] Unless otherwise defined, all technical and / or scientific terms used in this application have the same meaning as commonly understood by one of ordinary skill in the art to which the invention relates. The materials, methods, and examples mentioned in this application are illustrative only and not restrictive.

[0020] Although the present invention has been described in conjunction with specific embodiments, those skilled in the art may make appropriate substitutions, modifications and changes within the scope of the invention of this application, and such substitutions, modifications and changes shall still fall within the scope of protection of this application.

Claims

1. A method for detecting rapid eye movement (REM) based on chest and abdominal breathing signals during sleep, comprising: Calculating the correlation between the subject's chest respiratory signal and abdominal respiratory signal; By adding a first sliding window to the chest respiration signal and the abdominal respiration signal respectively, the correlation coefficient between the chest respiration signal and the abdominal respiration signal in different windows is calculated to form a correlation coefficient sequence; The second sliding window is added to the correlation coefficient sequence to obtain the mean sequence and standard deviation sequence, where the mean represents the amplitude and the standard deviation represents the fluctuation; the difference sequence is obtained by the difference between the mean sequence and the standard deviation sequence; The threshold of the difference sequence is set to 0.

5. When the value of the difference sequence is less than the threshold, the sleep stage is determined to be rapid eye movement.

2. The method for detecting rapid eye movement (REM) based on chest and abdominal breathing signals during sleep according to claim 1, wherein: The correlation coefficient is a Pearson correlation coefficient or a Spearman correlation coefficient.

3. The rapid eye movement (REM) detection method based on thoracic and abdominal breathing signals during sleep according to claim 1, characterized in that: The window size of the first sliding window is 10 seconds, and the step size is 1 second.

4. The method for rapid eye movement (REM) detection based on chest and abdominal breathing signals during sleep according to claim 1, wherein: The window size of the second sliding window is 100 seconds, and the step size is 1 second.

Images