Sleep follow-up visit monitoring system for patients with chronic obstructive pulmonary disease

Abstract

The invention relates to the technical field of sleep staging, in particular to a chronic obstructive pulmonary disease patient sleep follow-up visit monitoring system which comprises the steps that BCG signals and electroencephalogram data of a chronic obstructive pulmonary disease patient are continuously obtained, and through the modes of preprocessing, feature extraction, sleep stage type judgment, sleep stage determination and anomaly detection, sleep follow-up visit monitoring of the chronic obstructive pulmonary disease patient is achieved. The sleep stage can be more accurately staging; according to the method, the accuracy of sleep analysis is improved, the dependence on detailed data analysis is reduced, the equipment computing power is saved, specific sleep staging can be conveniently and rapidly implemented, meanwhile, the method has the self-adjusting capacity, the sleep analysis standard can be dynamically adjusted according to the relation between the stage time ratio and the threshold interval, and the sleep quality is improved. The accuracy and reliability of sleep staging of the chronic obstructive pulmonary disease patient are further optimized, meanwhile, the system evaluates the sleep condition of the chronic obstructive pulmonary disease patient, and effective clinical guidance data are provided for clinicians.

Description

Technical Field

[0001] The invention relates to the technical field of sleep staging, and in particular to a sleep follow-up monitoring system for patients with chronic obstructive pulmonary disease. Background Art

[0002] Since patients with chronic obstructive pulmonary disease have reduced sleep quality due to respiratory factors, sleep indicators are a key indicator during follow-up. Therefore, monitoring sleep is essential. Among them, sleep staging is essential for understanding sleep structure and quality. It can help doctors diagnose sleep disorders such as insomnia, sleep apnea and periodic leg movement syndrome. By monitoring physiological parameters such as brain waves, heart rate, breathing patterns and body movements, different sleep stages can be identified, including light sleep, deep sleep and rapid eye movement (REM) sleep. Existing technologies usually rely on polysomnography for sleep staging, which is a comprehensive monitoring performed in a sleep laboratory, but this method is costly and complex to operate. In recent years, deep learning-based methods have gradually become a research hotspot due to their high accuracy and low cost. They can simplify the monitoring process by analyzing single-channel signals, providing more convenient sleep analysis tools for home and clinical environments.

[0003] Chinese patent publication number: CN111783534A, discloses a sleep staging method based on deep learning, which includes the following steps: oversampling a single-channel EEG signal to obtain a data set; designing a convolutional neural network for sleep staging; pre-training on the oversampled data set and adjusting the model's hyperparameters according to Bayesian optimization; designing a penalty weight loss function to optimize the model twice; using the trained model to test the input single-channel EEG signal to obtain the predicted sleep period;

[0004] However, there are still the following problems in the prior art:

[0005] All data are processed in a unified manner, the analysis process is complicated, and the computing power required of the equipment is too high. The sleep evaluation of patients with chronic obstructive pulmonary disease is inaccurate and cannot provide effective data support for clinical practice. Summary of the invention

[0006] To this end, the present invention provides a sleep follow-up monitoring system for patients with chronic obstructive pulmonary disease, which is used to overcome the problems in the prior art of unified operation and processing of all data, complex analysis process, and excessively high computing power required for equipment.

[0007] To achieve the above object, the present invention provides a sleep follow-up monitoring system for patients with chronic obstructive pulmonary disease, comprising:

[0008] A data acquisition module is used to continuously obtain the initial signal of the detection target when it is sleeping, and continuously obtain the EEG data of the detection target, the initial signal is a BCG signal;

[0009] A signal preprocessing module, which is used to preprocess the initial signal, remove invalid signal data, obtain an input signal, and draw a signal timing image;

[0010] A timing analysis module, which is used to extract the fluctuation characteristics of the input signal and divide the signal timing image into a number of characteristic timing segments;

[0011] A feature determination module, which is used to determine the detection target body motion characterization parameter according to the initial signal and the signal time sequence image, so as to determine the sleep stage type of the current feature time sequence segment;

[0012] A stage identification module, which is used to determine the sleep stage of each characteristic time sequence segment according to the sleep stage type combined with the fluctuation characteristics, including:

[0013] Obtaining the data segment and the corresponding time point corresponding to the respiratory frequency in the characteristic time series segment, and calculating the respiratory fluctuation characterization parameter to determine the sleep stage of the detection target;

[0014] Or, obtaining all peak values ​​of the input signal in the characteristic time sequence segment, calculating the signal peak value difference between the maximum peak value and the minimum peak value, and determining the sleep stage of the detection target according to the signal peak value difference;

[0015] The abnormality determination module compares the time corresponding to each of the sleep stages with the time corresponding to the sleep stage determined according to the EEG data to determine whether the sleep stage is abnormal.

[0016] Furthermore, the signal preprocessing module is used to obtain the input signal and draw a signal timing image including:

[0017] For comparing all the acquired initial signals with the sleep signal threshold;

[0018] for recording the initial signal data exceeding the sleep signal threshold as invalid signal data;

[0019] for removing the invalid signal data in the initial signal to obtain the input signal;

[0020] It is used to draw a signal timing image according to the acquisition order of the input signal.

[0021] Furthermore, the target body motion characterization parameters used by the feature determination module to determine the detected target include:

[0022] It is used to solve the mean value of the difference between the input signal and the average value of the input signal at each moment of the current characteristic time series;

[0023] The ratio of the mean value to a preset mean value threshold is used to obtain a first target body motion characterization factor;

[0024] The ratio of the initial signal data volume to the input signal data volume used to solve the current characteristic time sequence segment is the second target body motion characterization factor;

[0025] The product of the first target body motion representation factor and the second target body motion representation factor is determined as a target body motion representation parameter.

[0026] Furthermore, the feature determination module is used to determine the sleep stage type of the current feature time segment according to the detection target body motion characterization parameter, including:

[0027] If the detected target body motion characterization parameter is greater than or equal to the standard body motion characterization parameter, the sleep stage type of the current characteristic time sequence segment is a non-sleep stage type;

[0028] If the detected target body motion characterization parameter is less than the standard body motion characterization parameter, the sleep stage type of the current characteristic time segment is the asleep stage type.

[0029] Furthermore, the stage identification module is used to determine the sleep stage of each characteristic time segment according to the sleep stage type combined with the fluctuation characteristics, including:

[0030] If the sleep stage type is a non-sleep stage type, then obtaining the data segment corresponding to the respiratory frequency in the characteristic time series segment and the corresponding time point, and calculating the respiratory fluctuation characterization parameter to determine the sleep stage of the detection target;

[0031] If the sleep stage type is the fallen asleep stage type, all peak values ​​of the input signal in the characteristic time sequence segment are obtained, the signal peak difference between the maximum peak value and the minimum peak value is calculated, and the sleep stage of the detection target is determined according to the signal peak difference.

[0032] Furthermore, the phase identification module is used to determine the respiratory fluctuation characterization parameters including:

[0033] To determine the time interval to reach the peak of the respiratory rate within the current characteristic time period;

[0034] The variance of the time interval is determined as the respiratory fluctuation characteristic variable.

[0035] Furthermore, the stage identification module is used to determine the sleep stage of the detection target according to the respiratory fluctuation characterization parameter, including:

[0036] If the respiratory fluctuation characterization parameter is greater than or equal to the standard respiratory fluctuation characterization parameter, the sleep stage of the detection target is the rapid eye movement stage;

[0037] If the respiratory fluctuation characterization parameter is less than the standard respiratory fluctuation characterization parameter, the sleep stage of the detection target is the awake stage.

[0038] Furthermore, the stage identification module determines the sleep stage of the detection target according to the signal peak difference, including:

[0039] If the signal of the detection target meets the sleep analysis standard, the sleep stage of the detection target is the light sleep stage;

[0040] If the signal of the detection target does not meet the sleep analysis standard, the sleep stage of the detection target is the deep sleep stage;

[0041] The sleep analysis standard is that the peak average value is greater than or equal to the standard average value, or the difference between the first and last peak values ​​of the characteristic time sequence segment is greater than a preset stage difference.

[0042] Furthermore, the abnormality determination module is used to determine whether the sleep stage is abnormal, including:

[0043] to calculate the stage time ratio between the time corresponding to each sleep stage and the time of the sleep stage determined by the corresponding EEG data;

[0044] To determine whether the stage time ratio is within the stage time ratio threshold interval;

[0045] If the stage time ratio is not within the stage time ratio threshold interval, it is determined that the sleep staging is abnormal.

[0046] Furthermore, the abnormality determination module is used to determine the adjustment method of each standard parameter according to the stage time ratio and the stage time ratio threshold range when determining that the sleep staging is abnormal.

[0047] Compared with the prior art, the beneficial effect of the present invention lies in that the present invention provides a sleep follow-up monitoring system for patients with chronic obstructive pulmonary disease, which can more accurately stage the sleep stages by continuously acquiring BCG signals and EEG data of patients with chronic obstructive pulmonary disease, through preprocessing, feature extraction, sleep stage type judgment, sleep stage determination and abnormality detection; the present invention improves the accuracy of sleep analysis, and saves equipment computing power by reducing dependence on detailed data analysis, facilitates the rapid implementation of specific sleep staging, and also has self-adjustment capabilities, and can dynamically adjust the sleep analysis standards according to the relationship between the stage time ratio and the threshold interval, further optimizing the accuracy and reliability of sleep staging for patients with chronic obstructive pulmonary disease.

[0048] Furthermore, in the present invention, breathing, heartbeat and body movement conditions may be different in different sleep stages. Before entering a sleep state, a person may have certain muscle activities, such as turning over or irregular breathing and heartbeat when preparing to fall asleep. The corresponding body movement characterization parameters are calculated to characterize the sleep stage of the detection target, thereby providing data support for dividing the sleep stage type of the detection target. While ensuring reliability, detailed data analysis is reduced, equipment computing power is saved, and subsequent specific sleep staging is facilitated, which facilitates the evaluation of the sleep condition of COPD patients and provides effective clinical guidance data for clinicians.

[0049] Furthermore, in the present invention, when the body motion characterization parameter of the detection target is high, the change in breathing frequency is selected to determine the sleep state of the detection target, because the breathing in the awake state stage is normal and regular, and the breathing in the rapid eye movement stage (REM) becomes more irregular, showing intermittent deep and shallow breathing. Therefore, based on the above-mentioned distinguishing features, the sleep stage is determined according to the breathing stability, which ensures the accuracy of sleep staging and provides effective clinical guidance data for clinicians.

[0050] Furthermore, in the present invention, when the body motion characterization parameters of the detection target are low, the change in the overall data frequency is selected to determine the sleep state of the detection target. This is because compared with the specific performance of the detection target in the shallow sleep stage, the breathing in the deep sleep stage becomes slower and deeper and the heartbeat slows down further, presenting a more stable level. Therefore, based on the above-mentioned distinguishing features, the sleep stage is determined according to the peak value of the overall data, which ensures the accuracy of sleep staging, improves the accuracy of sleep assessment for COPD patients, and provides effective clinical guidance data for clinicians.

[0051] Furthermore, when the sleep staging is abnormal, the analysis standards, such as body movement characterization parameters, respiratory fluctuation characterization parameters and peak average values, can be automatically adjusted to adapt to individual differences and changes in sleep patterns, thereby improving the accuracy and personalized adaptability of sleep analysis, and improving the accuracy of sleep assessment for COPD patients, providing clinicians with effective clinical guidance data. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a structural block diagram of a sleep follow-up monitoring system for patients with chronic obstructive pulmonary disease according to an embodiment of the present invention;

[0053] Figure 2 A logic diagram for determining sleep stage types for an embodiment of the present invention;

[0054] Figure 3 A logic diagram of determining the sleep stage according to the respiratory fluctuation characterization parameter according to an embodiment of the present invention;

[0055] Figure 4This is a logic diagram for determining sleep stages according to sleep analysis standards according to an embodiment of the present invention. DETAILED DESCRIPTION

[0056] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0057] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0058] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0059] See also Figure 1 As shown, it is a structural block diagram of a sleep follow-up monitoring system for patients with chronic obstructive pulmonary disease according to an embodiment of the present invention; the present invention provides a sleep follow-up monitoring system for patients with chronic obstructive pulmonary disease, comprising:

[0060] A data acquisition module is used to continuously obtain the initial signal of the detection target when it is sleeping, and continuously obtain the EEG data of the detection target, the initial signal is a BCG signal;

[0061] A signal preprocessing module, which is used to preprocess the initial signal, remove invalid signal data, obtain an input signal, and draw a signal timing image;

[0062] A timing analysis module, which is used to extract the fluctuation characteristics of the input signal and divide the signal timing image into a number of characteristic timing segments;

[0063] A feature determination module, which is used to determine the detection target body motion characterization parameter according to the initial signal and the signal time sequence image, so as to determine the sleep stage type of the current feature time sequence segment;

[0064] A stage identification module, which is used to determine the sleep stage of each characteristic time sequence segment according to the sleep stage type combined with the fluctuation characteristics, including:

[0065] Obtaining the data segment and the corresponding time point corresponding to the respiratory frequency in the characteristic time series segment, and calculating the respiratory fluctuation characterization parameter to determine the sleep stage of the detection target;

[0066] Or, obtaining all peak values ​​of the input signal in the characteristic time sequence segment, calculating the signal peak value difference between the maximum peak value and the minimum peak value, and determining the sleep stage of the detection target according to the signal peak value difference;

[0067] The abnormality determination module compares the time corresponding to each of the sleep stages with the time corresponding to the sleep stage determined according to the EEG data to determine whether the sleep stage is abnormal.

[0068] In implementation, the data acquisition module can be an embedded piezoelectric sensor pillow to collect BCG signals during human sleep, and obtain sleep stage labels through brain-computer synchronous collection of EEG data.

[0069] Specifically, there is no limitation on the specific structures of the data acquisition module, signal preprocessing module, timing analysis module, stage identification module and abnormality determination module, and they can all be composed of logical components, including field-programmable processors, computers or microprocessors in computers.

[0070] It is understandable that a person's breathing, heart rate and body movements will vary in different sleep stages. The sleep stages include: (1) Awake stage: Breathing: Normal and regular breathing. Heart rate: Usually at a high level, but can vary depending on the person's activity level and emotional changes. Body movement: There may be some muscle activity, but it is usually controlled by the person's behavior and posture; (2) Rapid eye movement (REM) stage: Breathing: Breathing becomes more irregular, manifested as intermittent deep and shallow breathing. Heart rate: The heart rate usually becomes more irregular, and the heart rate may increase. Body movement: Muscles relax, and except for rapid eye movements, the muscles in other parts of the body are almost completely paralyzed, with almost no obvious body movement; (3) Light sleep stage: Breathing: Breathing is relatively stable, but may be slightly slower than the awake state. Heart rate: The heart rate gradually slows down, but still maintains a relatively stable level. Body movement: There may be some slight muscle activity, but it is usually below the level of the awake state; (4) Deep sleep stage: Breathing: Breathing becomes slower and deeper. Heart rate: The heart rate slows down further and presents a more stable level. Body movement: Muscles are relaxed to the maximum extent, with almost no body movement and minimal muscle activity.

[0071] In implementation, the fluctuation characteristics of the input signal include breathing frequency, heart rate and all peak values ​​of the input signal. There is no limitation on the specific method of extracting the fluctuation characteristics. A model capable of processing the signal can be pre-trained, and logic components can be imported to implement corresponding functions, which will not be elaborated here.

[0072] The present invention provides a sleep follow-up monitoring system for patients with chronic obstructive pulmonary disease. By continuously acquiring BCG signals and EEG data of the detection target, and through preprocessing, feature extraction, sleep stage type judgment, sleep stage determination and abnormality detection, the sleep stage can be staged more accurately. The present invention not only improves the accuracy of sleep analysis, but also saves equipment computing power by reducing dependence on detailed data analysis, and facilitates the rapid implementation of specific sleep staging. At the same time, it also has self-adjustment capabilities and can dynamically adjust the sleep analysis standard according to the relationship between the stage time ratio and the threshold interval, thereby further optimizing the accuracy and reliability of sleep staging.

[0073] Specifically, the signal preprocessing module is used to obtain the input signal and draw a signal timing image including:

[0074] For comparing all the acquired initial signals with the sleep signal threshold;

[0075] for recording the initial signal data exceeding the sleep signal threshold as invalid signal data;

[0076] for removing the invalid signal data in the initial signal to obtain the input signal;

[0077] It is used to draw a signal timing image according to the acquisition order of the input signal.

[0078] In practice, the BCG signal is a signal that describes the force exerted by the heart on the human body surface during beating. The initial signal collected by the piezoelectric film sensor is a superposition of three signals: body movement, breathing and cardiac impact, and is accompanied by power frequency noise. Usually, the human breathing frequency ranges from 10 to 25 times per minute, corresponding to a breathing frequency of about 0.1 to 0.4 Hz, and the heart rate is between 50 and 120 times per minute, corresponding to a heart rate of 0.9 to 2 Hz.

[0079] The sleep signal threshold is selected in the interval [4, 6].

[0080] See also Figure 2 As shown, it is a logic diagram for determining the sleep stage type according to an embodiment of the present invention. The characteristic determination module is used to determine the target body motion characterization parameters of the detection target, including:

[0081] It is used to solve the mean value of the difference between the input signal and the average value of the input signal at each moment of the current characteristic time series;

[0082] The ratio of the mean value to the preset mean value threshold is used to obtain the first target body motion characterization factor;

[0083] The ratio of the initial signal data volume to the input signal data volume used to solve the current characteristic time sequence segment is the second target body motion characterization factor;

[0084] The product of the first target body motion representation factor and the second target body motion representation factor is determined as a target body motion representation parameter.

[0085] In implementation, the preset mean threshold is obtained by pre-determination, and input signals of several monitoring targets during sleep are obtained. The difference between the input signal at each moment of each monitoring target input signal and the average value of the input signal is recorded, the mean of each difference is solved, and 0.8 times the mean is set as the preset mean threshold.

[0086] In implementation, the time series of the signal timing image may be discontinuous, but the image is a continuous image. For the characteristic timing segments of the signal timing image, the data volume of each characteristic timing segment may be different, but the time intervals are the same, which facilitates subsequent prediction and detection of the target sleep stage.

[0087] Specifically, the feature determination module is used to determine the sleep stage type of the current feature time segment according to the detection target body motion characterization parameter, including:

[0088] If the detected target body motion characterization parameter is greater than or equal to the standard body motion characterization parameter, the sleep stage type of the current characteristic time sequence segment is a non-sleep stage type;

[0089] If the detected target body motion characterization parameter is less than the standard body motion characterization parameter, the sleep stage type of the current characteristic time segment is the asleep stage type.

[0090] In implementation, the standard body motion characterization parameters are selected in the interval [1.1,1.15];

[0091] The sleep stages included in the non-sleeping stage types include the wakefulness stage and the rapid eye movement stage, and the sleeping stage types include the light sleep stage and the deep sleep stage.

[0092] In the present invention, breathing, heartbeat and body movement will be different in different sleep stages. Before entering the sleep state, people may have certain muscle activities, such as turning over or irregular breathing and heartbeat when preparing to fall asleep. The corresponding body movement characterization parameters are calculated to characterize the sleep stage of the detection target, and then provide data support for dividing the sleep stage type of the detection target. Under the premise of ensuring reliability, detailed data analysis is reduced, equipment computing power is saved, and subsequent specific sleep staging is facilitated, which is convenient for evaluating the sleep condition of COPD patients and providing effective clinical guidance data for clinicians.

[0093] Specifically, the stage identification module is used to determine the sleep stage of each characteristic time segment according to the sleep stage type combined with the fluctuation characteristics, including:

[0094] If the sleep stage type is a non-sleep stage type, then obtaining the data segment corresponding to the respiratory frequency in the characteristic time series segment and the corresponding time point, and calculating the respiratory fluctuation characterization parameter to determine the sleep stage of the detection target;

[0095] If the sleep stage type is the fallen asleep stage type, all peak values ​​of the input signal in the characteristic time sequence segment are obtained, the signal peak difference between the maximum peak value and the minimum peak value is calculated, and the sleep stage of the detection target is determined according to the signal peak difference.

[0096] See also Figure 3 As shown, it is a logic diagram of determining the sleep stage according to the respiratory fluctuation characterization parameter according to an embodiment of the present invention, which is used to determine the time interval for reaching the peak value of the respiratory frequency in the current characteristic time period;

[0097] The variance of the time interval is determined as the respiratory fluctuation characteristic variable.

[0098] It is understandable that the breathing frequency is around 0.1 to 0.4 Hz, and the heart rate is between 0.9 and 2 Hz, so the numerical values ​​corresponding to the two performances are quite different, and the corresponding signals can be screened and separated.

[0099] Specifically, the stage identification module is used to determine the sleep stage of the detection target according to the respiratory fluctuation characterization parameter, including:

[0100] If the respiratory fluctuation characterization parameter is greater than or equal to the standard respiratory fluctuation characterization parameter, the sleep stage of the detection target is the rapid eye movement stage;

[0101] If the respiratory fluctuation characterization parameter is less than the standard respiratory fluctuation characterization parameter, the sleep stage of the detection target is the awake stage.

[0102] In implementation, the standard respiratory fluctuation characterization parameters are obtained in advance by measuring the respiratory frequencies and corresponding time points during several sleep processes of several monitoring targets, calculating the mean of the respiratory fluctuation characterization parameters, and setting the standard respiratory fluctuation characterization parameters to 1.15 times the mean of the respiratory fluctuation characterization parameters.

[0103] In the present invention, when the body motion characterization parameter of the detection target is high, the change in breathing frequency is selected to determine the sleep state of the detection target, because the breathing in the awake state stage is normal and regular, and the breathing in the rapid eye movement stage (REM) becomes more irregular, showing intermittent deep and shallow breathing. Therefore, based on the above-mentioned distinguishing features, the sleep stage is determined according to the breathing stability, which ensures the accuracy of sleep staging and provides effective clinical guidance data for clinicians.

[0104] See also Figure 4As shown, it is a logic diagram of determining the sleep stage according to the sleep analysis standard in an embodiment of the present invention, the stage identification module determines the sleep stage of the detection target according to the signal peak difference, including that if the signal of the detection target meets the sleep analysis standard, the sleep stage of the detection target is a light sleep stage;

[0105] If the signal of the detection target does not meet the sleep analysis standard, the sleep stage of the detection target is the deep sleep stage;

[0106] The sleep analysis standard is that the peak average value is greater than or equal to the standard average value, or the difference between the first and last peak values ​​of the characteristic time sequence segment is greater than a preset stage difference.

[0107] In implementation, the standard average value is determined based on the median of the peak average value of the light sleep stage and the peak average value of the deep sleep stage in the historical data;

[0108] The difference between the peak average value of the light sleep stage and the peak average value of the deep sleep stage in the historical data is recorded as the reference stage difference, and the preset stage difference is 0.8 to 0.9 times the reference stage difference.

[0109] In the present invention, when the body motion characterization parameters of the detection target are low, the change in the overall data frequency is selected to determine the sleep state of the detection target. Compared with the specific performance of the detection target in the shallow sleep stage, the breathing in the deep sleep stage becomes slower and deeper and the heartbeat slows down further, presenting a more stable level. Therefore, based on the above-mentioned distinguishing features, the sleep stage is determined according to the peak value of the overall data, which ensures the accuracy of sleep staging, improves the accuracy of sleep assessment for COPD patients, and provides effective clinical guidance data for clinicians.

[0110] Specifically, the abnormality determination module is used to determine whether the sleep stage is abnormal, including:

[0111] Calculating the stage time ratio between the time corresponding to each sleep stage and the time of the corresponding sleep stage determined by the EEG data;

[0112] whether the stage time ratio is within a stage time ratio threshold interval;

[0113] If the stage time ratio is not within the stage time ratio threshold interval, it is determined that the sleep staging is abnormal.

[0114] In the implementation, the threshold interval of the stage time ratio is [0.95, 1.05], that is, the times corresponding to the sleep stages based on deep learning and EEG data are similar.

[0115] Specifically, the abnormality determination module is used to determine the adjustment method of each standard parameter according to the stage time ratio and the stage time ratio threshold range when determining that the sleep stage is abnormal.

[0116] In implementation, if the ratio of the time corresponding to the sleep stage included in the non-sleep stage type is less than 0.95, the standard body movement characterization parameter is reduced to 0.9 times the original value; if the ratio of the time corresponding to the sleep stage included in the non-sleep stage type is greater than 1.05, the standard body movement characterization parameter is increased to 1.1 times the original value;

[0117] If the ratio of the time corresponding to the rapid eye movement stage is less than 0.95, the standard respiratory fluctuation characterization parameter is reduced to 0.9 times the original value; if the ratio of the time corresponding to the rapid eye movement stage is greater than 1.05, the standard respiratory fluctuation characterization parameter is increased to 1.1 times the original value;

[0118] If the ratio of the time corresponding to the light sleep stage is less than 0.95, the standard average value is reduced to 0.9 times the original value; if the ratio of the time corresponding to the light sleep stage is greater than 1.05, the standard average value is increased to 1.1 times the original value.

[0119] When the sleep staging is abnormal, it can automatically adjust the analysis standards, such as body movement characterization parameters, respiratory fluctuation characterization parameters and peak average values, to adapt to individual differences and changes in sleep patterns, thereby improving the accuracy and personalized adaptability of sleep analysis, and improving the accuracy of sleep assessment for COPD patients, providing effective clinical guidance data for clinicians.

[0120] Specifically, after the stage recognition module completes the sleep stage recognition, technical personnel in this field can further evaluate the sleep quality based on the duration of the sleep stage. For example, during the evaluation, the duration of each sleep stage of healthy people can be obtained in advance, and the sleep quality can be determined based on the average of the difference ratio between the duration of each sleep stage of the current monitoring target and the duration of the sleep stage of the healthy people. The sleep quality can be rated according to the interval in which the difference ratio lies. Optionally, if the difference ratio is in [0%, 15%), the sleep quality is excellent; if the difference ratio is in (15%, 25%), the sleep quality is medium; if the difference ratio is in (25%, +∞], the sleep quality is poor.

[0121] If the sleep follow-up monitoring system for patients with chronic obstructive pulmonary disease of the present invention is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read~Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0122] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0123] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A sleep follow-up monitoring system for patients with chronic obstructive pulmonary disease, characterized in that: include: A data acquisition module is used to continuously obtain the initial signal of the detection target when it is sleeping, and continuously obtain the EEG data of the detection target, the initial signal is a BCG signal; A signal preprocessing module, which is used to preprocess the initial signal, remove invalid signal data, obtain an input signal, and draw a signal timing image; A timing analysis module, which is used to extract the fluctuation characteristics of the input signal and divide the signal timing image into a number of characteristic timing segments; A feature determination module, which is used to determine the detection target body motion characterization parameter according to the initial signal and the signal time sequence image, so as to determine the sleep stage type of the current feature time sequence segment; A stage identification module, which is used to determine the sleep stage of each characteristic time sequence segment according to the sleep stage type combined with the fluctuation characteristics, including: Obtaining the data segment and the corresponding time point corresponding to the respiratory frequency in the characteristic time series segment, and calculating the respiratory fluctuation characterization parameter to determine the sleep stage of the detection target; Or, obtaining all peak values ​​of the input signal in the characteristic time sequence segment, calculating the signal peak difference between the maximum peak value and the minimum peak value, and determining the sleep stage of the detection target according to the signal peak difference; The abnormality determination module compares the time corresponding to each of the sleep stages with the time corresponding to the sleep stage determined according to the EEG data to determine whether the sleep stage is abnormal.

2. The sleep follow-up monitoring system for patients with chronic obstructive pulmonary disease according to claim 1, characterized in that: The signal preprocessing module is used to obtain the input signal and draw a signal timing image, including: For comparing all the acquired initial signals with the sleep signal threshold; for recording the initial signal data exceeding the sleep signal threshold as invalid signal data; for removing the invalid signal data in the initial signal to obtain the input signal; It is used to draw a signal timing image according to the acquisition order of the input signal.

3. The sleep follow-up monitoring system for patients with chronic obstructive pulmonary disease according to claim 2, characterized in that: The target body motion characterization parameters used by the feature determination module to determine the detected target include: It is used to solve the mean value of the difference between the input signal and the average value of the input signal at each moment of the current characteristic time series; The ratio of the mean value to the preset mean value threshold is used to obtain the first target body motion characterization factor; The ratio of the initial signal data volume to the input signal data volume used to solve the current characteristic time sequence segment is the second target body motion characterization factor; The product of the first target body motion representation factor and the second target body motion representation factor is determined as a target body motion representation parameter.

4. The sleep follow-up monitoring system for patients with chronic obstructive pulmonary disease according to claim 3, characterized in that: The feature determination module is used to determine the sleep stage type of the current feature time segment according to the detection target body motion characterization parameter, including: If the detected target body motion characterization parameter is greater than or equal to the standard body motion characterization parameter, the sleep stage type of the current characteristic time sequence segment is a non-sleep stage type; If the detected target body motion characterization parameter is less than the standard body motion characterization parameter, the sleep stage type of the current characteristic time segment is the asleep stage type.

5. The sleep follow-up monitoring system for patients with chronic obstructive pulmonary disease according to claim 4, characterized in that: The stage identification module is used to determine the sleep stage of each characteristic time segment according to the sleep stage type combined with the fluctuation characteristics, including: If the sleep stage type is a non-sleep stage type, then obtaining the data segment corresponding to the respiratory frequency in the characteristic time series segment and the corresponding time point, and calculating the respiratory fluctuation characterization parameter to determine the sleep stage of the detection target; If the sleep stage type is the fallen asleep stage type, all peak values ​​of the input signal in the characteristic time sequence segment are obtained, the signal peak difference between the maximum peak value and the minimum peak value is calculated, and the sleep stage of the detection target is determined according to the signal peak difference.

6. The sleep follow-up monitoring system for patients with chronic obstructive pulmonary disease according to claim 5, characterized in that: The phase identification module is used to determine the respiratory fluctuation characterization parameters including: To determine the time interval to reach the peak of the respiratory rate within the current characteristic time period; The variance of the time interval is determined as the respiratory fluctuation characteristic variable.

7. The sleep follow-up monitoring system for patients with chronic obstructive pulmonary disease according to claim 6, characterized in that: The stage identification module is used to determine the sleep stage of the detection target according to the respiratory fluctuation characterization parameter, including: If the respiratory fluctuation characterization parameter is greater than or equal to the standard respiratory fluctuation characterization parameter, the sleep stage of the detection target is the rapid eye movement stage; If the respiratory fluctuation characterization parameter is less than the standard respiratory fluctuation characterization parameter, the sleep stage of the detection target is the awake stage.

8. The sleep follow-up monitoring system for patients with chronic obstructive pulmonary disease according to claim 7, characterized in that: The stage identification module determines the sleep stage of the detection target according to the signal peak difference, including: If the signal of the detection target meets the sleep analysis standard, the sleep stage of the detection target is the light sleep stage; If the signal of the detection target does not meet the sleep analysis standard, the sleep stage of the detection target is the deep sleep stage; The sleep analysis standard is that the peak average value is greater than or equal to the standard average value, or the difference between the first and last peak values ​​of the characteristic time sequence segment is greater than a preset stage difference.

9. The sleep follow-up monitoring system for patients with chronic obstructive pulmonary disease according to claim 8, characterized in that: The abnormality determination module is used to determine whether the sleep stage is abnormal, including: to calculate the stage time ratio between the time corresponding to each sleep stage and the time of the sleep stage determined by the corresponding EEG data; To determine whether the stage time ratio is within the stage time ratio threshold interval; If the stage time ratio is not within the stage time ratio threshold interval, it is determined that the sleep staging is abnormal.

10. The sleep follow-up monitoring system for patients with chronic obstructive pulmonary disease according to claim 9, characterized in that: The abnormality determination module is used to determine the adjustment method of each standard parameter according to the stage time ratio and the stage time ratio threshold range when determining that the sleep stage is abnormal.

Images