Method for detecting sleep apnea and related events based on brain-computer interface

Abstract

The invention discloses a method for detecting sleep apnea and related events based on a brain-computer interface. The method comprises the following steps: collecting physiological signals during sleep; performing signal preprocessing operation on the physiological signal to obtain a preprocessed physiological signal; performing automatic sleep staging on electroencephalogram signals in the preprocessed physiological signals by adopting a deep learning model to obtain a sleep staging result; extracting features corresponding to the sleep staging results to obtain sleep feature data; positioning an apnea related event through the sleep feature data to obtain an event positioning result; calculating the event positioning result and the preprocessed physiological signal by adopting a statistical method to obtain event statistical data; calculating an apnea hypopnea index according to the sleep staging result, the event statistical data and the total sleep duration; and generating a sleep monitoring report according to the sleep staging result, the event statistical data and the apnea hypopnea index.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of brain-computer interface, in particular to a brain-computer interface method for detecting sleep apnea and related events in sleep medicine field. BACKGROUND

[0002] In the field of sleep medicine, the diagnosis of sleep apnea syndrome needs to rely on professional monitoring technology. The widely recognized and applied diagnostic method in the clinic is polysomnography. As a commonly used standard means for diagnosing sleep apnea syndrome, this device can monitor multiple physiological indicators during sleep and provide data support for clinical diagnosis.

[0003] However, the polysomnography has certain limitations in actual application. The device has high cost and large overall volume, and has specific requirements for the use environment, so it needs to be operated in a professional laboratory environment and is difficult to be deployed and used in daily scenes such as home, which limits the flexibility of the monitoring scene. At the same time, since many monitoring lines and sensors need to be connected to the patient when the device is used, it will cause certain interference to the patient's sleep, resulting in low cooperation degree of the patient in the monitoring process, and the long-term continuous monitoring of the patient's sleep condition cannot be realized, which is not conducive to the continuous tracking of the disease development trend.

[0004] In order to improve the above-mentioned problems, a portable sleep monitoring device appears in the prior art. This kind of device is optimized in volume and use convenience, and can meet the monitoring needs of non-laboratory scenes to a certain extent, but still has deficiencies in technical performance. On the one hand, this kind of portable device cannot combine electroencephalogram signals to work when detecting sleep breathing related events, and only relies on part of physiological indicators such as respiration and blood oxygen to make judgments, resulting in incomplete identification of respiratory events; on the other hand, for microstructure events in sleep process, such as awakening, micro-awakening, apnea related cortical response, etc., this kind of portable device lacks technical means to identify using electroencephalogram information, and it is difficult to capture key electroencephalogram features related to sleep quality and respiratory events, thereby affecting the accuracy of the diagnosis result of sleep apnea syndrome, and cannot meet the needs of clinical diagnosis accuracy and family long-term monitoring. SUMMARY

[0005] The purpose of the present application is to provide a sleep apnea and related event detection method based on brain-computer interface, which can improve the accuracy of the diagnosis result of sleep apnea syndrome and meet the needs of clinical diagnosis accuracy and family long-term monitoring.

[0006] In order to achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0007] A sleep apnea and related event detection method based on brain-computer interface, comprising the steps of:

[0008] acquiring a physiological signal during sleep;

[0009] performing a signal preprocessing operation on the physiological signal to obtain a preprocessed physiological signal;

[0010] performing automatic sleep staging on an electroencephalogram signal in the preprocessed physiological signal by using a deep learning model to obtain a sleep staging result;

[0011] extracting a feature corresponding to the sleep staging result to obtain sleep feature data;

[0012] locating a respiratory pause related event by using the sleep feature data to obtain an event location result;

[0013] performing calculation on the event location result and the preprocessed physiological signal by using a statistical method to obtain event statistical data;

[0014] calculating a respiratory pause hypopnea index according to the sleep staging result, the event statistical data, and a total sleep duration;

[0015] generating a sleep monitoring report according to the sleep staging result, the event statistical data, and the respiratory pause hypopnea index.

[0016] In a possible implementation, in the step of performing the signal preprocessing operation on the physiological signal, the signal preprocessing operation includes filtering processing, noise reduction processing, and artifact removal processing;

[0017] performing the filtering processing on the physiological signal to remove interference frequency components in the physiological signal;

[0018] performing the noise reduction processing on the physiological signal subjected to the filtering processing to reduce noise in the physiological signal;

[0019] performing the artifact removal processing on the physiological signal subjected to the noise reduction processing to remove artifact components in the physiological signal.

[0020] In a possible implementation, the step of performing the signal preprocessing operation on the physiological signal further includes resampling processing;

[0021] performing the resampling processing on the physiological signal subjected to the artifact removal processing to reduce the analysis and calculation amount of a subsequent deep learning model;

[0022] obtaining the preprocessed physiological signal by using the resampling processing.

[0023] In a possible implementation, in the step of automatically performing sleep staging on the electroencephalogram in the preprocessed physiological signal by using a deep learning model, the deep learning model is a pre-trained deep learning model.

[0024] The pre-trained deep learning model analyzes the electroencephalogram in the preprocessed physiological signal.

[0025] The sleep staging result is output through the analysis.

[0026] In a possible implementation, in the step of automatically performing sleep staging on the electroencephalogram in the preprocessed physiological signal by using a deep learning model, the sleep staging result includes a wake state and sleep states of different depths.

[0027] The pre-trained deep learning model classifies the electroencephalogram according to a set sleep staging standard.

[0028] The sleep staging result including the wake state and the sleep states of different depths is obtained through the classification.

[0029] In a possible implementation, in the step of extracting the features corresponding to the sleep staging result, the features include network features, spectral power features, time domain features, and frequency domain features.

[0030] The network features corresponding to the sleep staging result are extracted.

[0031] The spectral power features corresponding to the electroencephalogram in the preprocessed physiological signal are extracted.

[0032] The time domain features and the frequency domain features corresponding to the apnea-related events are extracted.

[0033] The network features, the spectral power features, the time domain features, and the frequency domain features are integrated to obtain the sleep feature data.

[0034] In a possible implementation, in the step of calculating the event positioning result and the preprocessed physiological signal by using a statistical method, the event statistical data includes the number of apnea events per unit time, the number of hypopnea events accompanied by specific physiological changes, and the number of micro-awakening events.

[0035] The total number of apnea events in the event positioning result is counted.

[0036] The total number of hypopnea events accompanied by specific physiological changes is counted according to the preprocessed physiological signal.

[0037] The total number of micro-awakening events in the event positioning result is counted.

[0038] Integrating the total number of apnea events, the total number of hypopnea events accompanied by specific physiological changes and the total number of micro-awakening events, the event statistics data is obtained.

[0039] In a possible implementation, in the step of calculating an apnea hypopnea index according to the sleep staging result, the event statistics data and the total sleep duration,

[0040] Adding the total number of apnea events and the total number of hypopnea events accompanied by specific physiological changes, the total number of events is obtained.

[0041] Dividing the total number of events by the total sleep duration;

[0042] Multiplying the quotient value by a set time unit coefficient, the apnea hypopnea index is obtained.

[0043] In a possible implementation, in the step of generating a sleep monitoring report according to the sleep staging result, the event statistics data and the apnea hypopnea index, the sleep monitoring report includes an all-night sleep staging result and a long-term sleep trend;

[0044] Extracting an all-night sleep stage distribution from the sleep staging result;

[0045] Analyzing a long-term event change according to the event statistics data and the apnea hypopnea index of multiple days;

[0046] Integrating the all-night sleep stage distribution and the long-term event change, the long-term sleep trend is formed.

[0047] Including the all-night sleep staging result, the long-term sleep trend and the apnea hypopnea index into the sleep monitoring report.

[0048] In a possible implementation, physiological signals during sleep are collected by using a portable device, the portable device is a portable brain-computer interface sleep instrument; the physiological signals include electroencephalogram signals, respiration signals and blood oxygen signals; the portable brain-computer interface sleep instrument collects the electroencephalogram signals through electrodes and collects the respiration signals and the blood oxygen signals through sensors.

[0049] In a possible implementation, the step of using the portable brain-computer interface sleep instrument to collect physiological signals during sleep further comprises a data transmission step; the physiological signals are transmitted to a mobile terminal through a wireless communication protocol; the mobile terminal performs preliminary processing on the physiological signals to obtain preliminary processed physiological signals; in the step of performing signal preprocessing operations on the physiological signals, the signal preprocessing operations are performed on the preliminary processed physiological signals.

[0050] In a possible implementation, the data transmission step further comprises cloud transmission; the preliminary processed physiological signals are transmitted from the mobile terminal to a cloud server through a network communication protocol; in the step of automatically performing sleep staging on the preprocessed physiological signals by using a deep learning model, the cloud server calls the deep learning model; the deep learning model performs automatic sleep staging based on the preprocessed physiological signals provided by the cloud server to obtain the sleep staging result.

[0051] In a possible implementation, the electrodes of the portable brain-computer interface sleep instrument comprise a plurality of channel electrodes; the plurality of channel electrodes are arranged at specific head positions; a reference electrode and a ground electrode of the portable brain-computer interface sleep instrument are arranged at specific positions behind the ears; the electroencephalogram signals are collected by the plurality of channel electrodes, the reference electrode, and the ground electrode working cooperatively.

[0052] In a possible implementation, the sensor of the portable brain-computer interface sleep instrument is an oximetry sensor; the oximetry sensor works at a set sampling rate; the oximetry sensor measures heart rate data and blood oxygen saturation data by using a photoplethysmogram technology; the heart rate data and the blood oxygen saturation data are included in the physiological signals for subsequent signal preprocessing operations.

[0053] In a possible implementation, the step of generating a sleep monitoring report further comprises a long-term data monitoring step; the physiological signals are continuously collected for multiple days; the signal preprocessing operations, automatic sleep staging, feature extraction, event positioning, statistical calculation, and apnea hypopnea index calculation steps are repeatedly performed on the physiological signals of each day; sleep staging results, event statistical data, and apnea hypopnea indexes of multiple days are obtained to form long-term sleep data; in the step of generating a sleep monitoring report, a change trend reflected by the long-term sleep data is included in the long-term sleep trend.

[0054] In a possible implementation, the step of generating a sleep monitoring report further comprises a pre-warning step;

[0055] Analyze the abnormal change in the long-term sleep trend; when the abnormal change reaches a set threshold, generate early warning information; and incorporate the early warning information into the sleep monitoring report for prompting a user or medical staff.

[0056] In a possible implementation, in the step of generating the sleep monitoring report, the sleep staging result, the event statistical data, the apnea hypopnea index, and the long-term sleep trend are presented in an easily understandable form.

[0057] The easily understandable form includes a chart and a textual description; and the sleep monitoring report is adapted for user viewing and clinical diagnosis reference through the chart and the textual description, to assist in making personalized treatment decisions.

[0058] In a second aspect, the present application further provides an electronic device, comprising a processor and a memory.

[0059] The memory is configured to store a program.

[0060] The processor executes the program to implement the method as described above.

[0061] In a third aspect, the present application further provides a computer readable storage medium, which stores a program, and the program is executed by a processor to implement the method as described above.

[0062] In a fourth aspect, the present application further provides a computer program product or a computer program, which comprises computer instructions stored in a computer readable storage medium. A processor of a computer device can read the computer instructions from the computer readable storage medium, and the processor executes the computer instructions to enable the computer device to perform the method as described above.

[0063] Compared with the prior art, the present application has the following beneficial effects:

[0064] The present application collects the electroencephalogram signal, the respiration signal, and the blood oxygen signal during sleep by using a portable brain-computer interface device. Compared with a traditional polysomnograph, the present application does not need to connect a large number of complex lines and sensors to the patient, greatly reduces the interference to the patient's sleep process, and is more easily adapted to and cooperated with the monitoring work by the patient, thereby enabling long-term continuous monitoring in a home scenario.

[0065] In the signal processing link, the collected physiological signals are sequentially subjected to filtering, noise reduction, artifact removal processing, and then resampling processing. The filtering processing can remove irrelevant frequency interference components in the physiological signals, the noise reduction processing can weaken the noise interference in the signals, the artifact removal processing can exclude the influence of abnormal artifacts on the data accuracy, and the data quality of the physiological signals is effectively improved. The resampling processing can reduce the data amount of the physiological signals, directly reduce the analysis and calculation amount of the subsequent deep learning model, speed up the data processing speed, and improve the efficiency of the overall detection process.

[0066] The present application introduces a deep learning model to automatically sleep stage the preprocessed electroencephalogram signals and extracts features corresponding to different sleep stages. The electroencephalogram signals can reflect information of sleep microstructure events such as wakefulness, micro-awakening, and respiratory pause related cortical responses during sleep. Combined with the analysis capability of the deep learning model for the electroencephalogram signals, the features corresponding to these sleep microstructure events can be captured, and events such as respiratory pause, hypopnea, hypoxemia, and micro-awakening can be comprehensively identified. The present application makes up for the defects of the traditional portable device that cannot combine electroencephalogram information and only relies on part of the physiological indicators, resulting in incomplete event identification, and improves the comprehensiveness and accuracy of the identification of sleep apnea and related events.

[0067] The present application supports long-term monitoring of sleep data, and can continuously collect sleep staging results, event statistical data, and respiratory pause hypopnea index for multiple days. Through integrated analysis of these long-term data, the trend of sleep apnea events can be formed, and warnings can be issued based on the trend changes. At the same time, a sleep monitoring report containing the whole night sleep staging results and long-term sleep trend is generated. The long-term data and report provide dynamic change information of the patient's sleep condition for the clinician, helping the doctor to more comprehensively grasp the patient's disease development, so as to develop a treatment plan more suitable for the individual situation of the patient, and improve the applicability of the treatment plan. For home users, they can also intuitively understand their own sleep health status through the report, meeting the daily health management needs. BRIEF DESCRIPTION OF DRAWINGS

[0068] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiments will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0069] Figure 1 The flowchart of the sleep apnea and related event detection method in the embodiment of the present application is shown in the figure.

[0070] Figure 2 The schematic diagram of collecting physiological signals during sleep by using a portable device in the embodiment of the present application is shown in the figure.

[0071] Figure 3 Flowchart of the process of collecting physiological signals during sleep and signal preprocessing operation in the embodiment of the present application;

[0072] Figure 4 Flowchart of the process of automatically sleep staging of the electroencephalogram signal in the preprocessed physiological signal by using the deep learning model in the embodiment of the present application;

[0073] Figure 5 Schematic diagram of the pre-trained deep learning model (UT-OSANet) in the embodiment of the present application;

[0074] Figure 6 Schematic diagram of the sleep monitoring report in the embodiment of the present application. DETAILED DESCRIPTION

[0075] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0076] Embodiment:

[0077] It should be noted that the terms "comprising" and "having" and any variations thereof in the embodiments of the present application are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device comprising a series of steps or units does not have to be limited to only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to the process, method, product or device.

[0078] Figure 1 Flowchart of the method for detecting sleep apnea and related events in the embodiment of the present application; see Figure 1 The embodiment of the present application provides a method for detecting sleep apnea and related events based on a brain-computer interface, comprising the steps of:

[0079] Step 1: Collecting physiological signals during sleep.

[0080] When specifically executed, the physiological signals during sleep can be collected by using a portable device, and the portable device is a portable brain-computer interface sleep instrument. The physiological signals include electroencephalogram signals, respiratory signals and blood oxygen signals. The portable brain-computer interface sleep instrument collects the electroencephalogram signals through electrodes and collects the respiratory signals and blood oxygen signals through sensors.

[0081] Figure 2A schematic diagram of collecting physiological signals during sleep using a portable device in embodiments of the present application; see Figure 2 The electrodes of the portable brain-computer interface sleep instrument include multiple channel electrodes; the multiple channel electrodes are arranged at specific head positions; the reference electrode and the ground electrode of the portable brain-computer interface sleep instrument are arranged at specific positions behind the ears; the electroencephalogram signals are collected through the cooperation of the multiple channel electrodes, the reference electrode and the ground electrode.

[0082] The sensor of the portable brain-computer interface sleep instrument is an oximetry sensor; the oximetry sensor works at a set sampling rate; the oximetry sensor measures heart rate data and blood oxygen saturation data through the photoplethysmography technology; the heart rate data and the blood oxygen saturation data are included in the physiological signals for subsequent signal preprocessing operations.

[0083] Exemplarily, the multi-modal physiological signal collection process of the portable brain-computer interface sleep instrument is shown in Figure 1 The device uses 4-channel (or single-channel) electroencephalogram electrodes, which are placed at positions F3, F4, E1, E2, etc., and the reference and ground electrodes are placed behind the mastoid of the ear. The sampling rate is 250 Hz, and the filter range is 0.5-45 Hz. In addition, the oximetry sensor equipped at the bottom operates at a sampling rate of 50 Hz, and uses the photoplethysmography technology to measure heart rate and blood oxygen saturation data.

[0084] Figure 3 A flowchart of collecting physiological signals during sleep and signal preprocessing operations in embodiments of the present application; see Figure 3 In some embodiments, the step of collecting physiological signals during sleep using the portable brain-computer interface sleep instrument further includes a data transmission step; the physiological signals are transmitted to a mobile terminal through a wireless communication protocol; the mobile terminal preliminarily processes the physiological signals to obtain preliminarily processed physiological signals; in the step of performing signal preprocessing operations on the physiological signals, the preliminarily processed physiological signals are used to perform signal preprocessing operations.

[0085] The data transmission step further includes cloud transmission; the preliminarily processed physiological signals are transmitted from the mobile terminal to a cloud server through a network communication protocol; in the step of performing automatic sleep staging on the electroencephalogram signals in the preprocessed physiological signals using a deep learning model, the cloud server calls the deep learning model; the deep learning model performs automatic sleep staging based on the preprocessed physiological signals provided by the cloud server to obtain sleep staging results.

[0086] Step 2: Perform signal preprocessing operations on the physiological signals to obtain preprocessed physiological signals.

[0087] In specific implementation, the signal preprocessing operation includes filtering processing, noise reduction processing and artifact removal processing; the physiological signal is filtered to remove interference frequency components in the physiological signal; the filtered physiological signal is subjected to noise reduction processing to reduce noise in the physiological signal; the noise-reduced physiological signal is subjected to artifact removal processing to remove artifact components in the physiological signal.

[0088] In some embodiments, the step of performing signal preprocessing operation on the physiological signal further includes resampling processing; the artifact-removed physiological signal is subjected to resampling processing to reduce the analysis and calculation amount of the subsequent deep learning model; the preprocessed physiological signal is obtained through resampling processing.

[0089] Step 3: The deep learning model is used to automatically perform sleep staging on the electroencephalogram in the preprocessed physiological signal to obtain a sleep staging result.

[0090] Figure 4 A flowchart of the process of using the deep learning model to automatically perform sleep staging on the electroencephalogram in the preprocessed physiological signal in the embodiments of the present application is shown in Figure 5 A schematic diagram of the pre-trained deep learning model (UT-OSANet) in the embodiments of the present application is shown in Figures 4-5 In specific implementation, the deep learning model is a pre-trained deep learning model; the pre-trained deep learning model analyzes the electroencephalogram in the preprocessed physiological signal; and the sleep staging result is output through analysis.

[0091] The sleep staging result includes a wake state and different depth sleep states; the pre-trained deep learning model classifies the electroencephalogram according to a set sleep staging standard; and the sleep staging result containing the wake state and different depth sleep states is obtained through classification.

[0092] Step 4: Features corresponding to the sleep staging result are extracted to obtain sleep feature data.

[0093] In specific implementation, the features include network features, spectral power features, time domain features and frequency domain features; the network features corresponding to the sleep staging result are extracted; the spectral power features corresponding to the electroencephalogram in the preprocessed physiological signal are extracted; the time domain features and frequency domain features corresponding to the apnea-related events are extracted; and the network features, spectral power features, time domain features and frequency domain features are integrated to obtain the sleep feature data.

[0094] Step 5: The sleep feature data is used to locate the apnea-related events to obtain an event location result.

[0095] Step 6: Statistical methods are used to calculate the event location result and the preprocessed physiological signal to obtain event statistical data.

[0096] The following is an overall description of steps 5-6. In specific implementation, the event statistical data includes the number of apnea events, the number of hypopnea events accompanied by specific physiological changes, and the number of micro-arousal events per unit time; the total number of apnea events in the event localization result; the total number of hypopnea events accompanied by specific physiological changes according to the pre-processed physiological signals; the total number of micro-arousal events in the event localization result; and the total number of apnea events, the total number of hypopnea events accompanied by specific physiological changes, and the total number of micro-arousal events are integrated to obtain the event statistical data.

[0097] Step 7: Calculate the apnea-hypopnea index according to the sleep staging result, the event statistical data, and the total sleep duration.

[0098] In specific implementation, the total number of apnea events and the total number of hypopnea events accompanied by specific physiological changes or the total number of micro-arousal events are added to obtain the total number of events; the total number of events is divided by the total sleep duration; and the quotient is multiplied by a set time unit coefficient to obtain the apnea-hypopnea index.

[0099] The specific formula is as follows:

[0100]

[0101] In the formula, Napnea is the total number of apnea events, Nhypopnea|desat or arousal is the total number of hypopnea events accompanied by specific physiological changes or the total number of micro-arousal events; Tsleep is the total sleep duration, 60 is the set time unit coefficient, and AHI is the apnea-hypopnea index.

[0102] Step 8: Generate a sleep monitoring report according to the sleep staging result, the event statistical data, and the apnea-hypopnea index.

[0103] Figure 6 FIG. 1 is a schematic diagram of a sleep monitoring report in an embodiment of the present application. Figure 6 In specific implementation, the sleep monitoring report includes the whole-night sleep staging result and the long-term sleep trend; the whole-night sleep stage distribution is extracted from the sleep staging result; the long-term event change is analyzed according to the event statistical data and the multi-day apnea-hypopnea index; the whole-night sleep stage distribution and the long-term event change are integrated to form the long-term sleep trend; and the whole-night sleep staging result, the long-term sleep trend, and the apnea-hypopnea index are included in the sleep monitoring report.

[0104] In some embodiments, the step of generating the sleep monitoring report further comprises a long-term data monitoring step; continuously collecting physiological signals for multiple days; repeatedly performing the signal preprocessing operation, automatic sleep staging, feature extraction, event positioning, statistical calculation and apnea hypopnea index calculation steps on the physiological signals of each day; obtaining sleep staging results, event statistical data and apnea hypopnea indexes for multiple days, forming long-term sleep data; in the step of generating the sleep monitoring report, the change trend reflected by the long-term sleep data is included in the long-term sleep trend.

[0105] The step of generating the sleep monitoring report further comprises a warning step; analyzing abnormal changes in the long-term sleep trend; generating warning information when the abnormal changes reach a set threshold; including the warning information in the sleep monitoring report for prompting the user or medical staff.

[0106] In the step of generating the sleep monitoring report, the sleep staging results, event statistical data, apnea hypopnea indexes and long-term sleep trend are presented in an easily understandable form; the easily understandable form includes charts and textual descriptions; the sleep monitoring report is suitable for user viewing and clinical diagnosis reference through the charts and textual descriptions, assisting in making personalized treatment decisions.

[0107] Based on the same inventive concept, the embodiments of the present application also provide an electronic device, which comprises a processor and a memory, the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set are loaded and executed by the processor to realize the brain-computer interface-based sleep apnea and related event detection method as described above.

[0108] It can be understood that the memory can include a random access memory (RAM) and a read-only memory (ROM). Optionally, the memory includes a non-transitory computer-readable storage medium. The memory can be used to store instructions, programs, codes, code sets or instruction sets. The memory can include a program storage area and a data storage area, wherein the program storage area can store instructions for implementing an operating system, instructions for at least one function, instructions for implementing various method embodiments described above, etc.; the data storage area can store data created according to the use of the server, etc.

[0109] The processor can include one or more processing cores. The processor connects various parts within the entire server by various interfaces and lines, executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory, and calling data stored in the memory. Alternatively, the processor can be implemented in at least one of a hardware form of a digital signal processing (DSP), a field-programmable gate array (FPGA), a programmable logic array (PLA). The processor can be integrated with one or several combinations of a central processing unit (CPU) and a modem. Among them, the CPU mainly processes operating systems and application programs; the modem is used to process wireless communication. It can be understood that the above-mentioned modem can also not be integrated into the processor, but be realized by a single chip.

[0110] Since the electronic device is the electronic device corresponding to the sleep apnea and related event detection method based on brain-computer interface of the embodiment of the application, and the principle of solving problems of the electronic device is similar to that of the method, the implementation of the electronic device can be referred to the implementation process of the above-mentioned method embodiment, and the repeated parts will not be described again.

[0111] Based on the same inventive concept, the embodiment of the present application also provides a computer readable storage medium, the storage medium stores at least one instruction, at least one program, code set or instruction set, the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to realize the sleep apnea and related event detection method based on brain-computer interface as described above.

[0112] Those skilled in the art can understand that all or part of the steps of various methods in the above embodiments can be completed by instructing relevant hardware through programs, and the programs can be stored in a computer readable storage medium, including Read-Only Memory (ROM), Random Access Memory (RAM), Programmable Read-only Memory (PROM), Erasable Programmable Read Only Memory (EPROM), One-time Programmable Read-Only Memory (OTPROM), Electrically-Erasable Programmable Read-Only Memory (EEPROM), Compact Disc Read-Only Memory (CD-ROM) or other optical disk storage, magnetic disk storage, magnetic tape storage, or any other medium that can be used to carry or store data in a computer readable manner.

[0113] Since the storage medium is the storage medium corresponding to the method for detecting sleep apnea and related events based on brain-computer interface according to the embodiments of the application, and the principle of solving problems of the storage medium is similar to the method, the implementation of the storage medium can be referred to the implementation process of the above method embodiments, and the repeated parts will not be described here.

[0114] In some possible implementation manners, various aspects of the method according to the embodiments of the application can also be implemented in the form of a program product, which includes program codes for causing a computer device to perform the steps of the method for detecting sleep apnea and related events based on brain-computer interface according to various exemplary embodiments of the present application described above in the specification when the program product is run on the computer device. Wherein, the executable computer program codes or "codes" for executing various embodiments can be written in high-level programming languages such as C, C++, C#, Smalltalk, Java, JavaScript, Visual Basic, Structured Query Language (for example, Transact-SQL), Perl, or in various other programming languages.

[0115] It should be understood that various aspects of the application can be implemented in hardware, software, firmware or a combination of them. In the above embodiments, various steps or methods can be implemented in software or firmware which is stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, and in another embodiment, any of the following technologies, known in the art, or their combination, can be used: discrete logic circuitry having logic gates for implementing logic functions upon an application of data signals, application-specific integrated circuits having appropriate combinational logic gates, programmable gate arrays (PGA), field-programmable gate arrays (FPGA), and so on.

[0116] In the description of the specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Also, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, a person skilled in the art can combine and combine the different embodiments or examples described in the specification and the features of the different embodiments or examples, without contradiction.

[0117] The above embodiments are only for the purpose of illustrating the technical concept and characteristics of the present application, and the purpose is to enable those skilled in the art to understand the content of the present application and to implement it, and cannot limit the protection scope of the present application. Any equivalent changes or modifications made in accordance with the essence of the present application should be covered within the protection scope of the present application.

Claims

1. A method for detecting sleep apnea and related events based on a brain-computer interface, characterized in that, The method comprises the steps of: collecting physiological signals during sleep; performing signal preprocessing operations on the physiological signals to obtain preprocessed physiological signals; using a deep learning model to automatically perform sleep staging on electroencephalogram signals in the preprocessed physiological signals to obtain sleep staging results; extracting features corresponding to the sleep staging results to obtain sleep feature data; locating apnea-related events through the sleep feature data to obtain event location results; calculating the event location results and the preprocessed physiological signals using statistical methods to obtain event statistics; calculating an apnea hypopnea index according to the sleep staging results, the event statistics, and the total sleep time; generating a sleep monitoring report according to the sleep staging results, the event statistics, and the apnea hypopnea index.

2. The method of claim 1, wherein the method is based on a brain-computer interface. In the step of performing signal preprocessing operations on the physiological signals, the signal preprocessing operations include filtering processing, noise reduction processing, and artifact removal processing; performing the filtering processing on the physiological signals to remove interference frequency components in the physiological signals; performing the noise reduction processing on the filtered physiological signals to reduce noise in the physiological signals; performing the artifact removal processing on the noise-reduced physiological signals to remove artifact components in the physiological signals.

3. The method of claim 2, wherein the method further comprises: The step of performing signal preprocessing operations on the physiological signals further includes resampling processing; performing the resampling processing on the artifact-removed physiological signals to reduce the analysis and calculation amount of subsequent deep learning models; obtaining the preprocessed physiological signals through the resampling processing.

4. The method of claim 1, wherein the method is based on a brain-computer interface. In the step of using a deep learning model to automatically perform sleep staging on electroencephalogram signals in the preprocessed physiological signals, the deep learning model is a pre-trained deep learning model; the pre-trained deep learning model analyzes the electroencephalogram signals in the preprocessed physiological signals; the sleep staging results are output through the analysis.

5. The method of claim 4, wherein the method further comprises: In the step of using a deep learning model to automatically perform sleep staging on electroencephalogram signals in the preprocessed physiological signals, the sleep staging results include wakefulness states and different depth sleep states; the pre-trained deep learning model classifies the electroencephalogram signals according to a set sleep staging standard; the sleep staging results containing wakefulness states and different depth sleep states are obtained through the classification.

6. The method of claim 4, wherein the method further comprises: In the step of extracting features corresponding to the sleep staging results, the features include network features, spectral power features, time domain features, and frequency domain features; extracting the network features corresponding to the sleep staging results; extracting the spectral power features corresponding to the electroencephalogram signals in the preprocessed physiological signals; extracting the time domain features and the frequency domain features corresponding to apnea-related events; integrating the network features, the spectral power features, the time domain features, and the frequency domain features to obtain the sleep feature data.

7. The method for detecting apnea and related events in sleep based on brain-computer interface according to claim 1, characterized in that, The event statistical data includes the number of apnea events per unit time, the number of hypopnea events accompanied by specific physiological changes, and the number of micro-awakening events. The total number of apnea events is counted from the event positioning results. The total number of hypopnea events accompanied by specific physiological changes is counted from the preprocessed physiological signals. The total number of micro-awakening events is counted from the event positioning results. The total number of apnea events, the total number of hypopnea events accompanied by specific physiological changes, and the total number of micro-awakening events are integrated to obtain the event statistical data.

8. The brain-computer interface-based sleep apnea and related event detection method of claim 1, wherein, In the step of calculating the apnea-hypopnea index according to the sleep staging results, the event statistical data, and the total sleep duration, the event statistical data includes the total number of apnea events and the total number of hypopnea events accompanied by specific physiological changes; The total number of apnea events and the total number of hypopnea events accompanied by specific physiological changes are added to obtain the total number of events; The total number of events is divided by the total sleep duration; The quotient is multiplied by a set time unit coefficient to obtain the apnea-hypopnea index.

9. The brain-computer interface-based sleep apnea and related event detection method of claim 1, wherein, In the step of generating a sleep monitoring report according to the sleep staging results, the event statistical data, and the apnea-hypopnea index, the sleep monitoring report includes the overnight sleep staging results and the long-term sleep trend; The overnight sleep stage distribution is extracted from the sleep staging results; The long-term event change is analyzed according to the event statistical data and the apnea-hypopnea index of multiple days; The overnight sleep stage distribution and the long-term event change are integrated to form the long-term sleep trend; The overnight sleep staging results, the long-term sleep trend, and the apnea-hypopnea index are included in the sleep monitoring report.

10. The brain-computer interface-based sleep apnea and related event detection method of claim 1, wherein, In the step of generating a sleep monitoring report, the sleep staging results, the event statistical data, the apnea-hypopnea index, and the long-term sleep trend are presented in an easily understandable form; The easily understandable form includes charts and textual descriptions; the sleep monitoring report is suitable for user viewing and clinical diagnosis reference through the charts and textual descriptions, and assists in making personalized treatment decisions.

Images