Sleep monitoring system and method based on multi-sensor fusion

Abstract

The invention provides a sleep monitoring system and method based on multi-sensor fusion. The sleep monitoring system comprises a host and the following modules arranged in the host: a main control module; the electroencephalogram acquisition module is connected with the main control module and is used for acquiring electroencephalogram signals; the multi-axis motion sensor module is connected with the main control module and used for collecting motion signals; the multi-axis motion sensor module at least comprises an accelerometer for collecting acceleration and a gyroscope for collecting angular velocity; the main control module is responsible for controlling the electroencephalogram acquisition module and the multi-axis motion sensor module to acquire data and can process the acquired data in real time. The sleep monitoring method is realized based on the sleep monitoring system. The system is provided with the multi-axis motion sensor module, acceleration data and angular velocity data can be obtained, the main control module can effectively combine the angular velocity data and the acceleration data, turning body motion and breathing motion are effectively distinguished, and misjudgment is reduced.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, and in particular to a sleep monitoring system and method based on multi-sensor fusion. Background Technology

[0002] Sleep apnea syndrome is a common sleep disorder that requires monitoring for apnea events during sleep. An apnea event is defined as the disappearance or significant reduction of airflow through the mouth and nose during sleep (a decrease of ≥90% from the limit) for a duration of ≥10 seconds.

[0003] Currently, there are two main types of devices on the market for monitoring sleep apnea events: One type is the traditional polysomnography (PSG) device, as shown in the existing patent document with publication number CN109745002A. It needs to be used in hospitals and has the drawbacks of being complex, expensive and inconvenient to use.

[0004] Another type is the portable sleep monitoring device shown in the existing patent document with publication number CN 109222961A. Compared with polysomnography (PSG), it has the advantages of small size and ease of use. However, most portable sleep monitoring devices only use a single sensor, the accelerometer, to monitor body movement, which has the following drawbacks: 1. Using only an accelerometer cannot effectively distinguish between body movements (such as turning over) and respiratory movements, which can easily lead to misjudgment of apnea events.

[0005] 2. When body movement occurs, the respiratory signal in the acceleration signal will be interfered with, leading to missed detection of apnea events.

[0006] 3. Lack of multi-sensor synchronous acquisition and processing mechanism results in low monitoring accuracy. Summary of the Invention

[0007] In view of the shortcomings of the prior art described above, the purpose of this invention is to provide a sleep monitoring system based on multi-sensor fusion, which can at least solve one of the problems in the prior art.

[0008] To achieve the above and other related objectives, the present invention provides a sleep monitoring system based on multi-sensor fusion, comprising a host and the following modules disposed on the host: Main control module; The EEG acquisition module, connected to the main control module, is used to acquire EEG signals; A multi-axis motion sensor module, connected to the main control module, is used to acquire motion signals; The multi-axis motion sensor module includes at least an accelerometer for acquiring acceleration and a gyroscope for acquiring angular velocity; The main control module is responsible for controlling the EEG acquisition module and the multi-axis motion sensor module to acquire data, and can process the acquired data in real time.

[0009] In some implementations, the EEG acquisition module can continuously acquire EEG data and transmit the acquired EEG data to the main control module, the accelerometer can continuously acquire acceleration data and transmit the acquired acceleration data to the main control module, and when the accelerometer detects an abnormal change in acceleration, the main control module controls the gyroscope to turn on.

[0010] In some implementations, the criterion for judging abnormal acceleration changes is an acceleration change greater than 0.3g (i.e. 0.3 times the gravitational acceleration).

[0011] In some implementations, the main control module can detect apnea events based on acceleration data collected by the accelerometer.

[0012] In some implementations, the main control module first extracts Y-axis data from the acceleration data collected by the multi-axis motion sensor module, then performs bandpass filtering on the Y-axis data to extract the breathing signal, monitors the amplitude of the breathing signal in real time, and determines a breathing apnea event when the amplitude is continuously lower than the first set threshold for more than 10 seconds, and records the type and timestamp of the breathing apnea event.

[0013] In some implementations, when the gyroscope is turned on, the main control module can detect rollover events based on the angular velocity data collected by the gyroscope.

[0014] In some implementations, the main control module determines whether a rolling over event has occurred based on the angular velocity data collected by the gyroscope. When a rolling over event occurs, the main control module suspends the determination of the breathing apnea event.

[0015] In some implementations, the criteria for determining a rolling over event are that the angular velocity in the Z-axis direction is continuously greater than 45° / s and exceeds 500ms.

[0016] In some implementations, the host includes a housing, a first circuit board, and a first connector. The first circuit board is installed inside the housing, and the first connector is installed at the bottom of the housing and electrically connected to the first circuit board. The EEG acquisition module includes a soft adhesive patch, a second circuit board, an EEG acquisition chip, a second connector, and EEG electrodes. The soft adhesive patch is attached to the user's forehead. The second circuit board is installed inside the soft adhesive patch. The EEG electrodes are installed at the bottom of the soft adhesive patch and connected to the second circuit board. The second connector is installed at the top of the soft adhesive patch and electrically connected to the second circuit board. The second connector cooperates with the first connector. The main control module, the EEG acquisition chip, and the multi-axis motion sensor module are all integrated on the first circuit board. The first connector is electrically connected to the EEG acquisition chip.

[0017] In some implementations, the sleep monitoring system based on multi-sensor fusion can be used in conjunction with a host computer. The main control module includes a first Bluetooth communication module, and the host computer includes a second Bluetooth communication module. The main control module and the host computer are wirelessly connected via Bluetooth. Thus, the main control module is responsible for controlling the EEG acquisition module to collect EEG data, and the multi-axis motion sensor module to collect acceleration and angular velocity data. The main control module then processes this data in real time and transmits the results to the host computer via the Bluetooth module to monitor the patient's sleep quality and whether sleep apnea occurs.

[0018] To achieve the above and other related objectives, the present invention also provides a sleep monitoring method based on multi-sensor fusion, which is based on the aforementioned sleep monitoring system. The sleep monitoring method includes at least the following steps: S1. Initialization and parameter settings: Main control module initialization The system includes an EEG acquisition module and a multi-axis motion sensor module, configured with sampling rate and range. Among them, the EEG acquisition module is configured with a sampling rate of 250Hz, and the accelerometer of the multi-axis motion sensor module is configured with a sampling rate of 20Hz and a range of ±2g. S2. Data Acquisition and Synchronization: The EEG acquisition module continuously acquires EEG data and transmits it to the main control module; The accelerometer continuously collects acceleration data and transmits it to the main control module; The main control module uses its internal RTC to add a precise timestamp to each data packet to ensure data synchronization; S3. Respiratory signal extraction: The main control module extracts Y-axis data from the acceleration data collected by the multi-axis motion sensor module, and performs bandpass filtering of 0.1-0.5Hz on the Y-axis data to extract the breathing signal; S4. Apnea event detection: The main control module monitors the amplitude of the respiratory signal in real time. When the amplitude remains below the first set threshold for more than 10 seconds, it is determined to be a breathing apnea event. The main control module records the type and timestamp of the apnea event; S5, Turning Over Event Detection: When the accelerometer detects motion exceeding the second set threshold, the gyroscope is activated, with a sampling rate of 50Hz. The gyroscope continuously collects angular velocity data and transmits it to the main control module; The main control module uses the angular velocity data from the gyroscope to determine whether a rollover event has occurred; Pause step S4 when a rolling over event is detected.

[0019] In some implementations, the method further includes the following after step S5: S6. Comprehensive Judgment and Data Transmission: The main control module can make a comprehensive judgment by combining EEG signals and rolling over events; The main control module can store and transmit detected apnea events and turning events along with their timestamps.

[0020] As described above, the sleep monitoring system and method based on multi-sensor fusion of the present invention have the following beneficial effects: 1. This invention provides a novel sleep monitoring system with a multi-axis motion sensor module that can acquire acceleration and angular velocity data. The main control module can effectively combine the angular velocity and acceleration data to effectively distinguish between turning over and breathing movements, reducing misjudgments.

[0021] 2. The sleep monitoring system of the present invention is a lower-level machine, which is attached to the user's forehead during use. The main control module can realize the detection algorithm for determining apnea events and turning events, and process the data obtained by the EEG signal acquisition module and the multi-axis motion sensor module in real time. It does not rely on the upper-level machine and has a fast response.

[0022] 3. By using a timestamp synchronization mechanism, the timing of EEG signals and body movement signals is aligned, thereby improving the accuracy of data fusion.

[0023] 4. The sleep monitoring system of the present invention integrates multiple sensors such as EEG collector, accelerometer and gyroscope. By working together, the accuracy and reliability of the results can be greatly improved. Attached Figure Description

[0024] Figure 1 This is a three-dimensional structural diagram of the sleep monitoring system based on multi-sensor fusion according to the present invention; Figure 2 for Figure 1 The diagram shows a wearable sleep monitoring system based on multi-sensor fusion. Figure 3 for Figure 1 The diagram shows a three-dimensional structure of a sleep monitoring system based on multi-sensor fusion from another angle. Figure 4 for Figure 1 The diagram shown is a top view of the sleep monitoring system based on multi-sensor fusion. Figure 5 for Figure 4 The diagram shows a cross-sectional structure of a sleep monitoring system based on multi-sensor fusion along the AA direction. Figure 6 for Figure 5 A three-dimensional structural schematic diagram of the first circuit board and the electronic components integrated on the first circuit board shown. Figure 7 for Figure 6 A bottom view of the structure of the first circuit board shown; Figure 8 This is a simplified system block diagram of the sleep monitoring system based on multi-sensor fusion according to the present invention; Figure 9 This is a logic diagram of the sleep monitoring system and method based on multi-sensor fusion according to the present invention. Figure 10 This is an architecture diagram of the sleep monitoring system based on multi-sensor fusion according to the present invention.

[0025] Figure 1-8 Figure labels in the diagram: 100 - Sleep monitoring system; 200 - Host computer; 200a - First Bluetooth communication module; 1-Main unit; 2-Main control module; 3-EEG acquisition module; 4-Multi-axis motion sensor module; 11-Casing; 12-First circuit board; 13-First connector; 14-First magnetic clasp; 15-Battery; 21-Second Bluetooth communication module; 31-Soft adhesive patch; 32-Second circuit board; 33-Electrode; 34-Second connector; 35-Second magnetic clasp; 36-EEG acquisition chip; 37-ADC converter; 38-Signal amplifier; 41-Accelerometer; 42-Gyroscope. Detailed Implementation

[0026] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0027] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0028] Please see Figures 1-10This invention provides a sleep monitoring system based on multi-sensor fusion, belonging to the field of medical device technology. It can overcome the shortcomings of traditional sleep monitoring systems, such as high probability of false positives and false negatives, poor accuracy, and inability to meet usage requirements.

[0029] like Figure 1-10 As shown, the present invention provides a sleep monitoring system based on multi-sensor fusion according to one embodiment. The sleep monitoring system 100 based on multi-sensor fusion includes a host 1 and the following modules disposed on the host 1: Main control module 2; EEG acquisition module 3, connected to main control module 2, is used to acquire EEG signals; The multi-axis motion sensor module 4 is connected to the main control module 2 and is used to collect motion signals.

[0030] The multi-axis motion sensor module 4 includes at least an accelerometer 41 for acquiring acceleration and a gyroscope 42 for acquiring angular velocity.

[0031] The main control module 2 is responsible for controlling the EEG acquisition module 3 and the multi-axis motion sensor module 4 to acquire data, and can process the acquired data in real time.

[0032] Preferably, the main control module 2 uses the nRF52840 chip, and the multi-axis motion sensor module 4 only needs to be able to collect acceleration and angular velocity data, and can use a commercially available ICM20948 motion sensor or MPU6050 motion sensor. As a further preferred embodiment, this embodiment uses an ICM20948 nine-axis motion sensor (3-axis accelerometer + 3-axis gyroscope + 3-axis magnetometer).

[0033] The EEG acquisition module 3 can continuously acquire EEG data and transmit the acquired EEG data to the main control module 2 via the SPI interface. The accelerometer 41 can continuously acquire acceleration data and transmit the acquired acceleration data to the main control module 2 via the I2C interface. When the accelerometer 41 detects an abnormal acceleration change, the main control module 2 controls the gyroscope 42 to turn on.

[0034] Specifically, when the accelerometer 41 detects motion exceeding the second preset threshold, the main control module 2 controls the gyroscope 42 to turn on.

[0035] As a further preferred option, the second preset threshold is 0.3g, that is, the criterion for judging abnormal acceleration changes is that the acceleration change is greater than 0.3g (i.e. 0.3 times the gravitational acceleration).

[0036] Preferably, the main control module 2 can detect apnea events based on the acceleration data collected by the accelerometer 41.

[0037] Preferably, the main control module 2 first extracts the Y-axis data (perpendicular to the chest and abdomen direction) from the acceleration data collected by the multi-axis motion sensor module 4, and then performs bandpass filtering on the Y-axis data to extract the breathing signal. The amplitude of the breathing signal is monitored in real time. When the amplitude is continuously lower than the first set threshold for more than 10 seconds, it is determined to be a breathing apnea event, and the type and timestamp of the breathing apnea event are recorded.

[0038] As a further preferred option, the first set threshold can be 50 count units.

[0039] In addition, it should be noted that the first set threshold can be adaptively adjusted according to individual differences of users (such as weight and age), rather than just a fixed threshold.

[0040] Preferably, when the gyroscope 42 is turned on, the main control module 2 can detect the overturning event based on the angular velocity data collected by the gyroscope 42.

[0041] Preferably, the main control module 2 determines whether a rolling event has occurred based on the angular velocity data collected by the gyroscope 42. When a rolling event occurs, the main control module 2 suspends the determination of the breathing apnea event.

[0042] As a preferred method, the criteria for determining a rolling over event is that the angular velocity in the Z-axis direction (the Z-axis of the gyroscope is parallel to the human spine) is continuously greater than 45° / s and exceeds 500ms.

[0043] like Figure 3-7 As shown, the host 1 in this embodiment includes a housing 11, a first circuit board 12, and a first connector 13. The first circuit board 12 is installed inside the housing 11, and the first connector 13 is installed at the bottom of the housing 11 and electrically connected to the first circuit board 12. The EEG acquisition module 3 includes a soft adhesive patch 31, a second circuit board 32, an EEG acquisition chip 36, a second connector 34, and EEG electrodes 33. The soft adhesive patch 31 is attached to the user's forehead. The second circuit board 32 is installed inside the soft adhesive patch 31. The EEG electrodes 33 are installed at the bottom of the soft adhesive patch 31 and connected to the second circuit board 32. The second connector 34 is installed at the top of the soft adhesive patch 31 and electrically connected to the second circuit board 32. The second connector 34 cooperates with the first connector 13. The main control module 2, the EEG acquisition chip 36, and the multi-axis motion sensor module 4 are all integrated into the first circuit board 12. The first connector 13 is electrically connected to the EEG acquisition chip 36.

[0044] As a preferred option, the EEG acquisition chip 36 uses the ADS1292.

[0045] Preferably, the first connector 13 and the second connector 34 are PogoPin connectors.

[0046] Preferably, the bottom of the housing 11 is provided with a first magnetic absorbing member 14, and the soft adhesive patch 31 is provided with a second magnetic absorbing member 35 that cooperates with the first magnetic absorbing member 14. The soft adhesive patch 31, the second circuit board 32, the second connector 34 and the EEG electrodes 33 can together form a detachable fitting part relative to the main unit 1 for easy storage.

[0047] Preferably, the housing 11 in this embodiment has an overall arc-shaped structure. The housing 11 includes a detachable upper housing and a lower housing. A first receiving cavity is formed between the upper housing and the lower housing. The first circuit board 12 is installed in the first receiving cavity. A first positioning groove is opened in the lower housing. The upper end of the first connector 13 is soldered to the first circuit board 12, and the lower end extends into the first positioning groove.

[0048] As a further preferred embodiment, there are two first magnetic suction members 14, which are symmetrically arranged at both ends of the lower housing along its length.

[0049] Preferably, the main unit 1 also includes a battery 15, which is installed inside the housing 11 and electrically connected to the main control module 2 to provide power to various electrical components.

[0050] Preferably, in this embodiment, the first circuit board 12 is a PCB board, the second circuit board 32 is an FPC circuit board, and the soft adhesive patch 31 is a soft bonding material, usually made of silicone or TPU (thermoplastic polyurethane), which can conform to human skin and has good comfort. The soft adhesive patch 31 covers the FPC circuit board, which plays a role in transmitting electroencephalogram (EEG) signals.

[0051] Preferably, the EEG electrode 33 is a silver chloride dry electrode 33.

[0052] The specific formation method of the EEG acquisition module 3 in this embodiment is as follows: the first end of the electrode 33 and the first end of the second connector 34 are respectively welded to the two sides of the second circuit board 32. Then, the whole assembly and the second magnetic accumulator 35 are coated with silicone to form a soft adhesive patch 31. When the soft adhesive patch 31 is attracted to the first magnetic accumulator 14 of the host 1, the second end of the second connector 34 is attached to the first end of the first connector 13, and the corresponding contacts are in contact to conduct the circuit. The second end of the first connector 13 is welded to the first circuit board 12.

[0053] like Figure 8As shown, the sleep monitoring system 100 based on multi-sensor fusion in this embodiment can be used in conjunction with the host computer 200. The main control module 2 includes a first Bluetooth communication module 200a, and the host computer 200 includes a second Bluetooth communication module 21. The main control module 2 and the host computer 200 are wirelessly connected via Bluetooth. Thus, the main control module 2 is responsible for controlling the EEG acquisition module 3 to collect EEG data, and the multi-axis motion sensor module 4 to collect acceleration and angular velocity data. The main control module 2 then processes these data in real time and transmits the results to the host computer 200 via the Bluetooth module, monitoring the patient's sleep quality and whether sleep apnea occurs.

[0054] The present invention also provides a sleep monitoring method based on multi-sensor fusion according to an embodiment. This method is based on the sleep monitoring system 100 described above and is specifically implemented through the following steps: S1. Initialization and parameter settings: The main control module 2 initializes the EEG acquisition module 3 and the multi-axis motion sensor module 4, and configures the sampling rate and range; Among them, the EEG acquisition module 3 is configured with a sampling rate of 250Hz, and the accelerometer 41 of the multi-axis motion sensor module 4 is configured with a sampling rate of 20Hz and a range of ±2g. S2. Data Acquisition and Synchronization: EEG acquisition module 3 continuously acquires EEG data and transmits it to main control module 2; Accelerometer 41 continuously collects acceleration data and transmits it to main control module 2; Main control module 2 uses its internal RTC to add a precise timestamp to each data packet to ensure data synchronization; S3. Respiratory signal extraction: The main control module 2 extracts Y-axis data from the acceleration data collected by the multi-axis motion sensor module 4, and performs a 0.1-0.5Hz bandpass filter on the Y-axis data to extract the breathing signal; S4. Apnea event detection: The main control module 2 monitors the amplitude of the respiratory signal in real time. When the amplitude is continuously lower than the first set threshold (e.g., 50 count units) for more than 10 seconds, it is determined to be a breathing apnea event. Main control module 2 records the type and timestamp of the apnea event; S5, Turning Over Event Detection: When the accelerometer 41 detects a motion with an acceleration change exceeding a second set threshold (e.g., 0.3g), the gyroscope 42 is activated and configured with a sampling rate of 50Hz. The gyroscope 42 continuously collects angular velocity data and transmits it to the main control module 2; The main control module 2 uses the angular velocity data from the gyroscope 42 to determine whether a flipping event has occurred; When a rolling over event is detected, pause step S4 to avoid interference.

[0055] In this step, the criteria for determining a rolling over event are that the angular velocity in the Z-axis direction (the Z-axis of the gyroscope is parallel to the human spine) is continuously greater than 45° / s and exceeds 500ms.

[0056] In some implementations, the method further includes the following after step S5: S6. Comprehensive Judgment and Data Transmission: The main control module 2 can make a comprehensive judgment by combining EEG signals and rolling events. For example, when abnormal fluctuations occur in the EEG signals, it can combine the rolling events to confirm whether the fluctuations in the EEG signals are caused by the rolling movements. The main control module 2 can store and transmit the detected apnea events and turning events along with timestamps to the host computer 200.

[0057] As described above, the advantages of the multi-sensor fusion-based sleep monitoring system 100 and method of the present invention over the prior art include at least the following: 1. The present invention provides a novel sleep monitoring system 100. The system is equipped with a multi-axis motion sensor module 4, which can acquire acceleration data and angular velocity data. The main control module 2 can effectively combine angular velocity data and acceleration data to effectively distinguish between turning over and breathing movements, thereby reducing misjudgment.

[0058] 2. The sleep monitoring system 100 of the present invention is a lower-level machine. When in use, it is attached to the user's forehead. The detection algorithm for determining apnea events and turning over events can be realized through the main control module 2. The data acquired by the EEG signal acquisition module and the multi-axis motion sensor module 4 are processed in real time without relying on the upper-level machine 200, and the response is fast.

[0059] 3. By using a timestamp synchronization mechanism, the timing of EEG signals and body movement signals is aligned, thereby improving the accuracy of data fusion.

[0060] 4. The sleep monitoring system 100 of the present invention integrates multiple sensors such as an EEG collector, an accelerometer 41, and a gyroscope 42. Through the coordinated work of multiple sensors such as the EEG collector, the accelerometer 41, and the gyroscope 42, the accuracy and reliability of the results can be greatly improved.

[0061] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A sleep monitoring system based on multi-sensor fusion, characterized in that, Includes a host (1) and the following modules disposed on the host (1): Main control module (2); The EEG acquisition module (3) is connected to the main control module (2) and is used to acquire EEG signals; A multi-axis motion sensor module (4) is connected to the main control module (2) and is used to collect motion signals; The multi-axis motion sensor module (4) includes at least an accelerometer (41) for acquiring acceleration and a gyroscope (42) for acquiring angular velocity; The main control module (2) is responsible for controlling the EEG acquisition module (3) and the multi-axis motion sensor module (4) to acquire data, and can process the acquired data in real time.

2. The sleep monitoring system based on multi-sensor fusion according to claim 1, characterized in that, The EEG acquisition module (3) can continuously acquire EEG data and transmit the acquired EEG data to the main control module (2). The accelerometer (41) can continuously acquire acceleration data and transmit the acquired acceleration data to the main control module (2). When the accelerometer (41) detects an abnormal acceleration change, the main control module (2) controls the gyroscope (42) to turn on.

3. The sleep monitoring system based on multi-sensor fusion according to claim 2, characterized in that, The main control module (2) can detect apnea events based on the acceleration data collected by the accelerometer (41).

4. The sleep monitoring system based on multi-sensor fusion according to claim 3, characterized in that, The main control module (2) first extracts Y-axis direction data from the acceleration data collected by the multi-axis motion sensor module (4), then performs bandpass filtering on the Y-axis data to extract the breathing signal, monitors the amplitude of the breathing signal in real time, and determines the apnea event when the amplitude is lower than the first set threshold for more than 10 seconds, and records the type and timestamp of the apnea event.

5. The sleep monitoring system based on multi-sensor fusion according to claim 4, characterized in that, When the gyroscope (42) is turned on, the main control module (2) can detect the overturning event based on the angular velocity data collected by the gyroscope (42).

6. The sleep monitoring system based on multi-sensor fusion according to claim 5, characterized in that, The main control module (2) determines whether a rolling event has occurred based on the angular velocity data collected by the gyroscope (42). When a rolling event occurs, the main control module (2) suspends the determination of the breathing apnea event. The criteria for determining the rolling over event is that the angular velocity in the Z-axis direction is continuously greater than 45° / s and exceeds 500ms.

7. The sleep monitoring system based on multi-sensor fusion according to any one of claims 1-6, characterized in that, The host (1) includes a housing (11), a first circuit board (12), and a first connector (13). The first circuit board (12) is installed inside the housing (11), and the first connector (13) is installed at the bottom of the housing (11) and electrically connected to the first circuit board (12). The EEG acquisition module (3) includes a soft adhesive patch (31), a second circuit board (32), an EEG acquisition chip (36), a second connector (34), and EEG electrodes (33). The soft adhesive patch (31) is attached to the user's forehead, and the second circuit board (32) is installed on the soft adhesive patch. Inside the adhesive patch (31), the EEG electrode (33) is installed at the bottom of the adhesive patch (31) and connected to the second circuit board (32). The second connector (34) is installed at the top of the adhesive patch (31) and electrically connected to the second circuit board (32). The second connector (34) cooperates with the first connector (13). The main control module (2), the EEG acquisition chip (36) and the multi-axis motion sensor module (4) are all integrated on the first circuit board (12). The first connector (13) is electrically connected to the EEG acquisition chip (36).

8. The sleep monitoring system based on multi-sensor fusion according to any one of claims 1-6, characterized in that, It can be used with a host computer (200). The main control module (2) includes a first Bluetooth communication module (200a), and the host computer (200) includes a second Bluetooth communication module (21). The main control module (2) and the host computer (200) are wirelessly connected via Bluetooth.

9. A sleep monitoring method based on multi-sensor fusion, characterized in that, Based on the sleep monitoring system (100) based on multi-sensor fusion as described in any one of claims 1-8, the sleep monitoring method includes at least the following steps: S1. Initialization and parameter settings: The main control module (2) initializes the EEG acquisition module (3) and the multi-axis motion sensor module (4), and configures the sampling rate and range; Among them, the EEG acquisition module (3) is configured with a sampling rate of 250Hz, and the accelerometer (41) of the multi-axis motion sensor module (4) is configured with a sampling rate of 20Hz and a range of ±2g. S2. Data Acquisition and Synchronization: The EEG acquisition module (3) continuously acquires EEG data and transmits it to the main control module (2); The accelerometer (41) continuously collects acceleration data and transmits it to the main control module (2); The main control module (2) uses its internal RTC to add a precise timestamp to each data packet to ensure data synchronization; S3. Respiratory signal extraction: The main control module (2) extracts Y-axis direction data from the acceleration data collected by the multi-axis motion sensor module (4), and performs 0.1-0.5Hz bandpass filtering on the Y-axis direction data to extract the breathing signal; S4. Apnea event detection: The main control module (2) monitors the amplitude of the respiratory signal in real time. When the amplitude is continuously lower than the first set threshold and exceeds 10 seconds, it is determined to be a respiratory arrest event. The main control module (2) records the type and timestamp of the apnea event; S5, Turning Over Event Detection: When the accelerometer (41) detects motion exceeding the second set threshold, the gyroscope (42) is activated, and the gyroscope (42) is configured with a sampling rate of 50Hz. The gyroscope (42) continuously collects angular velocity data and transmits it to the main control module (2); The main control module (2) determines whether a rollover event has occurred by using the angular velocity data of the gyroscope (42); Pause step S4 when a rolling over event is detected.

10. The sleep monitoring method based on multi-sensor fusion according to claim 9, characterized in that, The process after step S5 also includes: S6. Comprehensive Judgment and Data Transmission: The main control module (2) can make a comprehensive judgment by combining EEG signals and rolling events; The main control module (2) can store and transmit the detected apnea events and turning events along with timestamps.

Images