Fusion multi-mode biological signal detection method and fusion multi-mode biological signal detection system

Abstract

The invention relates to the technical field of biological signal detection, and particularly discloses a fusion multi-mode biological signal detection method and system. The detection system comprises an electroencephalogram signal collector, an electrocardiosignal collector, an electromyographic signal collector, a blood oxygen signal collector, a preprocessor, a wireless transmitting device, a wireless receiving device, an integrated processor and an upper computer. The electroencephalogram signal collector, the electrocardiosignal collector, the electromyographic signal collector, the blood oxygen signal collector and the wireless transmitting device are respectively connected to the preprocessor; the wireless receiving device is connected to the integrated processor, and the integrated processor is connected to the upper computer. The collectors can synchronously collect electroencephalogram, electrocardio, myoelectricity and blood oxygen signals, the system is portable and convenient to wear in a wireless transmission mode, the signals can be fused and displayed according to time, an analysis report is output, and synchronous detection and concentrated analysis of various biological signals are achieved.

Description

Technical Field

[0001] This application relates to the field of biosignal detection technology, specifically to a method and system for detecting multimodal biosignals. Background Technology

[0002] Brain-computer interface (BCI) technology is a technique that allows the human brain to interact with a computer system. It studies brain activity by capturing and analyzing electrical signals generated by the brain. The basic principle of this technology is to use electrodes to record changes in electrical potential on the scalp; these changes reflect the synchronized activity of groups of neurons in the brain. By placing multiple electrodes at specific locations on the head, electrical activity in different areas of the brain can be captured and recorded to form an electroencephalogram (EEG). This also enables the direct control of computer systems and other external devices via the human brain. BCI technology has been widely applied in fields such as medicine, gaming, assistive communication, and psychology.

[0003] Electrocardiogram (ECG) monitoring is an important tool for assessing cardiac function by recording and analyzing the heart's electrical activity. It is a non-invasive diagnostic technique that records an ECG by capturing the electrical signals generated by the heart with each beat. These signals reflect the depolarization and repolarization processes of the heart muscle, providing crucial information about heart health. The basic principle of ECG monitoring involves using electrode patches or conductive gels to detect subtle electrical changes on the skin surface, amplifying and recording these changes to create a visualized waveform, providing strong support for the prevention and treatment of cardiovascular diseases.

[0004] Electromyography (EMG) is a technique that analyzes muscle function and status by detecting and recording electrical activity in muscles. This technology is based on the phenomenon that muscles generate electrical signals during contraction. When muscle fibers are excited, they produce detectable changes in electrical potential. These changes can be captured and recorded by electrodes to form an electromyogram (EMG). EMG is widely used in clinical diagnostics, sports science, rehabilitation medicine, and bionics. In clinical diagnostics, EMG is used to assess neuromuscular diseases such as neuropathy and muscle atrophy, helping doctors determine the location and extent of the lesions. In sports science, EMG is used to analyze athletes' movement techniques and muscle usage, guiding training, improving athletic performance, and preventing sports injuries. In rehabilitation medicine, EMG is used to assess patients' rehabilitation progress, guide rehabilitation training, and promote the recovery of muscle function. Furthermore, EMG is also applied in bionics, such as prosthetic control, using the detection of EMG signals to control the movement of prostheses, helping people with disabilities regain their daily living abilities.

[0005] Finger oximetry is a non-invasive technology that measures blood oxygen saturation levels using a photoelectric sensor at the fingertip. Based on the principle of pulse oximetry, it estimates blood oxygen saturation (SpO2) by utilizing the different absorption characteristics of hemoglobin for red and near-infrared light. The sensor emits these two types of light, and the blood oxygen level is calculated by detecting the change in light intensity after absorption by the blood. In terms of applications, finger oximetry is widely used in hospitals, home care, sports and health monitoring, and high-altitude environments. In hospitals, it is a routine device for monitoring patients during and after surgery and in intensive care units, used to monitor patients' blood oxygen saturation in real time and promptly detect problems in the respiratory and circulatory systems. In home care, small pulse oximeters allow patients to self-monitor at home, especially suitable for patients with chronic respiratory diseases. In the sports and health field, athletes and outdoor enthusiasts use pulse oximeters to assess physical condition and adapt to altitude changes. Those working at high altitudes and mountaineers can also monitor blood oxygen levels to prevent altitude sickness.

[0006] Currently available biosignal detection devices such as EEG, ECG, EMG, and blood oxygen saturation are generally used for measuring specific types of signals, such as EEG or ECG. This specialization prevents the simultaneous measurement of multiple biosignals. For example, a device may only allow ECG or EMG measurements at a time, not both EEG and ECG simultaneously. This limits the efficiency and depth of multidimensional biological data analysis, making the comprehensive assessment of an individual's physiological state complex. Because multiple biosignals cannot be detected simultaneously, users may need to repeat measurements at different times using different devices, significantly increasing the total testing time and workload. Prolonged testing not only affects user experience but may also introduce more errors and variability. Especially during long-term monitoring, the natural fluctuations in biosignals can lead to data inconsistencies and difficulties in interpretation.

[0007] In conclusion, although biosignal detection technology provides valuable information for medical diagnosis and treatment, the limitations of related technologies have hindered its wider application and development. Summary of the Invention

[0008] To achieve simultaneous detection and centralized analysis of multiple biological signals and provide a more comprehensive and accurate assessment of physiological state, this application designs an integrated biological signal detection system, the key components of which and its workflow are as follows.

[0009] Firstly, this application proposes a multimodal biosignal detection system and adopts the following technical solution.

[0010] A multimodal biosignal detection system includes an electroencephalogram (EEG) signal acquisition unit, an electrocardiogram (ECG) signal acquisition unit, an electromyogram (EMG) signal acquisition unit, a blood oxygen saturation (POS) signal acquisition unit, a preprocessor, a wireless transmitter, a wireless receiver, an integrated processor, and a host computer. The EEG, ECG, EMG, POS, and wireless transmitter are each connected to the preprocessor; the wireless receiver is connected to the integrated processor, and the integrated processor is connected to the host computer.

[0011] By adopting the above technical solutions, multiple acquisition devices can simultaneously acquire EEG, ECG, EMG, and blood oxygenation signals. The preprocessor performs preprocessing such as signal enhancement and time sorting. This system adopts a wireless transmission mode, making the system portable and less restricted by wearable devices. The integrated processor performs data optimization and feature extraction, and the host computer fuses and displays various signals and outputs analysis reports, realizing the synchronous detection and centralized analysis of multiple biological signals.

[0012] Secondly, this application proposes a method for detecting multimodal biological signals, and adopts the following technical solution.

[0013] A multimodal biosignal detection method is implemented using the aforementioned detection system. The preprocessor simultaneously receives raw signals from the EEG, ECG, EMG, and POS acquisition devices, amplifies, filters, and performs analog-to-digital conversion on each raw signal, assigns identifiers and timestamps, then packages the data and transmits it via the wireless transmitter to the wireless receiver. The wireless receiver then transmits the data to the integrated processor. The integrated processor uses Fast Fourier Transform and / or Wavelet Transform algorithms to perform real-time data analysis, package the data, and transmit it to the host computer.

[0014] By employing the above-mentioned technical solution, this detection method simultaneously collects and processes multiple biological signals, including electroencephalogram (EEG), electrocardiogram (ECG), electromyogram (EMG), and blood oxygenation. This integrated approach to simultaneously detecting multiple biological signals significantly improves monitoring efficiency and response speed. Users can obtain multi-dimensional physiological information at the same time, enabling doctors to more comprehensively assess the patient's condition.

[0015] A preferred embodiment of the above-mentioned multimodal biosignal detection method is that the host computer includes a display; the host computer integrates and displays electroencephalogram (EEG) signals, electrocardiogram (ECG) signals, electromyogram (EMG) signals, and blood oxygenation signals in a view of the display, and these signals are aligned on the same time axis.

[0016] By adopting the above technical solution, information extracted from different signal sources can be accurately correlated in time, providing a foundation for comprehensive physiological state analysis.

[0017] A preferred embodiment of the above-mentioned multimodal biosignal detection method is as follows: the EEG signal acquisition device includes a reference electrode, a right leg drive electrode, and multiple acquisition electrodes; the reference electrode is placed at the mastoid process of the subject, and the multiple acquisition electrodes are placed in the non-mastoid region of the subject's scalp; the right leg drive electrode is placed at one position on one of the subject's limbs. The ECG signal acquisition device includes multiple ECG sensors attached to the subject's chest and limbs; the EMG signal acquisition device includes multiple EMG sensors attached to the skin outside the muscles; and the POS signal acquisition device is worn on the subject's finger or earlobe.

[0018] By employing the above technical solutions, multiple acquisition electrodes directly contact the scalp to capture weak electrical signals generated by the activity of neurons in the cerebral cortex; a reference electrode provides a unified voltage reference for all acquisition electrodes; the electroencephalogram (EEG) signal is essentially the potential difference between the acquisition electrode and the reference electrode. The right leg drive electrode actively eliminates common-mode interference by injecting environmental noise back into the body through a negative feedback circuit, thus achieving noise cancellation. The electrocardiogram (ECG) signal acquisition device is designed to monitor the electrical activity of the heart. Multiple ECG sensors are attached to the subject's skin, typically located on the chest and limbs, to capture signals from different leads of the ECG. The electromyography (EMG) signal acquisition device focuses on detecting and recording electrical signals from muscle activity. EMG sensors are directly attached to the skin surface, typically covering major muscle groups or specific muscles, and can be used to analyze muscle fatigue and movement patterns. The blood oxygen signal acquisition device combines blood oxygen saturation monitoring functionality, primarily used for real-time monitoring of blood oxygen levels and heart rate changes. It is worn on the subject's finger or earlobe, emitting infrared and red light, and measuring the absorption rate of these two types of light by blood vessels to calculate blood oxygen saturation.

[0019] A preferred embodiment of the above-mentioned multimodal biosignal detection method is that the detection method further includes correlation analysis between biosignals.

[0020] By employing the above technical solutions, the system can not only analyze single biological signals but also assess the interactions and correlations between different physiological signals. This cross-signal analysis method helps to reveal the complex interactions between various systems in the human body, providing a completely new perspective for a deeper understanding of human physiological mechanisms.

[0021] A preferred embodiment of the above-mentioned multimodal biosignal detection method is that the correlation analysis between the biosignals includes analyzing the relationship between increased heart rate and enhanced brain wave activity, and analyzing the correlation between respiratory rate and blood oxygen saturation.

[0022] By employing the above technical solutions and performing multi-signal correlation analysis, it is possible to accurately analyze health problems.

[0023] A preferred embodiment of the above-mentioned multimodal biosignal detection method is that the analysis of the relationship between increased heart rate and enhanced brainwave activity specifically includes: activating a mechanism that can induce an increase in the subject's heart rate, and recording brainwave and electrocardiogram signals.

[0024] For EEG signals acquired by multiple acquisition electrodes, multiple consecutive equal time intervals t are divided with time as the axis. The average power of the EEG signal in each equal time interval is calculated to obtain the EEG wave sequence corresponding to the average power in the equal time interval.

[0025] For multiple sets of ECG signals collected by multiple ECG sensors, the time is used as the axis to divide the signal into multiple consecutive equal time intervals t. The average heart rate of the ECG signal in each equal time interval is calculated to obtain the heart rate sequence corresponding to the average heart rate in the equal time interval.

[0026] Align the EEG and heart rate sequences over time and calculate the Pearson correlation coefficient r over the entire time period. r is equal to the quotient of the covariance and standard deviation of the average power and average heart rate. If r is between 0.8 and 1, it indicates that an increase in heart rate is accompanied by an increase in EEG power.

[0027] By employing the above-mentioned technical approach, the relationship between increased heart rate and enhanced brainwave activity can be confirmed.

[0028] A preferred embodiment of the above-mentioned multimodal biosignal detection method is that the detection method includes using machine learning and pattern recognition techniques to perform in-depth learning on historical data in order to identify common health problems and potential risk patterns.

[0029] By adopting the above technical solution, this method can improve the accuracy and timeliness of disease diagnosis and treatment.

[0030] In summary, the fusion multimodal biosignal detection method and system of this application have the following beneficial effects: the detection system can simultaneously collect and process multiple biosignals such as EEG, ECG, EMG, and blood oxygenation, improving monitoring efficiency. Users can obtain multidimensional physiological information at the same time, enabling doctors to more comprehensively assess the patient's condition. Attached Figure Description

[0031] Figure 1 This is a schematic diagram of the structure of a multimodal biosignal detection system.

[0032] Figure 2 A schematic diagram illustrating the fusion of multiple biosignal data. Detailed Implementation

[0033] The technical solutions in the embodiments will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the following embodiments, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0034] refer to Figure 1 A multimodal biosignal detection system includes an electroencephalogram (EEG) signal acquisition unit, an electrocardiogram (ECG) signal acquisition unit, an electromyogram (EMG) signal acquisition unit, a blood oxygen saturation (POS) signal acquisition unit, a preprocessor, a wireless transmitter, a wireless receiver, an integrated processor, and a host computer. The EEG, ECG, EMG, POS, and POS signal acquisition units and the wireless transmitter are each connected to the preprocessor; the wireless receiver is connected to the integrated processor, and the integrated processor is connected to the host computer. The preprocessor can be a microcontroller unit (MCU). Both the wireless transmitter and receiver can be Bluetooth devices. The host computer can be a computer.

[0035] The overall hardware design of the detection system is as follows: The EEG signal acquisition unit acquires human brain signals using eight acquisition electrodes, one reference electrode, and one right leg drive electrode. The reference electrode is placed on the mastoid process of the subject to provide a relatively constant voltage reference point. The eight acquisition electrodes are placed on the non-mastoid locations of the subject's head, covering different brain regions for comprehensive monitoring of brain activity. The right leg drive electrode is placed on one of the limbs away from the head to reduce interference from other parts of the body. The main function of the EEG signal acquisition unit is to capture and record the electrical signals generated by brain activity. All electrodes are connected to a preprocessor. The preprocessor communicates wirelessly with an integrated processor for synchronized data transmission and efficient processing.

[0036] An electrocardiogram (ECG) acquisition device is designed to monitor the electrical activity of the heart. It contains multiple ECG sensors that are attached to the subject's skin, typically located on the chest and limbs, to capture signals from different leads of the ECG. The device features automatic gain control and baseline stabilization to ensure clear ECG signals are obtained even when the subject is exercising.

[0037] Electromyography (EMG) signal acquisition devices focus on detecting and recording electrical signals related to muscle activity. The device consists of multiple miniature EMG sensors that are attached directly to the skin surface, typically covering major muscle groups or specific muscles. Wireless transmission technology facilitates data transmission quickly and easily, allowing for long-term dynamic monitoring to analyze muscle fatigue and movement patterns.

[0038] A pulse oximeter combines blood oxygen saturation monitoring with real-time monitoring of blood oxygen levels and heart rate changes. This device uses a non-invasive sensor, typically worn on the finger or earlobe of the subject. It emits infrared and red light and measures the absorption rate of these two types of light by the blood vessels to calculate blood oxygen saturation. After a period of measurement, the pulse oximeter also provides the velocity and morphology of the pulse wave, which can be used to further calculate parameters such as heart rate.

[0039] The preprocessor simultaneously receives raw signals from different biosignal acquisition devices, including EEG signals acquired by an EEG signal acquisition device, ECG signals acquired by an ECG signal acquisition device, ECG signals acquired by an ECG signal acquisition device, and blood oxygenation signals acquired by an EMG signal acquisition device.

[0040] The preprocessor processes, amplifies, filters, and converts the raw signals received from different biosignal acquisition devices. It enhances weak bioelectrical signals using a series of high-performance amplifiers, removes noise and artifacts using digital filtering techniques, and optimizes signal quality. Furthermore, the preprocessor performs analog-to-digital conversion, transforming analog signals into digital signals for subsequent digital processing and analysis. To ensure data synchronization and integrity, the preprocessed data undergoes time and sequence alignment in the packetization module. The preprocessor employs precise clock synchronization technology to ensure that signals from different acquisition devices are consistent in time, which is crucial for subsequent data fusion and comprehensive analysis. During packetization, each signal is assigned a specific identifier and timestamp, and then packaged according to a specific format for accurate parsing and storage by the integrated processor. The preprocessed and packetized data is then transmitted to the integrated processor via a Bluetooth transmitter. This Bluetooth transmitter is designed as a low-power, high-efficiency wireless transmission module, supporting the latest Bluetooth technology standards to ensure stable and secure data transmission. It employs encryption technology to protect data from unauthorized interception or tampering during transmission. In addition, the device features intelligent connectivity, enabling it to automatically locate and connect to the integrated processor, simplifying user operation and enhancing the ease of use of the system.

[0041] The Bluetooth receiver module of this application adopts high-performance Bluetooth receiving technology, which can stably receive data streams transmitted simultaneously from multiple collectors, ensuring high data transmission efficiency and low latency. Furthermore, the module supports automatic reconnection, enabling rapid restoration of the connection even in cases of weak signal or temporary interruption, ensuring data continuity and integrity.

[0042] After the signal is received via Bluetooth, it is sent to the integrated processor for real-time analysis. This part mainly performs data (digital) frame processing, including signal identification, fitting, and format conversion, to optimize data quality and meet the needs of subsequent processing. Subsequently, the integrated processor performs feature extraction and data compression to extract key information reflecting the user's health status. These processing steps employ advanced algorithms and technologies, such as Fast Fourier Transform and Wavelet Transform, to ensure the accuracy and reliability of the processing results. The processed data is packaged according to a specific communication protocol to form standardized data packets, ensuring the integrity and consistency of the data during transmission and facilitating further processing and parsing by the PC host computer. The data packets are sent to the PC host computer via a serial port (such as USB or UART).

[0043] The PC-based host computer has a display screen for showing EEG, ECG, EMG, and blood oxygenation signals.

[0044] refer to Figure 2 This diagram illustrates the fusion of multiple biosignal data. It displays the waveforms of electroencephalogram (EEG), electrocardiogram (ECG), electromyogram (EMG), and blood oxygenation signals. These signals are integrated into a unified view, visually representing the relationships and differences between different biosignals. This integrated display not only facilitates rapid assessment of patient conditions by medical professionals but also makes complex physiological information easier for non-professionals to understand.

[0045] To achieve effective data fusion, this application employs a unified time base for biosignal synchronization. All signal acquisition devices were designed with time consistency in mind, ensuring that signals acquired from various sources are aligned on the same timeline. This is crucial for subsequent data analysis, as it guarantees that information extracted from different signal sources can be accurately correlated, providing a foundation for comprehensive physiological state analysis.

[0046] Building upon data fusion, this application further enables the automatic detection of abnormal fluctuations. The system is equipped with advanced algorithms such as Fast Fourier Transform and Wavelet Transform, which can analyze each signal waveform individually and identify fluctuation patterns that deviate from the predetermined normal range. These abnormal fluctuations may be warning signals of poor health or early-stage diseases; therefore, timely detection and reporting of these abnormalities are of paramount importance for disease prevention and early treatment.

[0047] This application also provides a correlation analysis function between biological signals. This function enables the system not only to analyze single biological signals, but also to assess the interactions and correlations between different physiological signals. For example, the system can demonstrate the relationship between increased heart rate and enhanced brain wave activity, or analyze the correlation between respiratory rate and blood oxygen saturation. This cross-signal analysis method helps to reveal the complex interaction mechanisms between various systems in the human body, providing a new perspective for a deeper understanding of human physiological mechanisms.

[0048] The study analyzed the relationship between increased heart rate and enhanced brainwave activity, specifically by activating mechanisms that could induce an increase in the subject's heart rate, such as fright stimuli, and recording brainwave and electrocardiogram signals.

[0049] For EEG signals acquired by multiple acquisition electrodes, multiple consecutive equal time intervals t are divided with time as the axis. The average power of the EEG signal in each equal time interval is calculated to obtain the EEG wave sequence corresponding to the average power in the equal time interval.

[0050] For multiple sets of ECG signals collected by multiple ECG sensors, the time is used as the axis to divide the signal into multiple consecutive equal time intervals t. The average heart rate of the ECG signal in each equal time interval is calculated to obtain the heart rate sequence corresponding to the average heart rate in the equal time interval.

[0051] The two occurrences of 't' above indicate that the time interval between the EEG sequence and the heart rate sequence is equal. 't' can be any value between 2 and 10 seconds, but is not limited to this.

[0052] The EEG and heart rate sequences were aligned over time, and the Pearson correlation coefficient *r* was calculated for the entire time period. *r* is equal to the quotient of the covariance and standard deviation of the average power and average heart rate. If *r* is between 0.8 and 1, it indicates that an increase in heart rate is accompanied by an increase in EEG power. If *r* is close to -1, it indicates that an increase in heart rate is accompanied by a decrease in EEG power. If *r* is close to 0, it indicates that there is no linear correlation between heart rate and EEG power.

[0053] With the accumulation of massive amounts of data, this application further utilizes machine learning and pattern recognition technologies to perform in-depth learning on historical data to identify common health problems and potential risk patterns. This predictive analytics can help doctors intervene in the early stages of disease development, thereby preventing more serious health problems from occurring. Simultaneously, this also provides data support for personalized medicine, enabling the development of more individualized health management plans based on individual physiological characteristics and health history.

[0054] This application provides a comprehensive and efficient solution for medical monitoring and health management through highly integrated data display, synchronized time base points, intelligent anomaly detection, in-depth correlation analysis, and advanced predictive technology. This not only improves the efficiency of medical data utilization but also greatly enhances the accuracy and timeliness of disease diagnosis and treatment.

[0055] In the field of medical and health monitoring, the demand for accurate and simultaneous detection of multiple biological signals is increasing. However, existing equipment often only detects a single biological signal, such as electroencephalogram (EEG) or electrocardiogram (ECG), and is mostly wired, limiting the patient's range of motion and affecting the real-time nature and accuracy of the data. Data processing and analysis often require multiple steps, which are time-consuming and inefficient. Most importantly, correlation analysis is often independent and lacks correlation. This application's integrated multimodal biological signal detection system, by collecting and processing multiple biological signals such as EEG, ECG, EMG, and blood oxygenation, achieves the organic combination of multiple biological signals, resulting in the following advantages.

[0056] (1) Improve monitoring efficiency: By integrating multiple biological signals simultaneously, the monitoring efficiency and response speed are greatly improved. Users can obtain multi-dimensional physiological information at the same time, and doctors can more comprehensively assess the patient's condition.

[0057] (2) Enhanced data accuracy: Due to the adoption of high-quality wireless transmission technology, the noise and interference that may be introduced by wired connections are reduced, thereby improving the accuracy and reliability of the data.

[0058] (3) Improved portability and comfort: The miniaturized detection system and wireless design greatly enhance the system's portability and user comfort. Patients can perform daily activities with fewer restrictions while undergoing health monitoring, making it particularly suitable for long-term or home monitoring applications.

[0059] (4) User-friendly interface: It provides an intuitive user interface, supports multiple languages, and makes it easy for different user groups to understand and operate. In addition, automatically generated reports and analysis results simplify communication between doctors and patients.

[0060] (5) Reduced operation and maintenance costs: The maintenance and upgrades of this system can be completed remotely, reducing the need for on-site maintenance and related costs. This centralized management approach also makes system updates faster and more efficient.

[0061] (6) Promoting the development of personalized medicine: The large amount of high-quality data accumulated provides a reliable data foundation for personalized medicine. Doctors can formulate more personalized treatment plans and interventions based on the specific circumstances of each patient.

[0062] (7) Enhance patients' self-management ability: Through continuous and real-time health monitoring, patients can better understand their own health status and thus take proactive management measures. This sense of participation and control can significantly improve patient satisfaction and treatment adherence.

[0063] (8) Data security and privacy protection: The system design includes advanced data encryption and user authentication mechanisms to ensure the security and privacy of patient data.

[0064] (9) Strong environmental adaptability: This system is portable and wireless, and is suitable for applications in hospitals, clinics, homes and mobile scenarios, such as health monitoring on the road or at work.

[0065] (10) It can serve as a powerful tool for scientific research and clinical trials: Researchers and clinicians can use the integrated multi-parameter monitoring capabilities of this system to conduct a wide range of research, from drug development to cognitive science research, thereby promoting the development of the medical frontier.

[0066] The integrated multimodal biosignal detection method and system presented in this application improves the quality and efficiency of medical monitoring, has a wide range of applications in clinical and family health management, and possesses significant social and economic value. Furthermore, the widespread use of this integrated sensing and analysis technology will greatly promote innovation and progress in the health field, enabling more intelligent and humanized medical services.

[0067] Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A fusion multi-modal bio-signal detection system, characterized by, The device comprises an electroencephalogram signal collector, an electrocardiogram signal collector, an electromyogram signal collector, an oxygen saturation signal collector, a preprocessor, a wireless sending device, a wireless receiving device, an integrated processor and a host computer. The electroencephalogram signal collector, the electrocardiogram signal collector, the electromyogram signal collector, the oxygen saturation signal collector and the wireless sending device are connected to the preprocessor; the wireless receiving device is connected to the integrated processor, and the integrated processor is connected to the host computer.

2. A fusion multi-modal bio-signal detection method, characterized by, The preprocessor simultaneously receives the original signals collected by the electroencephalogram signal collector, the electrocardiogram signal collector, the electromyogram signal collector and the oxygen saturation signal collector, amplifies, filters and digitizes each original signal, assigns an identifier and a time stamp to each original signal, packs the data, and sends the data to the wireless sending device; the wireless receiving device sends the data to the integrated processor; the integrated processor performs real-time analysis on the data by using a fast Fourier transform and / or a wavelet transform algorithm, and sends the data to the host computer. 3.The fusion multi-modal bio-signal detection method of claim 2, wherein, The host computer comprises a display; the host computer displays the electroencephalogram signal, the electrocardiogram signal, the electromyogram signal and the oxygen saturation signal in one view of the display, and the signals are aligned on the same time axis. 4.The fusion multi-modal bio-signal detection method of claim 2, wherein, The electroencephalogram signal collector comprises a reference electrode, a right leg driving electrode and a plurality of collecting electrodes; the reference electrode is arranged at the mastoid of a subject, and the plurality of collecting electrodes are arranged at the scalp of the subject; the right leg driving electrode is arranged at a position of a limb of the subject. The electrocardiogram signal collector comprises a plurality of electrocardiogram sensors arranged on the chest and limbs of the subject; the electromyogram signal collector comprises a plurality of electromyogram sensors arranged on the skin outside the muscles; and the oxygen saturation signal collector is arranged at a finger or an earlobe of the subject. 5.The fusion multi-modal bio-signal detection method of claim 4, wherein, The detection method further comprises correlation analysis between the biological signals. 6.The fusion multi-modal bio-signal detection method of claim 5, wherein, The correlation analysis between the biological signals comprises analyzing the relationship between an increase in heart rate and an increase in electroencephalogram activity, and analyzing the correlation between the respiratory rate and the oxygen saturation. 7.The fusion multi-modal bio-signal detection method of claim 6, wherein, The analysis of the relationship between the increase in heart rate and the increase in electroencephalogram activity specifically comprises: starting a mechanism capable of inducing an increase in heart rate of the subject, and recording the electroencephalogram signal and the electrocardiogram signal; for the electroencephalogram signals collected by the plurality of collecting electrodes, dividing a plurality of continuous equal time intervals t based on time, calculating the average power of the electroencephalogram signals in each equal time interval, and obtaining an electroencephalogram sequence corresponding to the equal time intervals; for the plurality of groups of electrocardiogram signals collected by the plurality of electrocardiogram sensors, dividing a plurality of continuous equal time intervals t based on time, calculating the average heart rate of the electrocardiogram signals in each equal time interval, and obtaining a heart rate sequence corresponding to the equal time intervals; aligning the electroencephalogram sequence and the heart rate sequence based on time, calculating a Pearson correlation coefficient r in the overall time period, and determining that the increase in heart rate is accompanied by the increase in electroencephalogram power when the Pearson correlation coefficient r is 0.8-1.

8. The fusion multimodal bio-signal detection method according to any one of claims 2-7, characterized in that, The detection method includes in-depth learning of historical data using machine learning and pattern recognition techniques to identify common health issues and potential risk patterns.

Images