Information acquisition method of brain-computer interface device

Abstract

The invention provides an information acquisition method of a brain-computer interface device. The information acquisition method comprises the following steps: emitting multi-wavelength pulsed light to a target user wearing the head-mounted device by using a light source in the head-mounted device; processing the optical signal of the multi-wavelength pulse light through a time domain delay module and a wavelength correlation light splitting module to generate a sparse pulse light array; transmitting the sparse pulse light array to a sensing front end by using a sensing network, and sending the sparse pulse light array to a brain area of the target user through a sensing point of the sensing front end; acquiring an optical signal reflected by the brain area by utilizing a detector, and performing real-time processing through an FPGA (Field Programmable Gate Array) to obtain processing information; and driving an external device by using the processing information to complete identification and execution of the intention of the target user. The FPGA is adopted for signal processing, complex operation can be completed within extremely short time, real-time response is supported, and the method is suitable for application scenes with quick response.

Description

Technical Field

[0001] The present invention relates to the field of brain-computer interface, and in particular to an information acquisition method for a brain-computer interface device. Background Art

[0002] In recent years, brain-computer interface (BCI) has gradually become an important enabling technology for the future intelligent world. In recent years, brain-computer interface has gained wide attention in the fields of intelligent device control and virtual reality. Traditional BCI devices face problems such as signal attenuation, wavelength interference, bulky equipment and safety.

[0003] A brain-computer interface device and information acquisition method are disclosed in the prior art, which relates to the field of the Internet of Things. The brain-computer interface includes a light source, a time domain delay module, a wavelength-dependent spectroscopic module, a sensor network and a sensor front end; wherein the light source is used to provide a wide-spectrum pulse light train, the time domain delay module is used to convert the wide-spectrum pulse light train into a multi-wave pulse light train, the wavelength-dependent spectroscopic module is used to decompose the multi-wave pulse light train into a sparse pulse light train, the sensor network is used to send the sparse pulse light train to the sensing point of the sensor front end, and the sensor front end is used to send the sparse pulse light train to the target brain through the sensing point to obtain sensing information, thereby obtaining detection light of different wavelengths by using wide-spectrum pulse light and time domain delay technology, without setting up multiple light sources, and without adding corresponding wavelength control modules, thereby achieving high spatial resolution of the brain-computer interface on the basis of ensuring the small integration of the brain-computer interface. However, the use of the time domain delay module in the above scheme will bring about a delay problem, and the time domain delay causes the lag of the signal to affect the real-time performance, especially in applications involving rapid response. Summary of the invention

[0004] In order to overcome the deficiencies of the prior art, an object of the present invention is to provide an information acquisition method for a brain-computer interface device.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A method for acquiring information of a brain-computer interface device, comprising:

[0007] Using a light source in the head-mounted device, emitting multi-wavelength pulsed light to a target user wearing the head-mounted device;

[0008] Processing the optical signal of the multi-wavelength pulse light by a time domain delay module and a wavelength-dependent spectroscopic module to generate a sparse pulse light train;

[0009] The sparse pulse light train is transmitted to a sensing front end by using a sensing network, and is sent to a brain area of ​​the target user through a sensing point of the sensing front end;

[0010] Acquire the light signal reflected by the brain region by using a detector, and process it in real time by using an FPGA to obtain processing information;

[0011] The processing information is used to drive an external device to complete the recognition and execution of the target user's intention.

[0012] Preferably, the multi-wavelength pulse light includes near-infrared light, visible light and ultraviolet light.

[0013] Preferably, the optical signal of the multi-wavelength pulse light is processed by a time domain delay module and a wavelength-dependent spectroscopic module to generate a sparse pulse light train, including:

[0014] Processing the input optical signal of the multi-wavelength pulse light through a time domain delay module to generate a plurality of multi-wavelength pulse light trains;

[0015] Each multi-wave pulse light train is decomposed by a wavelength-dependent spectroscopic module to obtain multiple sparse pulse light trains.

[0016] Preferably, the sensor network includes a sensor server and multiple sensor nodes; the sensor server periodically obtains communication status information of each sensor node in the topological network structure and the sensor nodes within a preset distance range around it, and inputs the obtained communication status information into a pre-established node connection decision model to obtain a real-time decision result, and based on the real-time decision result, dynamically adjusts the connection relationship between each sensor node in the topological network structure and the sensor nodes within the preset distance range around it in real time to form an adaptive topological network structure; the real-time decision result includes sensor nodes that need to be connected and disconnected in the topological network structure; the sensor server transmits data with each sensor node based on the adaptive topological network structure.

[0017] Preferably, the communication status information includes at least one of signal strength, data transmission rate, packet loss rate, delay, energy consumption, moving speed and historical connection quality.

[0018] Preferably, the method for constructing the node connection decision model includes:

[0019] The sample data included in the training set is input into the machine learning model for iterative training until the loss function of the machine learning model meets the requirements, and the trained machine learning model is obtained as the node connection decision model; the sample data includes historical communication status information of each sensor node and the sensor nodes within a preset distance range around it in a known topological network structure, and the historical connection relationship between each sensor node and the sensor nodes within a preset distance range around it under different historical communication states.

[0020] Preferably, the detector comprises any one of a photomultiplier tube or an avalanche photodiode.

[0021] Preferably, obtaining the light signal reflected by the brain area and performing real-time processing by FPGA to obtain processing information includes:

[0022] amplifying and performing analog-to-digital conversion on the light signal reflected by the brain region to obtain a conversion signal;

[0023] The conversion signal is processed in parallel based on FGPA to obtain the processing information; the parallel processing process includes: filtering, feature extraction and pattern recognition.

[0024] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0025] The present invention provides an information acquisition method for a brain-computer interface device, comprising: using a light source in a head-mounted device to emit multi-wavelength pulsed light to a target user wearing the head-mounted device; processing the optical signal of the multi-wavelength pulsed light through a time domain delay module and a wavelength-dependent spectroscopic module to generate a sparse pulsed light train; using a sensor network to transmit the sparse pulsed light train to a sensing front end, and sending it to the brain area of ​​the target user through the sensing point of the sensing front end; using a detector to obtain the optical signal reflected by the brain area, and performing real-time processing through an FPGA to obtain processing information; using the processing information to drive an external device to complete the recognition and execution of the intention of the target user. The present invention effectively optimizes the multi-wavelength optical signal by combining the time domain delay module and the wavelength-dependent spectroscopic module, generates a sparse pulsed light train, reduces interference between optical signals, and improves signal clarity. In addition, the present invention uses FPGA for signal processing, which can complete complex calculations in a very short time, supports real-time response, and is suitable for application scenarios with rapid response. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0027] Figure 1 A flow chart of a method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0030] Figure 1 A flow chart of a method provided by an embodiment of the present invention, such as Figure 1 As shown, the present invention provides an information acquisition method for a brain-computer interface device, comprising:

[0031] Step 100: Using a light source in a head mounted device to emit multi-wavelength pulsed light to a target user wearing the head mounted device;

[0032] Step 200: Processing the optical signal of the multi-wavelength pulse light through a time domain delay module and a wavelength-dependent spectroscopic module to generate a sparse pulse light train;

[0033] Step 300: Transmitting the sparse pulse light train to the sensing front end by using the sensing network, and sending it to the brain area of ​​the target user through the sensing points of the sensing front end;

[0034] Step 400: using a detector to obtain the light signal reflected by the brain area, and performing real-time processing through an FPGA to obtain processing information;

[0035] Step 500: Utilize the processed information to drive an external device to complete the recognition and execution of the target user's intention.

[0036] Preferably, the multi-wavelength pulse light includes near-infrared light, visible light and ultraviolet light.

[0037] Specifically, the wavelength pulse light of this embodiment refers to light containing multiple different wavelengths in one pulse, and the application of this light depends on its response characteristics to different targets and materials. In brain-computer interface (BCI) and other optical applications, multi-wavelength pulse light includes the following types of light:

[0038] Near-infrared light (NIR): Mainly used in biological imaging and brain activity monitoring because it can penetrate biological tissues and effectively obtain signals deep in the brain.

[0039] Visible light: includes red, orange, yellow, green, blue, indigo, purple and other colors of light. It can be used for general biosensing and visual stimulation research.

[0040] Ultraviolet light (UV): Mainly used to kill bacteria or perform specific material analysis.

[0041] Furthermore, in practical applications, the multi-wavelength pulse light of this embodiment is usually composed of the following wavelength combinations:

[0042] Near infrared light wavelength: 750-1400nm (e.g. 800nm, 980nm, 1060nm, etc.)

[0043] Visible light wavelength: 400-700nm (e.g. 450nm, 532nm, 610nm, 650nm, etc.)

[0044] UV wavelength: 200-400nm (e.g. 280nm, 355nm, etc.)

[0045] Furthermore, the application scenarios of this embodiment are as follows:

[0046] Brain blood oxygen monitoring: monitors blood oxygen levels by using near-infrared light (such as 800nm ​​and 940nm) because hemoglobin has different absorption characteristics for different wavelengths of light.

[0047] Tissue Imaging: Using a combination of visible and near-infrared light, tissue imaging and spectral analysis are performed to obtain more detailed information.

[0048] Stimulus-response experiment: Combining visible light and near-infrared light, we study the stimulation and response of light on the nervous system and promote our understanding of brain function.

[0049] In summary, the multi-wavelength pulsed light of this embodiment is of great significance in brain-computer interfaces and other biomedical applications, and can improve the accuracy and reliability of signal acquisition through the synergistic effect of different wavelengths.

[0050] Preferably, the optical signal of the multi-wavelength pulse light is processed by a time domain delay module and a wavelength-dependent spectroscopic module to generate a sparse pulse light train, including:

[0051] Processing the input optical signal of the multi-wavelength pulse light through a time domain delay module to generate a plurality of multi-wavelength pulse light trains;

[0052] Each multi-wave pulse light train is decomposed by a wavelength-dependent spectroscopic module to obtain multiple sparse pulse light trains.

[0053] Specifically, the processing steps of this embodiment are as follows:

[0054] 1. Input multi-wavelength pulse light. First, the device generates pulse light containing multiple different wavelengths (multi-wavelength pulse light) through a light source.

[0055] 2. The time domain delay module processes the input multi-wavelength pulse light signal and generates multiple multi-wavelength pulse light trains. This module adjusts and delays the signal time of different wavelength pulse light so that each multi-wavelength pulse light train can exist in parallel. Each light train contains the same wavelength components but is staggered in time.

[0056] 3. The wavelength-dependent spectroscopic module further processes the above-mentioned multiple multi-wavelength pulse light trains and decomposes them into N sparse pulse light trains. Through wavelength-dependent spectroscopic technology (such as fiber optic splitters, diffractive optical elements, etc.), pulses of different wavelengths are spectrally separated and filtered, so that each output sparse pulse light train contains only pulse light of a single wavelength.

[0057] 4. Output sparse pulse light trains, and finally send the generated N sparse pulse light trains to multiple sensing points at the sensing front end, so as to send these light signals to the target area (such as the target brain).

[0058] Preferably, the sensor network includes a sensor server and multiple sensor nodes; the sensor server periodically obtains communication status information of each sensor node in the topological network structure and the sensor nodes within a preset distance range around it, and inputs the obtained communication status information into a pre-established node connection decision model to obtain a real-time decision result, and based on the real-time decision result, dynamically adjusts the connection relationship between each sensor node in the topological network structure and the sensor nodes within the preset distance range around it in real time to form an adaptive topological network structure; the real-time decision result includes sensor nodes that need to be connected and disconnected in the topological network structure; the sensor server transmits data with each sensor node based on the adaptive topological network structure.

[0059] Specifically, the machine learning model can be a classifier that determines whether the sensor nodes around each sensor node in the topological network structure are connected or not, or it can be a regression model that predicts the quality or utility of the connection of each sensor node. The machine learning algorithms used by the machine model can include decision trees, support vector machines (SVMs), neural networks, random forests, gradient boosting machines (GBMs), deep learning, etc. The choice of which algorithm depends on the specific application scenario, the characteristics of the data, and the available computing resources. In practical applications, the real-time nature of the algorithm also needs to be considered, that is, the inference speed of the model must be fast enough to adapt to the dynamically changing network environment.

[0060] Preferably, the communication status information includes at least one of signal strength, data transmission rate, packet loss rate, delay, energy consumption, moving speed and historical connection quality.

[0061] Preferably, the method for constructing the node connection decision model includes:

[0062] The sample data included in the training set is input into the machine learning model for iterative training until the loss function of the machine learning model meets the requirements, and the trained machine learning model is obtained as the node connection decision model; the sample data includes historical communication status information of each sensor node and the sensor nodes within a preset distance range around it in a known topological network structure, and the historical connection relationship between each sensor node and the sensor nodes within a preset distance range around it under different historical communication states.

[0063] Preferably, the detector comprises any one of a photomultiplier tube or an avalanche photodiode.

[0064] Preferably, obtaining the light signal reflected by the brain area and performing real-time processing by FPGA to obtain processing information includes:

[0065] amplifying and performing analog-to-digital conversion on the light signal reflected by the brain region to obtain a conversion signal;

[0066] The conversion signal is processed in parallel based on FGPA to obtain the processing information; the parallel processing process includes: filtering, feature extraction and pattern recognition.

[0067] Specifically, the processing steps of this embodiment are as follows:

[0068] 1. Light signal acquisition: The detector captures light signals reflected from brain areas, which may contain relevant information about brain electrical activity.

[0069] 2. Signal amplification: Since the detected optical signal is usually very weak, it is necessary to enhance the signal through an amplification circuit. The amplifier is used to increase the weak signal for subsequent processing. This embodiment uses the amplified signal to provide sufficient strength for further signal analysis and feature extraction.

[0070] 3. Analog-to-digital conversion (ADC) converts the amplified analog signal into a digital signal. This embodiment uses the ADC module to convert the analog signal into a digital format suitable for FPGA processing. Only digital signals can be processed by FPGA, so this is a key conversion step.

[0071] 4. FPGA real-time processing, using the parallel processing capability of FPGA to perform multiple processing on digital signals. The specific process is as follows:

[0072] Filtering, using digital filters to remove noise and interference from the signal, selectively retaining brain waves in specific frequency bands (such as α, β waves, etc.).

[0073] Feature extraction, extracting useful features from the filtered signal, such as power spectrum density, peak frequency, etc., to represent the user's EEG activity state.

[0074] Pattern recognition, applying machine learning algorithms (such as support vector machines, deep learning) to classify the extracted features and identify the user's intention or state.

[0075] 5. Data output and instruction generation: convert the FPGA processing results into operational control instructions, and generate corresponding control instructions based on the recognized EEG patterns, such as controlling external devices (such as wheelchairs, computer cursors, etc.).

[0076] 6. Feedback mechanism: the system can provide users with feedback on the results of the execution through visual or other means. Feedback information helps users adjust their EEG activities, thereby improving the efficiency and accuracy of subsequent operations.

[0077] Furthermore, the external devices of this embodiment include various types, which are mainly selected according to user needs and application scenarios. Examples of external devices are as follows:

[0078] 1. Medical equipment

[0079] Assistive rehabilitation equipment: For example, electric wheelchairs, gait trainers, etc., which control movement or operation through the user's brain signals.

[0080] Medical monitoring equipment: such as heart rate monitors and blood oxygen saturation monitoring devices, which can provide real-time feedback on the user's physiological status.

[0081] 2. Computers and smart devices

[0082] Computer control: Control the computer interface through the user's EEG signals to perform tasks such as text input, web browsing, and file management.

[0083] Smartphone or tablet: Use brain signals to control applications on the phone to make calls, send messages, etc.

[0084] 3. Smart home devices

[0085] Smart lamps: adjust the brightness or color of lights based on the user's brain signals.

[0086] Temperature control devices: Control the air conditioning or HVAC system in the home through the user's intention to adjust the indoor temperature.

[0087] Home appliance control: Multimedia devices such as televisions and stereos can be turned on and off and the volume adjusted through brain signals.

[0088] 4. Virtual reality and augmented reality devices

[0089] VR headsets: Users can directly influence interactions in a virtual environment through their EEG activity. Such applications are suitable for games and training simulations.

[0090] AR device: Augmented reality glasses that use brain signals to display content and operate functions.

[0091] 5. Robots

[0092] Personal assistant robots: For example, nursing robots or home assistants, which are able to assign and execute tasks according to the user's intentions.

[0093] Industrial robots: In industrial environments, robotic arms or other automated equipment are controlled through brain signals to improve operational flexibility and precision.

[0094] 6. Research and educational equipment

[0095] Experimental tools: Experimental equipment used for brain science research, such as brain wave analyzers, neurofeedback equipment, etc.

[0096] Educational software: Suitable for brain-computer interface education platform, helping students master learning content, and can detect learning status through EEG signals.

[0097] 7. Entertainment system

[0098] Music player: Select and play music through brain signals, and even directly control the type and order of songs played through intention.

[0099] Gaming equipment: can adjust the game plot or difficulty according to the player's brain wave response to enhance immersion.

[0100] In summary, the brain-computer interface device of this embodiment can provide connections with a variety of external devices to meet different application requirements. Users can control external devices through direct EEG signals, improve the convenience and efficiency of operation, and promote the breadth and depth of technology application.

[0101] The beneficial effects of the present invention are as follows:

[0102] (1) The multi-wavelength pulsed light of the present invention can penetrate biological tissues and respond to different biological characteristics, thereby improving the ability to detect brain activity. The use of multiple wavelengths provides more biological information, thereby enabling more complex EEG activity analysis.

[0103] (2) The present invention effectively optimizes multi-wavelength optical signals by combining a time-domain delay module and a wavelength-dependent spectroscopic module, generates sparse pulse light trains, reduces interference between optical signals, and improves signal clarity. Using FPGA for signal processing can complete complex calculations in a very short time, support real-time response, and is suitable for fast-response application scenarios.

[0104] (3) The present invention uses a highly sensitive detector to accurately capture weak reflected light signals, so that brain signal information can be effectively obtained even under low signal strength conditions. And through precise pulse light signal control and detection, noise caused by environmental factors can be effectively eliminated, thereby improving the quality of data collection.

[0105] (4) The head-mounted device of the present invention is convenient for users to wear and is suitable for use in different environments, such as medical, home and workplace. The technology can be connected to a variety of external devices (such as wheelchairs, computers, smart homes, etc.), enhancing the breadth and flexibility of application.

[0106] (5) The present invention can achieve more efficient and accurate human-computer interaction and provide a more natural user experience by detecting the user's brain electrical activity and processing the output in real time as a control signal. The device can be adjusted in real time according to the user's intention, forming a good feedback loop and enhancing the user's sense of control and fluency over the device.

[0107] (6) The present invention uses optical methods to obtain signals, avoiding the discomfort and risks caused by traditional invasive methods. The device can be made of biocompatible materials to improve the comfort and safety of wearing and is suitable for long-term use.

[0108] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0109] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for acquiring information of a brain-computer interface device, characterized in that: include: Using a light source in the head-mounted device, emitting multi-wavelength pulsed light to a target user wearing the head-mounted device; Processing the optical signal of the multi-wavelength pulse light by a time domain delay module and a wavelength-dependent spectroscopic module to generate a sparse pulse light train; The sparse pulse light train is transmitted to a sensing front end by using a sensing network, and is sent to a brain area of ​​the target user through a sensing point of the sensing front end; Acquire the light signal reflected by the brain region by using a detector, and process it in real time by using an FPGA to obtain processing information; The processing information is used to drive an external device to complete the recognition and execution of the target user's intention.

2. The information acquisition method of the brain-computer interface device according to claim 1, characterized in that: The multi-wavelength pulse light includes near-infrared light, visible light and ultraviolet light.

3. The information acquisition method of the brain-computer interface device according to claim 1, characterized in that: The optical signal of the multi-wavelength pulse light is processed by a time domain delay module and a wavelength-dependent spectroscopic module to generate a sparse pulse light train, including: Processing the input optical signal of the multi-wavelength pulse light through a time domain delay module to generate a plurality of multi-wavelength pulse light trains; Each multi-wave pulse light train is decomposed by a wavelength-dependent spectroscopic module to obtain multiple sparse pulse light trains.

4. The information acquisition method of the brain-computer interface device according to claim 1, characterized in that: The sensor network includes a sensor server and a plurality of sensor nodes; the sensor server periodically obtains communication status information of each sensor node in the topological network structure and sensor nodes within a preset distance range around it, and inputs the obtained communication status information into a pre-established node connection decision model to obtain a real-time decision result, and based on the real-time decision result, dynamically adjusts the connection relationship between each sensor node in the topological network structure and sensor nodes within a preset distance range around it in real time to form an adaptive topological network structure; The real-time decision result includes sensor nodes that need to be connected and sensor nodes that need to be disconnected in the topological network structure; The sensor server performs data transmission with each sensor node based on an adaptive topology network structure.

5. The information acquisition method of the brain-computer interface device according to claim 4, characterized in that: The communication status information includes at least one of signal strength, data transmission rate, packet loss rate, delay, energy consumption, moving speed and historical connection quality.

6. The information acquisition method of the brain-computer interface device according to claim 4, characterized in that: The method for constructing the node connection decision model includes: The sample data included in the training set is input into the machine learning model for iterative training until the loss function of the machine learning model meets the requirements, and the trained machine learning model is obtained as the node connection decision model; the sample data includes historical communication status information of each sensor node and the sensor nodes within a preset distance range around it in a known topological network structure, and the historical connection relationship between each sensor node and the sensor nodes within a preset distance range around it under different historical communication states.

7. The information acquisition method of the brain-computer interface device according to claim 1, characterized in that: The detector includes any one of a photomultiplier tube or an avalanche photodiode.

8. The information acquisition method of the brain-computer interface device according to claim 1, characterized in that: The optical signal reflected by the brain region is obtained and processed in real time by FPGA to obtain processing information, including: amplifying and performing analog-to-digital conversion on the light signal reflected by the brain region to obtain a conversion signal; The conversion signal is processed in parallel based on FGPA to obtain the processing information; the parallel processing process includes: filtering, feature extraction and pattern recognition.

Images