Skin-like brain-computer flexible electrode, preparation method, interface signal acquisition system and processing method

Abstract

The skin-like electroencephalogram flexible electrode is high in biocompatibility and fitting degree, the electrode can be efficiently fitted with the scalp surface, interference of motion artifacts on signals is remarkably reduced, and stability and reliability of signal collection are improved. The electrode adopts a full-band low-gain direct-current coupling type pre-amplification technology, so that the signal quality can be effectively improved, and the capability of capturing weak electroencephalogram signals is enhanced. And meanwhile, precise analysis and real-time feedback of the brain function state are realized by combining a personalized coding and decoding technology with high temporal-spatial resolution. According to the technology, the electroencephalogram signals can be accurately captured and analyzed in real time, dynamic adjustment and optimization can be carried out according to individual differences of patients, and the monitoring accuracy and individuation level are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of brain function monitoring, and particularly to a skin-like brain-machine flexible electrode and a preparation method thereof, an interface signal acquisition system and a processing method thereof. Background Art

[0002] At present, rehabilitation intervention has become an important core strategy for treating brain function disorders. Especially in the recovery process of neurological diseases such as stroke and traumatic brain injury, precise and efficient rehabilitation intervention plays a crucial role in the functional reconstruction of patients and the improvement of their quality of life. However, the current rehabilitation medical industry still faces severe challenges such as the excessive workload of therapists, shortage of medical resources, and the difficulty in quantifying and ensuring the rehabilitation effect. There is an urgent need for efficient and precise technical means to improve the effect and accessibility of rehabilitation intervention.

[0003] Brain function area monitoring is an important technical means for stroke rehabilitation. By real-time monitoring the functional activities of different brain regions, it can provide a scientific basis for adjusting treatment plans and evaluating treatment effects during the rehabilitation process. However, existing wearable monitoring devices such as electroencephalogram (EEG) and near-infrared spectroscopy (NIRS) have problems such as poor adhesion and significant motion artifacts. In particular, traditional cap structures are prone to noise interference when patients move or have large movements, significantly affecting the accuracy and stability of signals. In addition, traditional devices have certain limitations in signal amplification and decoding technologies, and it is difficult to meet the requirements of high-precision brain function monitoring. Summary of the Invention

[0004] To achieve the above objects and other advantages of the present invention, the first object of the present invention is to provide a skin-like brain-machine flexible electrode, which adopts a skin-like nano-coupled electrode-skin interface electrical model, and the skin-like nano-coupled electrode-skin interface electrical model includes a skin-like layer, a fractal meandering interconnect layer, and a cerebral cortex.

[0005] Furthermore, the skin-like layer is made of conductive hydrogel by a physical cross-linking method.

[0006] The second object of the present invention is to provide a preparation method of a skin-like brain-machine flexible electrode for preparing the above-mentioned skin-like brain-machine flexible electrode, which includes the following steps:

[0007] Obtain the curved surface data of the human head to establish a 3D curve model of the human brain;

[0008] Based on the isometric transformation theory from a developable surface to a plane, perform mathematical modeling on the 3D curve model of the human brain to generate a Gaussian curvature contour map;

[0009] Adopt a near-zero strain 3D curved surface transfer printing process to accurately transfer the prefabricated skin-like nano-coupled electrode structure from a two-dimensional plane to a three-dimensional curved surface head mold, ensuring a high degree of fit between the electrode and the scalp.

[0010] The third object of the present invention is to provide a skin-like brain-computer interface signal acquisition system, including an electrode module, a signal processing module, and a main controller. The electrode module includes a plurality of the above-mentioned skin-like brain-computer flexible electrodes. The signal processing module includes a plurality of amplifiers, an analog multiplexer, and an analog-to-digital converter. The skin-like brain-computer flexible electrodes of each channel are respectively connected to the analog multiplexer through two-stage amplifiers, and the analog multiplexer is connected to the main controller through the analog-to-digital converter.

[0011] The fourth object of the present invention is to provide a method for processing skin-like brain-computer interface signals, which processes the signals collected by the above-mentioned system, including the following steps:

[0012] Decompose the multi-dimensional electroencephalogram signals by the NA-FMEMD method, and decompose the original multi-dimensional electroencephalogram signals into a group of intrinsic mode components with neat modes;

[0013] Extract and analyze the features of the intrinsic mode components obtained by decomposition in the spatial domain through the CCA method, and identify the relevant components of the electroencephalogram signals and artifact signals on different spatial channels;

[0014] Identify the artifact components according to the characteristic expression differences between the electroencephalogram signals and the artifact signals in the spatio-temporal domain, and realize the removal of artifacts in the brain-computer interface signals;

[0015] Realize personalized and accurate motion intention recognition by dynamically connecting multi-band causal brain networks and with the aid of a deep learning framework.

[0016] Furthermore, the step of realizing personalized and accurate motion intention recognition by dynamically connecting multi-band causal brain networks and with the aid of a deep learning framework includes:

[0017] Use a one-dimensional convolutional kernel to perform time-domain filtering on the electroencephalogram signals of each lead to capture the rhythm features in different time dimensions;

[0018] Use the common spatial pattern algorithm to perform feature mapping on the filtered signals and extract a frequency-sensitive feature matrix;

[0019] In the source space dimension, pay attention to the causal sequence of neural activities and construct a multi-band causal dynamic brain network;

[0020] Obtain the spatial position and time information of the cortical neural activity sources through source localization algorithms with different rhythms;

[0021] Based on the source localization information, a causal dynamic network between brain regions is established to capture the high-dimensional brain functional connectivity features in terms of frequency, time, and space;

[0022] The high-dimensional features in the measurement space and the source space are integrated by a deep learning framework based on a convolutional neural network;

[0023] All feature matrices are flattened and concatenated to form a one-dimensional feature vector.

[0024] Further, the step of performing time-domain filtering on the EEG signals of each lead using a one-dimensional convolutional kernel to capture the rhythm features in different time dimensions includes:

[0025] The neural network training process is accelerated through a batch normalization module to prevent gradient vanishing or gradient explosion;

[0026] The overfitting phenomenon is reduced through a dropout module to improve the robustness of the network;

[0027] The EEG signals are divided into multiple frequency bands by a filter bank to enhance the detection sensitivity of the movement-related frequency bands.

[0028] Further, the deep learning framework based on the convolutional neural network includes a first-layer convolutional kernel, a max-pooling layer, and a deep convolutional layer; among them,

[0029] The first-layer convolutional kernel is used to extract the shallow features of the feature matrix;

[0030] The max-pooling layer is used to remove redundant information, reduce the computational amount, and improve the generalization ability of the network;

[0031] The deep convolutional layer is used to further extract deep information and capture the complex feature patterns across time, space, and frequency.

[0032] The fifth object of the present invention is to provide a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, the steps of the above method are implemented.

[0033] The sixth object of the present invention is to provide a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0034] Compared with the prior art, the beneficial effects of the present invention are:

[0035] In view of the above problems, the present invention has developed a skin-like EEG flexible electrode with high biocompatibility and high fitting degree. This electrode can be efficiently attached to the scalp surface, significantly reducing the interference of motion artifacts on signals and improving the stability and reliability of signal acquisition. The electrode adopts a full-band low-gain DC-coupled preamplification technology, which can effectively improve the signal quality and enhance the ability to capture weak EEG signals. At the same time, combined with a personalized encoding and decoding technology with high spatio-temporal resolution, it realizes the accurate analysis and real-time feedback of brain function states. This technology can not only capture and analyze EEG signals in real time and accurately, but also dynamically adjust and optimize according to the individual differences of patients, significantly improving the accuracy and personalization level of monitoring.

[0036] The above description is only an overview of the technical solution of the present invention. In order to be able to more clearly understand the technical means of the present invention and implement it in accordance with the content of the specification, the following will be described in detail with reference to the preferred embodiments of the present invention and the accompanying drawings. The specific implementation manner of the present invention is given in detail by the following embodiments and their accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The drawings described herein are used to provide a further understanding of the present invention, and constitute a part of this application. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0038] Figure 1 It is a schematic diagram of the electrical model of the skin-like nano-coupled electrode-skin interface;

[0039] Figure 2 It is a flowchart of the preparation method of the skin-like brain-computer flexible electrode;

[0040] Figure 3 It is a 3D curved surface near-zero strain transfer strategy for the fractal snake-shaped network conductive structure;

[0041] Figure 4 It is a schematic diagram of the EEG acquisition circuit architecture;

[0042] Figure 5 It is a flowchart of the signal processing method for the skin-like brain-computer interface;

[0043] Figure 6 It is a technical roadmap for artifact removal;

[0044] Figure 7 It is an overall roadmap for personalized encoding and decoding based on deep learning;

[0045] Figure 8 It is a schematic diagram of a computer device;

[0046] Figure 9 It is a schematic diagram of a computer-readable storage medium. DETAILED DESCRIPTION OF THE INVENTION

[0047] The present invention is further described below in conjunction with the accompanying drawings and specific implementation methods. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. It should be noted that, under the premise of no conflict, the embodiments or technical features described below can be arbitrarily combined to form a new embodiment.

[0048] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0049] The figure numbers in this application are only used to distinguish the various steps in the scheme, and are not used to limit the execution order of the various steps. The specific execution order is subject to the description in the specification.

[0050] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0051] Example 1

[0052] A skin-like brain-machine flexible electrode adopts a skin-like nano-coupled electrode-skin interface electrical model, such as Figure 1 As shown, the skin-like nano-coupled electrode-skin interface electrical model includes a multi-layer structure including a skin-like layer (hydrogel layer), a fractal meandering interconnect layer and a cerebral cortex, forming an efficient electrical signal conduction channel.

[0053] Through theoretical calculation and experimental testing, the model was accurately mapped and parameterized, and a single-electrode electrical model with clear performance parameters was established. In order to improve the signal-to-noise ratio (SNR) of the acquired signal, the system introduced the electrode effective area compensation rate δ for electrode points at different locations. Based on the 10-20 international standard lead system, the construction of a distributed electrode network for the whole brain was achieved.

[0054] In order to improve the biocompatibility of EEG acquisition equipment, conductive hydrogel is used as the core material of the skin-like layer. The main components of the hydrogel are selected from biosafety materials certified by the US FDA (Food and Drug Administration), including polyvinyl alcohol (PVA), polyethylene glycol (PEG) and chitosan derivatives. These materials have excellent biocompatibility, low toxicity and mechanical flexibility, ensuring that they will not cause skin allergies or inflammatory reactions when in contact with the skin for a long time.

[0055] The forming process of the hydrogel relies on its physically cross-linkable functional groups and is formed by physical cross-linking methods. During the forming process, the intermolecular forces between hydrogel molecules are mainly achieved through hydrogen bonds, significantly enhancing the structural stability of the material while maximizing the retention of the characteristic functional groups of the material. This property enables the hydrogel to maintain good electrical conductivity and mechanical stability under mechanical stretching, compression, and repeated use, providing high stability and reusability for the skin-like electrode in practical application scenarios.

[0056] Based on the skin-like nano-coupled sensor design and processing technology, the present invention forms a skin-like brain-computer interface signal acquisition system.

[0057] In the construction of the brain-computer interface signal acquisition system, a distributed electrode-skin electrical model and a mechanical-electrical performance coupling model are introduced. Combining material science and electronic engineering technologies, the geometric dimensions, material properties, and mechanical coupling attributes of the electrode are standardized and optimized using the step-by-step method, ensuring a high degree of fit between the electrode and the skin, effectively reducing noise interference caused by motion artifacts and small displacements, and providing technical support for stable and high-quality signal acquisition.

[0058] Example 2

[0059] A preparation method for a skin-like brain-computer flexible electrode is used to prepare the above-mentioned skin-like brain-computer flexible electrode. For a detailed description of the skin-like brain-computer flexible electrode, reference can be made to the corresponding description in Example 1, which will not be elaborated here. As Figure 2 、 Figure 3 shown, it includes the following steps:

[0060] S110. Obtain the curved surface data of the human head to establish a 3D curve model of the human brain;

[0061] Optionally, a high-precision scan of the participant's head is performed using a handheld 3D scanner to obtain complete head curved surface data and reconstruct a three-dimensional brain curve surface model, ensuring that the electrode can highly match the individual's head curved surface during subsequent preparation, reducing signal loss and artifact interference caused by geometric mismatch of traditional electrodes.

[0062] S120. Based on the isometric transformation theory from a developable surface to a plane, perform mathematical modeling on the 3D curve model of the human brain to generate a Gaussian curvature contour map; specifically, taking the central region of the brain as the center point, establish multiple transformable curves, and map the three-dimensional surface to a two-dimensional plane in a near-zero strain manner in space, thereby avoiding microcracks or stress concentration caused by material deformation during two-dimensional material preparation and maximizing the integrity and electrical conductivity of the electrode.

[0063] S130. Adopt the near-zero strain 3D curved surface transfer printing process to accurately transfer the prefabricated skin-like nano-coupled electrode structure from a two-dimensional plane to a three-dimensional curved surface head mold, ensuring a high degree of fit between the electrode and the scalp. For details, see Figure 3 .

[0064] The present invention adopts a near-zero strain 3D curved surface transfer printing strategy to achieve the precise preparation of a large-area, highly biocompatible skin-like nano-coupled electrophysiological sensor, effectively solving the problems of poor adhesion and unstable signal transmission of traditional electrodes on complex curved surfaces (such as the scalp surface).

[0065] Example 3

[0066] A skin-like brain-computer interface signal acquisition system, as Figure 4 shown, includes an electrode module, a signal processing module, and a main controller. The electrode module includes a plurality of the above-mentioned skin-like brain-computer flexible electrodes. For a detailed description of the skin-like brain-computer flexible electrodes, reference can be made to the corresponding description in Example 1, which will not be elaborated here. The signal processing module includes a plurality of amplifiers, an analog multiplexer, and an analog-to-digital converter. The skin-like brain-computer flexible electrodes of each channel are respectively connected to the analog multiplexer through two-stage amplifiers, and the analog multiplexer is connected to the main controller through the analog-to-digital converter. In this embodiment, the analog-to-digital converter communicates with the main controller using the SPI communication protocol; the analog-to-digital converter uses a 16-bit analog-to-digital converter.

[0067] For brain-computer interface signal acquisition, in order to improve signal quality and reduce noise interference, a full-band low-gain DC-coupled preamplification technology is adopted.

[0068] By analyzing the bandwidth of the electroencephalogram (EEG) signal, clarifying the characteristic distribution and physiological significance of signals in different frequency bands during brain activities, and determining the specific frequency band parameters of band-pass filtering to ensure the effective extraction and amplification of signals in the target frequency band. According to the signal characteristics of different frequency bands, the amplification factor parameters are designed to achieve high-fidelity amplification of weak EEG signals and minimize the interference of non-target frequency components to the greatest extent.

[0069] In the context of the demand for wireless communication, the system, based on theoretical calculations, clarifies the relationship between analog-to-digital conversion (AD sampling frequency) and the data throughput of wireless communication to ensure that data will not be distorted or lost during transmission. The system introduces a high-speed AD converter to ensure fast and high-precision analog-to-digital conversion of multi-channel EEG signals.

[0070] In terms of the hardware architecture, the system has carried out multiple rounds of circuit design optimization to ensure the electrical isolation between circuit modules and the efficiency of the signal transmission path. After theoretical calculations, simulation tests, and actual verification, the design scheme of the unit circuit is finally determined.

[0071] Example 4

[0072] A method for processing signals of a skin-like brain-computer interface processes the signals collected by the above system. For a detailed description of the skin-like brain-computer interface signal acquisition system, reference can be made to the corresponding description in Example 3, which will not be elaborated here. Regarding the artifact problem, a method for removing the hybrid artifacts of brain-computer interface signals based on multi-dimensional spatio-temporal-frequency related component decomposition is proposed. Specifically, as Figures 5 - 7 shown, it includes the following steps:

[0073] S210. Decompose the multi-dimensional electroencephalogram (EEG) signals by the NA-FMEMD (Improved Noise-Assisted Multi-Dimensional Ensemble Empirical Mode Decomposition) method, and decompose the original multi-dimensional EEG signals into a group of intrinsic mode functions (IMFs) with neat modes. While retaining the time, space, and frequency characteristics of the signals, NA-FMEMD can effectively reduce the boundary effect and mode mixing problems existing in traditional decomposition methods.

[0074] S220. After obtaining the intrinsic mode functions with neat modes, perform feature extraction and analysis on the decomposed intrinsic mode functions in the spatial domain by the CCA (Canonical Correlation Analysis) method, and effectively identify the relevant components of the EEG signals and artifact signals on different spatial channels;

[0075] The spatio-temporal-frequency related component decomposition of multi-dimensional signals is realized based on NA-FMEMD and CCA.

[0076] S230. According to the difference in the characteristic expressions of the EEG signals and artifact signals in the spatio-temporal domain, effectively identify the artifact components, and realize the fast and high-quality removal of artifacts in the brain-computer interface signals;

[0077] S240. Dynamically connect the multi-band causal brain network and, with the help of a deep learning framework, realize personalized and accurate recognition of motion intentions.

[0078] The present invention proposes a personalized encoding and decoding technology with high spatio-temporal resolution, deeply fuses the brain-computer interface signals in the measurement space and the source space, combines the time-domain characteristics, rhythm characteristics, and spatial distribution characteristics, constructs a multi-band causal connection dynamic brain network, and with the help of a deep learning framework, realizes personalized and accurate recognition of motion intentions.

[0079] In some embodiments, the step of dynamically connecting the multi-band causal brain network and, with the help of a deep learning framework, realizing personalized and accurate recognition of motion intentions includes:

[0080] In the measurement space, the system uses a one-dimensional convolutional kernel to perform time-domain filtering on the electroencephalogram (EEG) signals of each lead in order to capture the rhythm characteristics in different time dimensions. The specific modules are as follows:

[0081] Batch normalization module: Accelerates the neural network training process and prevents gradient vanishing or gradient explosion.

[0082] Dropout module: Reduces the overfitting phenomenon and improves the robustness of the network.

[0083] For the multi-band signals in the measurement space, the system introduces a filter bank to divide the EEG signals into multiple frequency bands and enhance the detection sensitivity of the motion-related frequency bands.

[0084] The common spatial pattern (CSP) algorithm is used to perform feature mapping on the filtered signals to extract a frequency-sensitive feature matrix;

[0085] In the source space dimension, the causal sequence of neural activities is concerned, and a multi-band causal dynamic brain network is constructed;

[0086] Through source localization algorithms with different rhythms, the spatial positions and time information of the cortical neural activity sources are obtained;

[0087] Based on the source localization information, a causal dynamic network between brain regions is established to capture the high-dimensional brain functional connection characteristics in terms of frequency, time, and space;

[0088] In order to fully integrate the high-dimensional features of the measurement space and the source space, the system designs a deep learning framework based on a convolutional neural network (CNN). The specific architecture is as follows:

[0089] The first-layer convolutional kernel: Extracts the shallow features of the feature matrix.

[0090] Max pooling layer: Removes redundant information, reduces the computational amount, and improves the generalization ability of the network.

[0091] Deep convolutional layer: Further extracts deep information and captures complex feature patterns across time, space, and frequency.

[0092] After the feature extraction is completed, the system flattens and concatenates all the feature matrices to form a one-dimensional feature vector.

[0093] To support the personalized encoding and decoding technology with high spatio-temporal resolution, the system designs a core computing platform based on System on Chip (SoC) to achieve efficient neural network computing and data transmission. For complex neural network computing, the data storage is optimized to ensure high bandwidth on-chip storage, so that on-chip data transmission will not become the performance bottleneck of this architecture. Finally, each module is integrated and connected, and the control logic is designed to achieve collaborative work with the external host. At the same time, the driver program and compiler for the neural network need to be completed to achieve convenient and flexible configuration and invocation of the algorithm accelerator. For the designed on-chip system architecture, a reasonable configuration of the hardware computing unit is completed, the computing tasks are split, and different parts of the computing are allocated to different hardware computing resources to complete the computing, so that the whole system can efficiently complete the whole computing task.

[0094] Embodiment 5

[0095] A computer device 300, as Figure 8 shown, includes a memory 310, a processor 320, and a computer program 330 stored on the memory and executable on the processor. When the processor executes the computer program, it implements the steps of a skin-like brain-computer interface signal processing method. For a detailed description of the method, reference can be made to the corresponding description in the above method embodiments, which will not be elaborated here.

[0096] Embodiment 6

[0097] A computer-readable storage medium, as Figure 9 shown, stores a computer program thereon. When the computer program is executed by a processor, it implements the steps of a skin-like brain-computer interface signal processing method. For a detailed description of the method, reference can be made to the corresponding description in the above method embodiments, which will not be elaborated here.

[0098] The number of devices and the scale of processing described here are used to simplify the description of the present invention. Applications, modifications, and variations of the present invention will be obvious to those skilled in the art.

[0099] Although the embodiments of the present invention have been disclosed as above, they are not limited to the applications listed in the specification and embodiments. It can be fully applied to various fields suitable for the present invention. For those familiar with the field, additional modifications can be easily made. Therefore, without departing from the general concept defined by the claims and the equivalent scope, the present invention is not limited to specific details and the examples shown and described here.

[0100] The device, computer device, non-volatile computer storage medium, and method provided by the embodiments of this specification are corresponding. Therefore, the device, computer device, and non-volatile computer storage medium also have beneficial technical effects similar to those of the corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the corresponding device, computer device, and non-volatile computer storage medium will not be elaborated here.

[0101] Those skilled in the art also know that in addition to implementing the controller in the form of pure computer-readable program code, the method steps can be logically programmed to enable the controller to be implemented in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, embedded microcontrollers, etc. to achieve the same function. Therefore, such a controller can be regarded as a hardware component, and the devices included therein for implementing various functions can also be regarded as the structures within the hardware component. Or even, the devices for implementing various functions can be regarded as both software units for implementing the method and structures within the hardware component.

[0102] The systems, devices, or units illustrated in the above embodiments can be specifically implemented by computer chips or entities, or by products with certain functions. For the convenience of description, when describing the above devices, they are described as various units according to their functions. Of course, when implementing one or more embodiments of this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0103] Those skilled in the art should understand that the embodiments of this specification can be provided as a method, system, or computer program product. Therefore, the embodiments of this specification can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of this specification can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program code.

[0104] This specification is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0105] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction device that implements the functions specified in one or more of the processes and / or blocks Figure 1 one or more of the processes and / or blocks Figure 1 specified in the block or blocks.

[0106] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one or more of the processes and / or blocks Figure 1 one or more of the processes and / or blocks Figure 1 specified in the block or blocks.

[0107] It should also be noted that the term "comprising", "including" or any other variation thereof is intended to cover a non-exclusive inclusion, such that a process, method, article or apparatus that comprises a list of elements does not include only those elements but also other elements not expressly listed, or elements that are inherent to such process, method, article or apparatus. Without further limitation, an element defined by the phrase "comprising a …" does not exclude the presence of additional identical elements in the process, method, article or apparatus that comprises the element.

[0108] This specification can be described in the general context of computer-executable instructions executed by a computer, such as program units. Generally, program units include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The specification can also be practiced in a distributed computing environment where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program units can be located in both local and remote computer storage media including storage devices.

[0109] Each embodiment in this specification is described in a progressive manner, and the same or similar parts among the embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, it is described relatively simply, and the relevant parts can be referred to the partial description of the method embodiment.

[0110] The above is only for the embodiments of this specification and is not intended to limit one or more embodiments of this specification. For those skilled in the art, various changes and modifications can be made to one or more embodiments of this specification. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of one or more embodiments of this specification shall be included within the scope of the claims of one or more embodiments of this specification.

Claims

1. A skin-like brain-computer flexible electrode, characterized in that: A skin-like nano-coupled electrode-skin interface electrical model is adopted, and the skin-like nano-coupled electrode-skin interface electrical model includes a skin-like layer, a fractal meandering interconnection line layer and a cerebral cortex.

2. The skin-like brain-computer flexible electrode according to claim 1, wherein: The skin-like layer is made of conductive hydrogel through a physical cross-linking method.

3. A preparation method of a skin-like brain-computer flexible electrode for preparing the skin-like brain-computer flexible electrode according to any one of claims 1 to 2, characterized in that, The following steps are involved: Acquire the human head surface data to build a 3D curve model of the human brain; Based on the isometric transformation theory from developable surface to plane, mathematical modeling is performed on the 3D curve model of the human brain to generate a Gaussian curvature contour map; Using a near-zero strain 3D curved surface transfer process, the prefabricated skin-like nano-coupling electrode structure is accurately transferred from a two-dimensional plane to a three-dimensional curved head mold, ensuring a high fit between the electrode and the scalp.

4. A skin-like brain-computer interface signal acquisition system, characterized in that: It includes an electrode module, a signal processing module, and a main controller. The electrode module includes a plurality of skin-like brain-machine flexible electrodes as described in any one of claims 1 to 2. The signal processing module includes a plurality of amplifiers, an analog multiplexer, and an analog-to-digital converter. The skin-like brain-machine flexible electrodes of each channel are respectively connected to the analog multiplexer via a two-stage amplifier, and the analog multiplexer is connected to the main controller via the analog-to-digital converter.

5. A method for processing signals of a skin-like brain-computer interface, which processes the signals collected by the system as described in claim 4, characterized in that, The following steps are involved: The multidimensional EEG signal is decomposed by NA-FMEMD method, and the original multidimensional EEG signal is decomposed into a set of eigenmodal components with neat modalities. The CCA method is used to extract and analyze the features of the intrinsic modal components obtained by decomposition in the spatial domain, and the related components of the EEG signal and the artifact signal in different spatial channels are identified; According to the characteristic expression differences between EEG signals and artifact signals in the spatiotemporal domain, the artifact components are identified to achieve artifact removal in the brain-computer interface signal. Through multi-band causal connection of dynamic brain networks and with the help of deep learning framework, personalized and accurate movement intention recognition can be achieved.

6. The skin-like brain-computer interface signal processing method according to claim 5, characterized in that The steps of realizing personalized and accurate movement intention recognition by connecting a dynamic brain network through multi-band causality and using a deep learning framework include: A one-dimensional convolution kernel is used to perform time-domain filtering on the EEG signal of each lead to capture the rhythmic features in different time dimensions; The common space pattern algorithm is used to perform feature mapping on the filtered signal and extract the frequency-sensitive feature matrix; In the source space dimension, we focus on the causal order of neural activities and construct a multi-band causal dynamic brain network. Through the source localization algorithm of different rhythms, the spatial location and time information of the source of cortical neural activity are obtained; Based on source localization information, a causal dynamic network between brain regions is established to capture high-dimensional brain functional connectivity characteristics in frequency, time and space; A deep learning framework based on convolutional neural networks integrates high-dimensional features of the measurement space and the source space; All feature matrices are flattened and concatenated to form a one-dimensional feature vector.

7. The method for processing signals of a skin-like brain-computer interface according to claim 6, wherein The step of using a one-dimensional convolution kernel to perform time domain filtering on the EEG signal of each lead to capture rhythm features in different time dimensions includes: Accelerate the neural network training process through the batch normalization module to prevent gradient disappearance or gradient explosion; Reduce overfitting by discarding modules and improve the robustness of the network; The electroencephalogram signal is divided into multiple frequency bands through a filter bank to enhance the detection sensitivity of the movement-related frequency band.

8. The skin-like brain-computer interface signal processing method according to claim 6, characterized in that, The deep learning framework based on a convolutional neural network includes a first-layer convolutional kernel, a max-pooling layer, and a deep convolutional layer; among them, The first-layer convolutional kernel is used to extract the shallow features of the feature matrix; The max-pooling layer is used to remove redundant information, reduce the amount of computation, and improve the generalization ability of the network; The deep convolutional layer is used to further extract deep information and capture complex feature patterns across time, space, and frequency.

9. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 5 to 8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 5 to 8 are implemented.

Images