Brain-computer interface terminal device

Abstract

The application relates to the technical field of brain-computer interfaces, and discloses a brain-computer interface terminal device. The brain-computer interface terminal device comprises a flexible main body frame, a frontal lobe collection array, a dual-function ear clip module, a magnetic suction modular electrode and a master control integrated module; the frontal lobe collection array is arranged on the flexible main body frame and adopts a multi-point potential distribution design; the dual-function ear clip module integrates a reference electrode and an environmental noise sensor and is used for synchronously collecting brain electrical reference potential and environmental electromagnetic interference data; the magnetic suction modular electrode adopts a magnetic suction female buckle interface design to be compatible with dry electrodes and flexible patch electrodes and to realize plug and play; the master control integrated module is electrically connected with the frontal lobe collection array, the dual-function ear clip module and the magnetic suction modular electrode respectively, so as to realize brain electrical signal collection and interference elimination, adaptive adjustment of electrode collection parameters, dynamic switching of a collection channel and a sampling rate, linkage configuration of an electrode working mode and a collection point.

Description

Technical Field

[0001] This application relates to the field of brain-computer interface technology, and in particular to a brain-computer interface terminal device. Background Technology

[0002] With the industrialization of brain-computer interface technology, wearable EEG signal acquisition terminals have gradually expanded from medical research scenarios to diverse scenarios such as home rehabilitation, daily physiological monitoring, and human-computer interaction, which has placed higher comprehensive requirements on the portability, anti-interference, adaptability, battery life, and scenario scalability of the devices.

[0003] Currently, the hardware structure and signal processing logic of existing brain-computer interface (BCI) terminal devices generally adopt independent design patterns, performing unidirectional optimization for a single usage scenario. This results in several technical shortcomings that cannot be addressed collaboratively: First, insufficient anti-interference performance. Existing devices can only handle interference through single software filtering or hardware shielding methods, failing to simultaneously collect data on environmental electromagnetic interference and human dynamic interference sources. In complex unshielded environments such as home and mobile settings, dynamic noise cannot be effectively eliminated, making it difficult to guarantee the stability and accuracy of EEG signal acquisition. Second, poor human adaptation. Existing devices mostly adopt rigid main structures, which cannot adapt to the differences in head physiological configurations of different users, easily leading to poor electrode-skin contact and contact impedance. The excessive size of the device leads to significant differences in signal acquisition quality among different users, making stable cross-user adaptation impossible. Thirdly, the device lacks scalability; the electrode types, acquisition points, and operating modes of existing devices are all fixed designs, unable to be flexibly adjusted according to the needs of different usage scenarios. This makes it difficult to simultaneously meet the high-precision acquisition requirements of medical research scenarios and the low-power consumption requirements of everyday home use, limiting the device's applicability. Fourthly, acquisition performance and battery life cannot be balanced; existing devices generally adopt a fixed operating mode across all channels and all times, unable to dynamically adjust their operating status according to the acquisition needs of EEG signals. This results in high overall power consumption for portable devices, and the battery life cannot meet the core requirements for long-term wearable use. Summary of the Invention

[0004] This application provides a brain-computer interface terminal device, which aims to solve the current problem that the hardware structure design and signal processing logic of existing brain-computer interface terminal devices generally adopt independent design patterns, and are optimized in a single direction for a single use scenario, resulting in many technical defects that cannot be solved collaboratively.

[0005] In a first aspect, embodiments of this application provide a brain-computer interface terminal device, including a flexible main frame, a prefrontal cortex acquisition array, a dual-function ear clip module, magnetically attached modular electrodes, and a main control integrated module. The flexible main frame is made of flexible and elastic materials; the prefrontal cortex acquisition array is deployed on the flexible main frame and adopts a multi-point potential distribution design; the dual-function ear clip module integrates a reference electrode and an environmental noise sensor for synchronously acquiring EEG reference potential and environmental electromagnetic interference data; the magnetically attached modular electrodes adopt a magnetically attached female buckle interface design to be compatible with plug-and-play dry electrodes and flexible patch electrodes; the main control integrated module is respectively connected to the flexible main frame. The device is electrically connected to a prefrontal lobe acquisition array, a dual-function ear clip module, and magnetically attached modular electrodes. The flexible main frame maintains the overall structural stability of the device through an internally embedded support structure. Combined with flexible and elastic materials, it adapts to the forehead curves of different users, allowing the prefrontal lobe acquisition array to form a basic physical fit with the user's forehead skin. The main control integrated module acquires the electrode-skin contact impedance data of each acquisition channel in real time through the prefrontal lobe acquisition array. Based on the impedance data, it identifies the user's forehead shape and electrode fit status, and synchronously adjusts the acquisition parameters of the corresponding acquisition channels. Together with the flexible main frame, it completes the precise adaptation to different users' forehead shapes.

[0006] In some embodiments, the flexible main frame has an embedded support structure to adapt to the forehead curves of different users.

[0007] In some embodiments, the flexible main frame maintains the overall structural stability of the device through an internally embedded support structure, and combines flexible and elastic materials to adapt to the forehead curves of different users, so that the prefrontal lobe acquisition array forms a basic physical fit with the user's forehead skin; the main control integration module acquires the electrode-skin contact impedance data of each acquisition channel in real time through the prefrontal lobe acquisition array, identifies the user's forehead configuration and electrode fit status based on the impedance data, and synchronously adjusts the acquisition parameters of the corresponding acquisition channel, so as to complete the precise adaptation to the forehead configuration of different users in conjunction with the flexible main frame.

[0008] In some embodiments, the prefrontal cortex acquisition array adopts a six-point potential distribution design.

[0009] In some embodiments, the dual-function ear clip module synchronously acquires EEG reference potential and environmental electromagnetic interference data, including: while acquiring EEG reference potential, simultaneously acquiring human body environmental induced current and spatial electromagnetic noise data through an induction circuit.

[0010] In some embodiments, the magnetic female connector can be directly connected to a dry electrode, or connected to a flexible patch electrode via a matching male connector, enabling plug-and-play and quick switching of the electrode.

[0011] In some embodiments, the main control integrated module receives EEG reference potential data and environmental electromagnetic interference data synchronously collected by the dual-function ear clip module, compares and analyzes the two sets of data through a differential amplification processing algorithm, removes environmental interference noise in the EEG signal, and forms an anti-interference closed loop of hardware acquisition and software processing.

[0012] In some embodiments, the main control integrated module monitors the contact impedance between the electrodes and the skin of each acquisition channel of the prefrontal lobe acquisition array in real time. Through the forehead configuration recognition algorithm and the electrode contact status monitoring algorithm, it identifies abnormal channels and calculates the corresponding compensation amount. It then automatically adjusts the gain and filtering parameters of the acquisition channels in real time to ensure the stability of the acquisition signal.

[0013] In some embodiments, during the non-whole EEG feature extraction stage, the main control integration module executes a point-based dynamic sleep and wake-up algorithm to control the device to prioritize low-power mode with fewer channels and a low sampling rate; when a preset specific EEG feature is detected, all acquisition channels are woken up and switched to a high-precision mode with full channels and a high sampling rate; when the acquisition task is completed or timed out, the corresponding acquisition channel is controlled to sleep and switch back to low-power mode.

[0014] In some embodiments, the main control integrated module has a built-in electrode configuration switching module, which controls the working mode of the magnetic modular electrode and the number of activated acquisition points according to the preset acquisition accuracy requirements and usage scenarios, so as to realize the synchronous matching of hardware electrode configuration and software acquisition logic.

[0015] In some embodiments, the main control integration module has a built-in intelligent usage scenario recognition and automatic adaptation algorithm module to automatically identify the user's current usage scenario based on the collected data and automatically switch the configuration of the brain-computer interface terminal device. During the operation of the brain-computer interface terminal device, the main control integration module continuously collects multi-dimensional data, including environmental electromagnetic interference data acquired by the dual-function ear clip module, EEG signal feature data acquired by the prefrontal cortex acquisition array, operating status data of the brain-computer interface terminal device, and user operation habit data. Through the scene intelligent recognition and automatic adaptation algorithm module, based on the EEG signal feature data, device operating status data, and user operation habit data, and through a pre-trained scene classification model, the module identifies the user's current usage scenario, including high-precision acquisition scenarios for medical research, long-term monitoring scenarios for home rehabilitation, real-time human-computer interaction control scenarios, and daily physiological state monitoring scenarios. After identifying the corresponding usage scenario, the module matches the optimal device configuration scheme for the corresponding usage scenario, and adjusts the working mode and number of activated acquisition points of the magnetic modular electrodes, the sampling rate and working mode of the acquisition channel, the parameter configuration of the signal processing algorithm, and the power consumption management mode of the device, while simultaneously adjusting the acquisition logic and feature parsing strategy at the software level.

[0016] In some embodiments, the main control integration module incorporates an EEG intention intelligent prediction and closed-loop pre-linkage algorithm module to predict the user's operational intention based on EEG preparatory potential characteristics, thereby achieving pre-linkage and low-latency closed-loop control of external devices. When the brain-computer interface terminal device is in human-computer interaction or rehabilitation training mode, the main control integration module continuously collects the user's EEG signals at a high sampling rate through a prefrontal cortex acquisition array, capturing EEG preparatory potential characteristics related to the user's action intention and operational decision. Based on the real-time collected EEG signals, the EEG intention intelligent prediction and closed-loop pre-linkage algorithm module identifies characteristic changes in EEG preparatory potentials through a pre-trained temporal prediction model, predicting the user's upcoming operational intention and corresponding control command. After predicting the user's operational intention, the main control integration module sends a pre-linkage command to the corresponding external device through a wireless communication unit, controlling the external device to enter a preparatory running state. After completing the full EEG intention recognition and confirmation, an execution command is sent to control the external device to complete the corresponding action.

[0017] This application utilizes a dual-function ear clip module integrating a reference electrode and an environmental noise sensor to simultaneously acquire EEG reference potential and environmental electromagnetic interference data. Combined with the collaborative processing of the main control integrated module, it can effectively eliminate environmental electromagnetic interference in complex scenarios. Simultaneously, relying on the device's integrateable dual-wavelength optical monitoring module, it acquires human pulse and blood circulation-related electrical signals through 660nm and 940nm dual-wavelength optical signals. This helps identify and accurately eliminate inherent physiological artifacts in EEG signals caused by heartbeat and vascular pulsation. Compared to existing single anti-interference solutions, this approach simultaneously addresses the two core noise sources: environmental interference and human physiological artifacts, significantly improving the cleanliness and acquisition stability of EEG signals.

[0018] By employing a flexible main frame made of flexible and elastic materials, it can adapt to the physiological configuration of different users' foreheads. Combined with the adaptive adjustment function of electrode acquisition parameters in the main control integrated module, it solves the technical problems of poor electrode contact and large differences in signal quality among different users. At the same time, this flexible load-bearing structure can be simultaneously compatible with the integration of dual-wavelength light emitting diodes and photoelectric detectors, ensuring a stable fit between the optical module and the forehead skin, avoiding data distortion caused by light signal leakage. This enables synchronous and stable acquisition of EEG signals and optical physiological signals when worn by different users, eliminating the need for multiple devices to be worn separately, and greatly improving the convenience of wearing the device and the consistency of data acquisition.

[0019] Modular electrodes designed with magnetic female buckle interfaces, combined with the electrode working mode and acquisition point linkage configuration function of the main control integrated module, can be compatible with multiple types of electrodes and acquisition needs in multiple scenarios. At the same time, relying on the infrared acquisition interface of the device, a dual-wavelength optical physiological monitoring module can be seamlessly integrated, enabling the same set of devices to simultaneously realize the acquisition of EEG signals and the monitoring of multi-dimensional physiological parameters such as blood oxygen saturation, heart rate, and pulse wave. This takes into account diverse scenarios such as medical research, neurorehabilitation, sleep monitoring, daily health assessment, and human-computer interaction, greatly expanding the application boundaries and scope of application of brain-computer interface terminal devices.

[0020] By using a prefrontal cortex acquisition array with multi-point potential distribution, combined with the dynamic switching function of the acquisition channel and sampling rate of the main control integrated module, the basic power consumption of the device is significantly reduced while ensuring the accuracy of the acquisition of core EEG features. At the same time, this dynamic power consumption management logic can also cover the dual-wavelength optical monitoring module. In the daily low-power monitoring mode, only the core acquisition channel and the optical monitoring module can be used. When the optical monitoring detects an abnormal physiological state of the user, the full-channel high-precision acquisition mode can be quickly activated to achieve linkage capture of abnormal states. While ensuring the complete acquisition of multimodal data, the overall power consumption of the device is further optimized, effectively extending the device's battery life.

[0021] Simultaneously, an integrated hardware and software collaborative architecture of "sensing-computing-adjustment" has been constructed, which can simultaneously collect EEG neural signals and human physiological state signals obtained by dual-wavelength optical monitoring. Compared with existing terminal devices that can only collect single EEG signals, it can realize the correlation analysis between EEG characteristics and human physiological state, and can more accurately identify the user's neural activity state, emotional state, fatigue level, sleep stage, etc., providing a more comprehensive data source for medical diagnosis, rehabilitation assessment, health management, and precise human-computer interaction, and greatly improving the analysis accuracy and practical application value of the device.

[0022] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic block diagram of the structure of a brain-computer interface terminal device provided in an embodiment of this application; Figure 2 This is a schematic diagram of the structure of a brain-computer interface terminal device provided in one embodiment of this application; Figure 3 This is a block diagram of hardware and software collaborative anti-interference logic provided in an embodiment of this application; Figure 4 This is a flowchart of adaptive impedance monitoring and parameter adjustment provided in one embodiment of this application; Figure 5 This is a flowchart of a dynamic power management algorithm provided in an embodiment of this application.

[0025] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Detailed Implementation

[0026] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0027] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0028] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, the terms "first" and "second" are used in the embodiments of the present invention to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and the terms "first" and "second" are not necessarily different.

[0029] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0030] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0031] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0032] With the industrialization of brain-computer interface technology, wearable EEG signal acquisition terminals have gradually expanded from medical research scenarios to diverse scenarios such as home rehabilitation, daily physiological monitoring, and human-computer interaction, which has placed higher comprehensive requirements on the portability, anti-interference, adaptability, battery life, and scenario scalability of the devices.

[0033] Currently, the hardware structure and signal processing logic of existing brain-computer interface (BCI) terminal devices generally adopt independent design patterns, performing unidirectional optimization for a single usage scenario. This results in several technical shortcomings that cannot be addressed collaboratively: First, insufficient anti-interference performance. Existing devices can only handle interference through single software filtering or hardware shielding methods, failing to simultaneously collect data on environmental electromagnetic interference and human dynamic interference sources. In complex unshielded environments such as home and mobile settings, dynamic noise cannot be effectively eliminated, making it difficult to guarantee the stability and accuracy of EEG signal acquisition. Second, poor human adaptation. Existing devices mostly adopt rigid main structures, which cannot adapt to the differences in head physiological configurations of different users, easily leading to poor electrode-skin contact and contact impedance. The excessive size of the device leads to significant differences in signal acquisition quality among different users, making stable cross-user adaptation impossible. Thirdly, the device lacks scalability; the electrode types, acquisition points, and operating modes of existing devices are all fixed designs, unable to be flexibly adjusted according to the needs of different usage scenarios. This makes it difficult to simultaneously meet the high-precision acquisition requirements of medical research scenarios and the low-power consumption requirements of everyday home use, limiting the device's applicability. Fourthly, acquisition performance and battery life cannot be balanced; existing devices generally adopt a fixed operating mode across all channels and all times, unable to dynamically adjust their operating status according to the acquisition needs of EEG signals. This results in high overall power consumption for portable devices, and the battery life cannot meet the core requirements for long-term wearable use.

[0034] Crucially, existing solutions to address these shortcomings are all isolated, single-point optimizations, focusing only on individual hardware or software improvements to address specific technical defects. They fail to form an integrated hardware and software architecture capable of simultaneously solving all the aforementioned technical problems. Those skilled in the art cannot gain technical insights from existing technologies to collaboratively integrate anti-interference hardware design, flexible adaptability structures, modular electrode configurations, and dynamic intelligent control logic. Consequently, existing brain-computer interface terminal devices have consistently failed to simultaneously meet the comprehensive requirements of high anti-interference, high adaptability, high scalability, and low power consumption, thus limiting the widespread application of brain-computer interface technology across multiple scenarios.

[0035] To solve the above problem, please refer to Figures 1 to 2This application provides a brain-computer interface terminal device, comprising a dual-function ear clip module 1, a flexible main frame 2, magnetically attached modular electrodes 3, a prefrontal cortex acquisition array 4, and a main control integrated module 5. The flexible main frame is made of flexible and elastic materials; the prefrontal cortex acquisition array is deployed on the flexible main frame and employs a multi-point potential distribution design; the dual-function ear clip module integrates a reference electrode and an environmental noise sensor for synchronously acquiring EEG reference potentials and environmental electromagnetic interference data; and the magnetically attached modular electrodes employ a magnetic female connector design to accommodate both dry electrodes and flexible patch electrodes. Ready to use; the main control integrated module is electrically connected to the frontal lobe acquisition array, the dual-function ear clip module, and the magnetic modular electrode, respectively. The flexible main frame maintains the overall structural stability of the device through the internally embedded support structure, and combines flexible and elastic materials to adapt to the forehead curve of different users, so that the frontal lobe acquisition array forms a basic physical fit with the user's forehead skin. The main control integrated module acquires the electrode-skin contact impedance data of each acquisition channel in real time through the frontal lobe acquisition array, identifies the user's forehead shape and electrode fit status based on the impedance data, and synchronously adjusts the acquisition parameters of the corresponding acquisition channel, and completes the precise adaptation to the forehead shape of different users in conjunction with the flexible main frame.

[0036] Specifically, this application addresses the core technical shortcomings of existing brain-computer interface (BCI) terminal devices, such as weak anti-interference capabilities, poor human adaptability, insufficient scenario scalability, and the inability to balance power consumption and performance. It provides a highly integrated hardware and software BCI terminal device, which is built on the core of an integrated collaborative architecture of "sensing-computation-adjustment". It adopts a portable integrated design and can realize closed-loop control of the entire process of "EEG acquisition-preprocessing-analysis-command output-device linkage-status feedback". At the same time, it is suitable for the use needs of diverse scenarios such as medical research, home rehabilitation, daily monitoring, and human-computer interaction.

[0037] The flexible main frame, as the core supporting body of the device, is made of flexible and elastic materials with biosafety. It is used to support the frontal lobe acquisition array, magnetic modular electrodes and supporting circuit structure, providing a wearable carrier for the device and serving as the hardware foundation for achieving human wearability.

[0038] The prefrontal lobe acquisition array, as the core acquisition unit of EEG signals, is deployed on the inner wearing surface of the flexible main frame. It adopts a multi-point potential distribution design and can simultaneously acquire raw EEG signals from the user's prefrontal lobe region through multiple channels, providing a basic data source for subsequent signal processing and feature analysis.

[0039] The dual-function ear clip module serves as the core of the device's reference acquisition and interference sensing. It integrates a reference electrode and an environmental noise sensor and is fixed to the user's ear when worn. On the one hand, it acquires the EEG reference potential through the reference electrode to provide a potential reference for EEG signal acquisition. On the other hand, it captures environmental electromagnetic interference data in real time through the environmental noise sensor to provide synchronous data source support for subsequent interference elimination.

[0040] The magnetic modular electrode serves as an expandable data acquisition unit for the device. Its hardware interface adopts a magnetic female connector design, enabling quick switching with plug-and-play functionality. It is compatible with both dry electrodes and flexible patch electrodes, allowing for electrode replacement and configuration adjustments without additional tools, thus adapting to the data acquisition needs of different scenarios.

[0041] The main control integrated module, as the core control and computing center of the device, is electrically connected to the prefrontal cortex acquisition array, the dual-function ear clip module, and the magnetic modular electrodes. It can receive the raw data acquired by each hardware module, complete signal processing, feature analysis, logical operations, and instruction output. It realizes four major collaborative control functions: EEG signal acquisition and interference elimination, adaptive adjustment of electrode acquisition parameters, dynamic switching of acquisition channels and sampling rates, and linkage configuration of electrode working modes and acquisition points. It is the core unit for realizing the integrated collaborative operation of software and hardware.

[0042] The specific implementation method of the overall operation of the device is as follows: after the device is worn, it first achieves basic physical contact with the user's forehead through the flexible main frame. The prefrontal lobe acquisition array and magnetic modular electrodes collect the raw EEG signals of the user's prefrontal lobe. The dual-function ear clip module simultaneously collects the EEG reference potential and environmental electromagnetic interference data. All data are transmitted synchronously to the main control integrated module. The main control integrated module first completes the interference removal to obtain clean EEG signals, and then adjusts the acquisition parameters in real time according to the electrode contact state to ensure signal stability. At the same time, according to the current usage scenario and EEG characteristics requirements, it dynamically adjusts the acquisition channel, sampling rate and electrode working mode, ultimately achieving stable, low-power and high-precision EEG signal acquisition and processing in different scenarios.

[0043] Meanwhile, based on the overall technical solution of the device, the structure of the flexible main frame is further optimized and limited. The core technology is to embed a support structure inside the flexible main frame to solve the technical problems of insufficient structural stability and inability to stably adapt to the forehead curves of different users when wearing pure flexible materials, thereby further improving the wearability and structural reliability of the device.

[0044] The flexible main frame is integrally molded from flexible and elastic materials, with a support structure embedded in its pre-set cavity. The support structure is arranged along the length of the flexible main frame, which not only does not hinder the flexible main frame from elastically deforming with the user's forehead curve, but also avoids excessive bending and structural deformation of the flexible main frame, maintaining the overall structural stability of the device. When worn, the flexible main frame, relying on its own flexible and elastic properties, combined with the shaping effect of the internal support structure, adapts to the forehead curve of different users, so that the prefrontal cortex acquisition array arranged on the flexible main frame can form a stable physical fit with the forehead skin of different users, adapting to the wearing needs of users with different head shapes and forehead configurations.

[0045] By further defining the hardware and software collaborative implementation method of adapting the flexible main frame to the user's forehead curve, the core technology is to solve the technical problems of not being able to achieve accurate adaptation by hardware deformation alone and the signal quality instability caused by abnormal electrode contact impedance through the linkage of hardware physical fit and software parameter adaptive adjustment, so as to achieve high-precision wearing adaptation across users.

[0046] First, basic physical adaptation is achieved at the hardware level. The flexible main frame maintains the overall structural stability of the device through internally embedded support structures. At the same time, relying on the deformation characteristics of its own flexible and elastic materials, it adapts to the forehead curve of different users, allowing the prefrontal lobe acquisition array to form a basic physical fit with the user's forehead skin. On this basis, precise adaptation compensation is achieved at the software level. The main control integrated module acquires the contact impedance data between the electrodes and the skin of each acquisition channel in real time through the prefrontal lobe acquisition array. Based on the impedance data, it identifies the user's forehead configuration characteristics and the actual fit of each electrode. For acquisition channels with poor fit or abnormal contact impedance, the corresponding acquisition parameters are adjusted synchronously. Through software parameter compensation combined with the physical deformation fit of the hardware, precise adaptation to the forehead configuration of different users is finally achieved, ensuring that different users can obtain stable EEG signal acquisition quality when wearing the device.

[0047] In some embodiments, the prefrontal cortex acquisition array adopts a six-point potential distribution design.

[0048] Based on the overall technical solution of the device, this embodiment further limits the point design of the prefrontal cortex acquisition array. The core technical content is to adopt a simplified design with a six-point potential distribution to solve the technical problems of excessive power consumption and insufficient portability caused by too many acquisition channels in existing devices. Under the premise of ensuring the acquisition capability of core EEG features, the basic power consumption of the device hardware is reduced.

[0049] The prefrontal cortex acquisition array, deployed on a flexible main frame, employs a six-point potential distribution design. The six acquisition points are arranged along the length of the flexible main frame according to the optimal physiological positions for prefrontal EEG signal acquisition, forming an acquisition array covering the core EEG feature acquisition area of ​​the prefrontal cortex. This simplified six-point design significantly reduces the number of acquisition channels compared to multi-channel full-point acquisition schemes, lowering the device's basic hardware power consumption. At the same time, it can completely acquire the core EEG feature signals of the prefrontal cortex region, balancing acquisition accuracy and low power consumption design requirements, making it suitable for use in portable wearable devices.

[0050] In some embodiments, the dual-function ear clip module synchronously acquires EEG reference potential and environmental electromagnetic interference data, including: while acquiring EEG reference potential, simultaneously acquiring human body environmental induced current and spatial electromagnetic noise data through an induction circuit.

[0051] Based on the overall technical solution of the device, this embodiment further defines the synchronous data acquisition method of the dual-function ear clip module. The core technical content is to clarify the implementation path of the dual-function ear clip module to synchronously acquire reference potential and environmental interference data, solve the technical problem that existing devices cannot synchronously acquire interference data from the same source and have poor interference rejection effect, and provide accurate data source support for subsequent software and hardware collaborative anti-interference.

[0052] After being worn on the user's ear, the dual-function ear clip module's integrated reference electrode maintains stable contact with the skin, continuously acquiring the reference potential corresponding to the EEG signal. This provides a stable potential reference for the EEG signal acquisition of the prefrontal cortex acquisition array. Simultaneously, the dual-function ear clip module works synchronously with the environmental noise sensor through its built-in induction circuit. The induction circuit acquires real-time data on the environmental induced current carried by the user's body, while the environmental noise sensor captures real-time data on the electromagnetic noise in the user's surrounding space. These two sets of data together constitute the environmental electromagnetic interference data. The dual-function ear clip module synchronously transmits the EEG reference potential data and the environmental electromagnetic interference data to the main control integrated module, ensuring the temporal coherence of the two sets of data and providing a synchronous and accurate data source for subsequent differential interference removal.

[0053] In some embodiments, the magnetic female connector can be directly connected to a dry electrode, or connected to a flexible patch electrode via a matching male connector, enabling plug-and-play and quick switching of the electrode.

[0054] Based on the overall technical solution of the device, this embodiment further defines the compatibility switching method of the magnetic modular electrode. The core technical content is to clarify the specific implementation method of the magnetic female buckle interface being compatible with dry electrodes and flexible patch electrodes, so as to solve the technical problem that the electrode type of the existing equipment is fixed and cannot be quickly switched to adapt to different scenarios, and realize the tool-free quick switching and plug-and-play of electrode components.

[0055] The hardware interface of the magnetic modular electrode adopts a standardized magnetic female connector design. The magnetic female connector has built-in magnetic components and conductive contacts, and has two docking modes: The first is a direct dry electrode connection mode, in which the matching dry electrode assembly can be directly connected to the magnetic female connector. The magnetic force of the magnetic component achieves quick fixation between the dry electrode and the female connector, and the conductive contacts complete the circuit connection. The dry electrode can be installed and used without additional tools. The second is a flexible patch electrode docking mode, in which flexible patch electrodes with matching male connectors can be connected to the magnetic female connector through the attraction of the male connectors, achieving quick installation of the flexible patch electrodes and circuit connection. Both modes can achieve plug-and-play and quick switching of electrodes. Users can choose the corresponding electrode type according to the needs of the application scenario and complete the adaptation and adjustment without changing the main body of the device.

[0056] In some embodiments, the main control integrated module receives EEG reference potential data and environmental electromagnetic interference data synchronously collected by the dual-function ear clip module, compares and analyzes the two sets of data through a differential amplification processing algorithm, removes environmental interference noise in the EEG signal, and forms an anti-interference closed loop of hardware acquisition and software processing.

[0057] Based on the overall technical solution of the device, this embodiment further defines the EEG signal acquisition and interference elimination function of the main control integrated module. The core technical content is to clarify the specific implementation method of the software and hardware collaborative anti-interference closed loop, solve the technical problem that the single anti-interference method of the existing device is ineffective and cannot effectively eliminate dynamic interference in complex environments, and greatly improve the anti-interference capability and signal acquisition stability of the device in complex environments.

[0058] The main control integrated module simultaneously receives the raw EEG signals transmitted from the prefrontal cortex acquisition array, as well as the EEG reference potential data and environmental electromagnetic interference data synchronously acquired by the dual-function ear clip module. The main control integrated module first preprocesses the raw EEG signals and reference potential data to obtain the effective EEG signal with interference. Then, using the synchronously acquired environmental electromagnetic interference data as a reference, it compares and analyzes the effective EEG signal with interference and the environmental electromagnetic interference data through a differential amplification processing algorithm. It accurately identifies and removes noise components in the EEG signal that are of the same origin as the environmental interference, and finally obtains a clean effective EEG signal. The whole process forms a hardware acquisition and software processing collaborative anti-interference closed loop of "hardware synchronous acquisition of reference and interference data - software differential processing to remove noise - clean signal output". It can effectively remove power frequency interference and dynamic electromagnetic interference in complex home and mobile scenarios, and ensure the accuracy and stability of EEG signal acquisition.

[0059] In some embodiments, the main control integrated module monitors the contact impedance between the electrodes and the skin of each acquisition channel of the prefrontal lobe acquisition array in real time. Through the forehead configuration recognition algorithm and the electrode contact status monitoring algorithm, it identifies abnormal channels and calculates the corresponding compensation amount. It then automatically adjusts the gain and filtering parameters of the acquisition channels in real time to ensure the stability of the acquisition signal.

[0060] Based on the overall technical solution of the device, this embodiment further defines the adaptive adjustment function of the electrode acquisition parameters of the main control integrated module. The core technical content is to clarify the entire process of parameter adaptive adjustment based on impedance monitoring, solve the technical problems of poor electrode contact in existing devices leading to unstable signal quality and inability to adapt to different users, and ensure the stability of signal acquisition under different wearing conditions.

[0061] During device operation, the main control integrated module monitors the contact impedance data between the electrodes and the skin of each acquisition channel in real time through the prefrontal cortex acquisition array. Simultaneously, the forehead configuration recognition algorithm and the electrode contact status monitoring algorithm built into the main control integrated module run synchronously. Based on the real-time acquired contact impedance data, it identifies the electrode fit status of each acquisition channel, locates abnormal channels with abnormal contact impedance, and calculates the corresponding parameter compensation amount for the abnormal channels. Based on the calculated compensation amount, the main control integrated module automatically adjusts the gain and filtering parameters of the abnormal channels in real time to compensate for signal attenuation and distortion caused by poor electrode fit and excessive contact impedance. The entire process requires no manual intervention, which can ensure the signal acquisition quality of each acquisition channel and ultimately achieve stable acquisition of EEG signals for different users and under different fit conditions.

[0062] In some embodiments, during the non-whole EEG feature extraction stage, the main control integration module executes a point-based dynamic sleep and wake-up algorithm to control the device to prioritize low-power mode with fewer channels and a low sampling rate; when a preset specific EEG feature is detected, all acquisition channels are woken up and switched to a high-precision mode with full channels and a high sampling rate; when the acquisition task is completed or timed out, the corresponding acquisition channel is controlled to sleep and switch back to low-power mode.

[0063] Based on the overall technical solution of the device, this embodiment further defines the dynamic switching function of the acquisition channel and sampling rate of the main control integrated module. The core technical content is to clarify the entire process of dynamic power consumption management based on EEG feature recognition, solve the technical problems of excessive power consumption and insufficient battery life caused by the full-channel and full-time operation of existing devices, and achieve a dynamic balance between acquisition accuracy and device battery life.

[0064] After the device is started, the main control module defaults to the non-full EEG feature extraction stage, executing a point-based dynamic sleep and wake-up algorithm. This controls the device to prioritize low-power operation with fewer channels and a lower sampling rate. In this mode, only some core acquisition channels are activated, continuously monitoring preset specific trigger features in the EEG signal at a lower sampling rate. When the main control module detects the presence of a preset specific EEG feature in the EEG signal, it immediately wakes up all acquisition channels, switching the device to a full-channel, high-sampling-rate, high-precision mode to complete full-dimensional EEG feature acquisition and analysis. When the main control module determines that the high-precision acquisition task is complete, or the high-precision mode has timed out, it immediately controls the non-core acquisition channels to enter sleep mode, switching the device back to low-power mode for continuous operation. Through this dynamic switching mechanism, while ensuring the accuracy of key EEG feature recognition, the overall power consumption of the device is significantly reduced, effectively extending the battery life of the portable device.

[0065] In some embodiments, the main control integrated module has a built-in electrode configuration switching module, which controls the working mode of the magnetic modular electrode and the number of activated acquisition points according to the preset acquisition accuracy requirements and usage scenarios, so as to realize the synchronous matching of hardware electrode configuration and software acquisition logic.

[0066] Based on the overall technical solution of the device, this embodiment further defines the electrode working mode and the linkage configuration function of the acquisition point of the main control integrated module. The core technical content is to clarify the implementation method of scenario-based software and hardware linkage configuration, solve the technical problem that the existing device has fixed acquisition configuration and cannot adapt to the needs of multiple scenarios, and realize the flexible adaptation of the same device to different usage scenarios.

[0067] The main control integrated module has a built-in electrode configuration switching module, allowing users to preset corresponding acquisition configuration schemes based on the acquisition accuracy requirements of the actual usage scenario. When the user switches usage scenarios or selects a corresponding configuration scheme, the electrode configuration switching module generates corresponding hardware control commands based on the preset acquisition accuracy requirements and usage scenario needs. This commands control the working mode of the magnetic modular electrodes and the number of acquisition points activated, allowing for flexible selection of any combination of 1 to 6 acquisition points. Simultaneously, the acquisition logic and signal processing parameters at the software level are adjusted to ensure complete synchronization and matching between the hardware electrode configuration and the software acquisition logic. Through this linkage configuration mechanism, the same device can flexibly switch to adapt to different scenarios. It can enable a full-point, high-precision working mode to meet the high-precision acquisition needs of medical research scenarios, or enable a fewer-point, low-power working mode to meet the long-term use needs of home daily monitoring scenarios, significantly expanding the applicability of the device.

[0068] In some embodiments, the main control integrated module incorporates a hardware and software collaborative anti-interference process that links with the dual-function ear clip module and the prefrontal cortex acquisition array. This process is consistent with the specification.Figure 3 Hardware and software collaborative anti-interference logic framework Figure 1 One-to-one correspondence enables the synchronous acquisition, hierarchical processing, and precise removal of EEG signals and interference noise.

[0069] To address the technical problem of existing brain-computer interface terminal devices where hardware acquisition and software interference suppression are disconnected, making it impossible to achieve synchronous interference acquisition and cancellation, resulting in poor noise removal and severe loss of effective EEG signals in complex environments; this embodiment is based on Figure 3 The closed-loop logic constructs a full-process anti-interference closed loop of "hardware synchronous dual-channel acquisition - software hierarchical parallel processing - adaptive noise precise cancellation", which greatly improves the anti-interference capability and signal acquisition stability of the equipment in complex electromagnetic environments.

[0070] During device operation, synchronous data acquisition is first completed through the hardware acquisition layer. The prefrontal cortex acquisition array continuously acquires raw EEG signals from the user's prefrontal cortex region, while the dual-function ear clip module simultaneously acquires EEG baseline potential data and environmental electromagnetic interference data. After the two data streams undergo synchronous analog-to-digital conversion, they are transmitted to the main control integration module in the form of synchronous digital data streams. The main control integration module temporarily stores the two synchronously transmitted digital data streams through its built-in raw data receiving buffer to ensure data timing consistency and avoid noise cancellation failure caused by timing misalignment. Subsequently, the main control integration module processes the two data streams in parallel from the buffer. One stream undergoes baseline correction and other processing through the EEG signal preprocessing module. The preprocessing of the frequency-based filtering yields the effective EEG signal with interference. Another path passes through a noise channel extraction module to extract features from the environmental electromagnetic interference data, obtaining noise feature data synchronized with the original EEG signal. Then, the main control integration module uses a built-in adaptive noise cancellation algorithm to compare and analyze the effective EEG signal with interference and the synchronously extracted noise feature data, accurately separating and removing environmental interference noise components from the EEG signal while fully preserving the physiological characteristics of the effective EEG signal. Finally, the clean EEG signal obtained after noise cancellation processing is transmitted through the clean data output interface of the main control integration module to the subsequent EEG feature analysis and command generation module, completing the entire anti-interference processing closed loop.

[0071] In some embodiments, the main control integrated module incorporates an electrode contact impedance adaptive monitoring and parameter closed-loop adjustment process, which is consistent with... Figure 4 Adaptive impedance monitoring and parameter adjustment process Figure 1 With one-to-one correspondence, the entire cycle of electrode contact status monitoring and dynamic closed-loop adjustment of acquisition parameters can be achieved.

[0072] To address the technical problem that existing brain-computer interface terminal devices cannot perform periodic closed-loop monitoring of electrode skin contact impedance and can only adjust the acquisition parameters manually, thus failing to compensate for signal quality degradation caused by impedance anomalies in real time; this embodiment is based on Figure 4 The closed-loop logic of the entire process enables fully automatic operation of impedance monitoring, anomaly detection, parameter adjustment, and status feedback, ensuring the stability of signal acquisition under different wearing conditions without manual intervention.

[0073] After the device enters the EEG signal acquisition state, the main control integrated module automatically initiates the electrode contact impedance monitoring process according to a preset fixed cycle. Within each monitoring cycle, the main control integrated module measures the contact impedance between the electrodes and the human skin of all acquisition channels through the prefrontal cortex acquisition array, obtaining the real-time impedance value of each channel. Subsequently, the main control integrated module compares the real-time impedance value of each channel with a preset normal impedance threshold range to determine whether the impedance of all channels is within the acceptable range. If the impedance of all channels is deemed acceptable, the main control integrated module maintains the current acquisition parameters of each channel unchanged, continues to perform EEG signal acquisition, and waits for the next monitoring cycle to arrive, repeating the impedance monitoring process. If an abnormal channel with impedance exceeding the acceptable range is detected, the main control integrated module first accurately locates the abnormal channel, identifies the degree of impedance deviation, and calculates the corresponding acquisition parameter compensation amount based on the quality of the currently acquired EEG signal. Then, based on the calculated compensation amount, it automatically adjusts the gain and filter parameters of the programmable gain amplifier of the abnormal channel in real time to compensate for signal attenuation and distortion caused by impedance abnormality. After the parameter adjustment is completed, the main control integrated module automatically records the abnormal channel information, impedance value, parameter adjustment content, and adjustment time for this monitoring, forming a complete adjustment log. After completing the entire process of this monitoring cycle, it waits for the next monitoring cycle to arrive and repeats the above closed-loop process.

[0074] In some embodiments, the main control integration module incorporates a dynamic power consumption management process for the EEG acquisition channel and operating mode, which is consistent with... Figure 5 Dynamic power management algorithm flow Figure 1 One-to-one correspondence enables intelligent dynamic switching between low-power mode and high-precision mode of the device.

[0075] To address the technical problem of existing brain-computer interface (BCI) terminal devices employing a fixed, all-channel, all-time operating mode, which fails to dynamically adjust their operating status according to EEG acquisition needs, resulting in excessively high overall power consumption and insufficient battery life for long-term wearable use; this embodiment is based on the appendix Figure 5 The state switching logic significantly reduces device power consumption while ensuring the accuracy of key EEG feature recognition, achieving a dynamic balance between acquisition accuracy and battery life.

[0076] After the device is powered on and enters EEG monitoring mode, the main control integrated module defaults to a low-power operation mode. In this mode, only a small number of core acquisition channels are activated, running continuously at a low sampling rate, monitoring only preset specific EEG trigger features, significantly reducing the device's basic power consumption. During low-power mode operation, the main control integrated module continuously analyzes the EEG signals acquired by the activated core channels in real time to determine whether the preset specific EEG trigger features are detected. If no preset specific EEG trigger features are detected, the main control integrated module maintains low-power mode and continues to perform real-time monitoring of trigger features. If a preset specific EEG trigger feature is detected, the main control integrated module immediately wakes up all dormant acquisition channels and hardware acquisition modules, switching the device from low-power mode to high-precision operation. In this high-precision mode, all acquisition channels are activated, operating at a high sampling rate to perform full-dimensional EEG feature acquisition and analysis. During high-precision mode operation, the main control integrated module continuously determines whether the current high-precision acquisition task is completed or whether the high-precision mode runtime exceeds the preset timeout threshold. If it is determined that the acquisition task is not completed and the timeout threshold is not exceeded, the main control integrated module maintains the high-precision mode to continue running, completing the high-precision acquisition and processing of EEG signals. If it is determined that the acquisition task is completed or the runtime exceeds the preset timeout threshold, the main control integrated module immediately controls the non-core acquisition channels and hardware modules to enter a sleep state, switching the device from high-precision mode back to low-power mode, continuing to perform monitoring of specific EEG trigger features, completing a complete mode switching closed loop.

[0077] In some embodiments, the main control integration module has a built-in intelligent motion artifact recognition and removal algorithm module, which can realize intelligent recognition and accurate removal of human motion artifacts based on the synchronous acquisition data of the prefrontal cortex acquisition array and the dual-function ear clip module.

[0078] To address the technical challenge that existing brain-computer interface (BCI) terminal devices can only handle environmental electromagnetic interference and cannot effectively eliminate electromyographic (EMG) artifacts and motion artifacts caused by users' own physical movements such as blinking, frowning, and head shaking; especially in mobile and home uncontrolled scenarios, where motion artifacts are the core factor causing EEG signal distortion, and existing technologies cannot achieve precise separation of artifacts from valid EEG signals, this embodiment achieves adaptive removal of motion artifacts through intelligent feature recognition of multi-channel synchronous data, further improving the signal acquisition quality in complex dynamic scenarios.

[0079] During operation, the prefrontal cortex acquisition array continuously collects multi-channel raw EEG signals, while the dual-function ear clip module simultaneously collects electromyographic interference signals from the ear and potential fluctuation data caused by human movement. Both sets of data are transmitted synchronously to the main control integration module. The main control integration module's built-in intelligent motion artifact recognition algorithm first extracts features from the synchronously collected multi-channel data, distinguishing between the characteristic frequency bands of valid EEG signals and the characteristic frequency bands of motion artifacts. Simultaneously, through a pre-trained deep learning model, it identifies the characteristic waveforms of typical motion artifacts such as blinking, frowning, and head shaking. When motion artifacts are identified, the algorithm module accurately separates and removes the artifact components in the EEG signal based on the artifact reference data from the synchronously collected ear clip module, while preserving the complete features of the valid EEG signal. The entire process requires no manual intervention and can realize real-time processing of motion artifacts in dynamic scenarios, significantly improving the acquisition accuracy of EEG signals in uncontrolled environments.

[0080] In some embodiments, the main control integration module has a built-in intelligent compensation algorithm module for long-term acquisition signal drift, which can realize dynamic identification and adaptive compensation of EEG signal baseline drift during long-term wear.

[0081] To address the technical challenges of baseline drift and continuous signal quality degradation in existing brain-computer interface (BCI) terminal devices during prolonged continuous wear, caused by factors such as drying of the electrode conductive medium, changes in user skin condition, and slight displacement of the wearing position, existing technologies can only perform baseline correction with fixed parameters and cannot adapt to dynamic drift changes during long-term wear. This embodiment utilizes a time-series prediction intelligent algorithm to predict drift trends and dynamically compensate for them, ensuring signal stability during long-term continuous operation of the device.

[0082] After the device enters the long-term continuous acquisition mode, the main control integrated module collects and records the baseline data of EEG signals from each channel and the electrode-skin contact impedance data according to a preset cycle. At the same time, through the built-in intelligent signal drift compensation algorithm module, a dynamic change model of the signal baseline of each channel is constructed based on the historical time-series data to predict the drift trend and amount of the signal baseline. When the algorithm module detects that the signal baseline drift exceeds the preset threshold or that the electrode contact impedance shows a continuous gradual change, it automatically adjusts the DC offset compensation parameters, baseline correction parameters, and acquisition gain of the corresponding channel to offset the signal drift caused by long-term wear. Meanwhile, the algorithm module can continuously optimize the dynamic change model based on the continuously acquired data to adapt to the drift characteristics of different users and different wearing times, ensuring the stability and consistency of EEG signal acquisition when the device is worn continuously for several hours or even longer.

[0083] In some embodiments, the main control integration module has a built-in personalized EEG feature adaptive learning algorithm module, which can realize online incremental learning and personalized adaptation of EEG feature models for different users.

[0084] To address the technical challenges of low accuracy and poor generalization ability in existing brain-computer interface terminal devices that use a general EEG feature recognition model due to significant individual differences in EEG physiological characteristics, head structure, and signal characteristics among different users, and the fact that existing technologies require cumbersome offline calibration by professionals and cannot meet the needs of ordinary home users, this embodiment utilizes an online incremental learning algorithm to automatically construct and continuously optimize personalized user models, significantly improving the accuracy and user adaptability of EEG decoding.

[0085] When a user uses the device for the first time, the main control integration module collects the user's basic EEG feature data through a preset guided calibration process, constructing an initial personalized EEG feature model for the user. During subsequent daily use, the main control integration module continuously collects the user's EEG signal data, corresponding scene labels, and command feedback data. Through the built-in personalized EEG feature adaptive learning algorithm module, it performs online incremental learning to continuously optimize the user's personalized feature model and adapt to the user's EEG signal characteristics that change over time. At the same time, the algorithm module can distinguish between the user's valid EEG features and interference signals, automatically selecting high-quality learning samples and completing iterative optimization of the model without manual annotation. For different usage scenarios, the algorithm module can also construct scenario-specific personalized sub-models to adapt to the EEG decoding needs of different scenarios such as medical rehabilitation, daily monitoring, and human-computer interaction, significantly improving the accuracy of EEG intent recognition for different users and in different scenarios.

[0086] In some embodiments, the main control integrated module has a built-in intelligent recognition and self-correction algorithm module for wearing status, which can identify the wearing status of the device based on multi-channel impedance data, and realize self-correction of the collected parameters and intelligent early warning of wearing abnormalities.

[0087] To address the technical problem that existing brain-computer interface terminal devices cannot promptly identify and correct when users experience misalignment, loosening, or poor fit on one side, leading to a sharp drop in signal quality and preventing users from promptly noticing and adjusting the device; existing technologies can only monitor single-channel impedance anomalies and cannot identify the overall wearing misalignment state or achieve targeted self-correction, this embodiment achieves accurate judgment of wearing status, parameter self-correction, and anomaly warning through intelligent identification of multi-channel impedance distribution, thereby improving the ease of use and acquisition stability of the device.

[0088] After the device is worn, the main control integrated module simultaneously collects the contact impedance data between the electrodes and the skin at each point through multiple acquisition channels of the frontal lobe acquisition array, constructing an impedance distribution matrix for all channels. The built-in intelligent wearing status recognition and self-correction algorithm module, based on this impedance distribution matrix and combined with a pre-trained wearing status recognition model, accurately judges the wearing status of the device, including various states such as normal fit, overall wear offset, unilateral looseness, and local poor fit. When a slight abnormality such as local poor fit is detected, the algorithm module automatically calculates the parameter compensation amount of the corresponding channel, adjusts the acquisition gain, sensitivity, and filtering parameters of the abnormal channel, completes the self-correction of the acquisition parameters, and compensates for the signal attenuation caused by poor fit. When abnormal states such as overall wear offset or severe looseness that cannot be corrected by parameters are detected, the main control integrated module sends a wearing abnormality warning prompt to the user's terminal through the wireless communication unit, guiding the user to adjust the wearing position, and at the same time pausing the high-precision acquisition mode and switching to the low-power monitoring mode to avoid invalid data acquisition and power waste.

[0089] In some embodiments, the main control integration module has a built-in intelligent usage scenario recognition and automatic adaptation algorithm module to automatically identify the user's current usage scenario based on the collected data and automatically switch the configuration of the brain-computer interface terminal device. During the operation of the brain-computer interface terminal device, the main control integration module continuously collects multi-dimensional data, including environmental electromagnetic interference data acquired by the dual-function ear clip module, EEG signal feature data acquired by the prefrontal cortex acquisition array, operating status data of the brain-computer interface terminal device, and user operation habit data. Through the scene intelligent recognition and automatic adaptation algorithm module, based on the EEG signal feature data, device operating status data, and user operation habit data, and through a pre-trained scene classification model, the module identifies the user's current usage scenario, including high-precision acquisition scenarios for medical research, long-term monitoring scenarios for home rehabilitation, real-time human-computer interaction control scenarios, and daily physiological state monitoring scenarios. After identifying the corresponding usage scenario, the module matches the optimal device configuration scheme for the corresponding usage scenario, and adjusts the working mode and number of activated acquisition points of the magnetic modular electrodes, the sampling rate and working mode of the acquisition channel, the parameter configuration of the signal processing algorithm, and the power consumption management mode of the device, while simultaneously adjusting the acquisition logic and feature parsing strategy at the software level.

[0090] To address the technical issues of existing brain-computer interface terminal devices requiring users to manually switch scene configuration modes, which is cumbersome and prevents users from accurately matching optimal configuration parameters, thus hindering the device's performance; and because existing technologies only support preset fixed mode switching and cannot automatically adapt to the optimal configuration based on the user's actual usage scenario, this embodiment uses a multi-dimensional data scene classification intelligent algorithm to achieve automatic identification of usage scenarios and fully automatic linkage adaptation of device hardware and software configurations, thereby improving the device's ease of use and scene adaptability.

[0091] During device operation, the main control integration module continuously collects multi-dimensional data, including environmental electromagnetic interference data acquired by the dual-function ear clip module, EEG signal feature data acquired by the prefrontal cortex acquisition array, device operating status data, and user operation habit data. The built-in intelligent recognition and automatic adaptation algorithm module for usage scenarios, based on the above multi-dimensional data, automatically identifies the user's current usage scenario through a pre-trained scenario classification model, including high-precision acquisition scenarios for medical research, long-term monitoring scenarios for home rehabilitation, real-time human-computer interaction control scenarios, and daily physiological state monitoring scenarios. After identifying the corresponding usage scenario, the algorithm module automatically matches the optimal device configuration scheme for that scenario, and adjusts the working mode and number of activated acquisition points of the magnetic modular electrodes, the sampling rate and working mode of the acquisition channel, the parameter configuration of the signal processing algorithm, and the device's power consumption management mode in a coordinated manner. At the same time, it simultaneously adjusts the acquisition logic and feature parsing strategy at the software level, realizing fully automatic synchronous adaptation of hardware configuration and software logic. Without manual operation by the user, the device can be in the optimal working state in different scenarios.

[0092] In some embodiments, the main control integration module incorporates an EEG intention intelligent prediction and closed-loop pre-linkage algorithm module to predict the user's operational intention based on EEG preparatory potential characteristics, thereby achieving pre-linkage and low-latency closed-loop control of external devices. When the brain-computer interface terminal device is in human-computer interaction or rehabilitation training mode, the main control integration module continuously collects the user's EEG signals at a high sampling rate through a prefrontal cortex acquisition array, capturing EEG preparatory potential characteristics related to the user's action intention and operational decision. Based on the real-time collected EEG signals, the EEG intention intelligent prediction and closed-loop pre-linkage algorithm module identifies characteristic changes in EEG preparatory potentials through a pre-trained temporal prediction model, predicting the user's upcoming operational intention and corresponding control command. After predicting the user's operational intention, the main control integration module sends a pre-linkage command to the corresponding external device through a wireless communication unit, controlling the external device to enter a preparatory running state. After completing the full EEG intention recognition and confirmation, an execution command is sent to control the external device to complete the corresponding action.

[0093] To address the significant latency issues in existing brain-computer interface (BCI) terminal devices, which require complete recognition of EEG intentions before outputting control commands and thus fail to meet the low-latency requirements of real-time human-computer interaction and rehabilitation training, this embodiment utilizes a timing prediction intelligent algorithm to achieve advance prediction of user operation intentions and pre-linkage with external devices. This significantly reduces the latency of BCI interaction and improves the smoothness and real-time performance of closed-loop control.

[0094] When the device is in human-computer interaction or rehabilitation training mode, the main control integration module continuously collects the user's EEG signals at a high sampling rate through the prefrontal cortex acquisition array, focusing on capturing the EEG preparatory potential characteristics related to the user's action intentions and operational decisions. The built-in EEG intention intelligent prediction and closed-loop pre-linkage algorithm module, based on the real-time collected EEG signals, identifies the characteristic changes of EEG preparatory potentials through a pre-trained temporal prediction model, and predicts in advance the user's upcoming operation intention and corresponding control command without waiting for the complete intention action to be completed. After predicting the user's operation intention, the main control integration module sends a pre-linkage command to the corresponding external device in advance through the wireless communication unit, controlling the external device to enter the preparatory operation state in advance. After the complete EEG intention recognition and confirmation is completed, the execution command is immediately sent to control the external device to complete the corresponding action. At the same time, the algorithm module can continuously optimize the prediction model based on the user's historical operation data, improve the accuracy of intention prediction, reduce the probability of false triggering, and ultimately significantly reduce the control latency of brain-computer interaction, achieving highly smooth brain-computer closed-loop control.

[0095] It should be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. It should be understood that when an element or layer is referred to as “on,” “adjacent to,” “connected to,” or “coupled to” other elements or layers, it may be directly on, adjacent to, connected to, or coupled to other elements or layers, or there may be intervening elements or layers. Conversely, when an element is referred to as “directly on,” “directly adjacent to,” “directly connected to,” or “directly coupled to” other elements or layers, there are no intervening elements or layers. It should be understood that although the terms first, second, third, etc., may be used to describe various elements, components, areas, layers, and / or portions, these elements, components, areas, layers, and / or portions should not be limited by these terms. These terms are merely used to distinguish one element, component, area, layer, or portion from another element, component, area, layer, or portion. Therefore, without departing from the teachings of this application, the first element, component, area, layer, or portion discussed below may be referred to as a second element, component, area, layer, or portion.

[0096] Spatial relation terms such as “below,” “under,” “below,” “under,” “above,” “above,” etc., are used herein for convenience of description to describe the relationship between one element or feature shown in the figure and other elements or features. It should be understood that, in addition to the orientation shown in the figure, spatial relation terms are intended to also include different orientations of the device in use and operation. For example, if the device in the figure is flipped, then the element or feature described as “below,” “under,” or “below” other elements or features will be oriented “above” other elements or features. Therefore, the exemplary terms “below” and “under” can include both above and below orientations. The device may be otherwise oriented (rotated 90 degrees or otherwise) and the spatial descriptive terms used herein will be interpreted accordingly.

[0097] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. When used herein, the singular forms “a,” “an,” and “the” are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the terms “comprising” and / or “including,” when used in this specification, identify the presence of the stated features, integers, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups. When used herein, the term “and / or” includes any and all combinations of the associated listed items.

[0098] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0099] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A brain-computer interface terminal device, comprising a flexible main frame, a prefrontal cortex acquisition array, a dual-function ear clip module, magnetic modular electrodes, and a main control integrated module, characterized in that, The flexible main frame is made of flexible and elastic materials; The prefrontal lobe acquisition array is deployed on a flexible main frame and adopts a multi-point potential distribution design. The dual-function ear clip module integrates a reference electrode and an environmental noise sensor for synchronously acquiring EEG reference potential and environmental electromagnetic interference data. The magnetic modular electrode adopts a magnetic female buckle interface design to be compatible with both dry electrodes and flexible patch electrodes for plug-and-play use. The main control integrated module is electrically connected to the frontal lobe acquisition array, the dual-function ear clip module and the magnetic modular electrode respectively. The flexible main frame maintains the overall structural stability of the device through the internally embedded support structure, and combines flexible and elastic materials to adapt to the forehead curve of different users, so that the frontal lobe acquisition array forms a basic physical fit with the user's forehead skin. The main control integration module acquires the contact impedance data between the electrodes and the skin of each acquisition channel in real time through the frontal lobe acquisition array. Based on the impedance data, it identifies the user's forehead shape and the electrode fit status, and synchronously adjusts the acquisition parameters of the corresponding acquisition channel. Together with the flexible main frame, it completes the precise adaptation to different users' forehead shapes.

2. The brain-computer interface terminal device according to claim 1, characterized in that, The prefrontal cortex acquisition array adopts a six-point potential distribution design.

3. The brain-computer interface terminal device according to claim 1, characterized in that, The dual-function ear clip module synchronously acquires EEG reference potential and environmental electromagnetic interference data, including: While collecting the EEG reference potential, the system simultaneously acquires data on the human body's environmental induced current and spatial electromagnetic noise through an induction circuit.

4. The brain-computer interface terminal device according to claim 1, characterized in that, The magnetic female connector can be directly connected to a dry electrode, or connected to a flexible patch electrode via a matching male connector, enabling plug-and-play and quick switching of electrodes.

5. The brain-computer interface terminal device according to claim 1, characterized in that, The main control integrated module receives EEG reference potential data and environmental electromagnetic interference data synchronously collected by the dual-function ear clip module. It compares and analyzes the two sets of data through differential amplification processing algorithm to remove environmental interference noise in the EEG signal, forming an anti-interference closed loop of hardware acquisition and software processing.

6. The brain-computer interface terminal device according to claim 1, characterized in that, The main control integrated module monitors the contact impedance between the electrodes and the skin of each acquisition channel of the frontal lobe acquisition array in real time. Through the forehead configuration recognition algorithm and the electrode contact status monitoring algorithm, it identifies abnormal channels and calculates the corresponding compensation amount. It automatically adjusts the gain and filtering parameters of the acquisition channels in real time to ensure the stability of the acquisition signal.

7. The brain-computer interface terminal device according to claim 1, characterized in that, During the non-whole EEG feature extraction stage, the main control integrated module executes a point-based dynamic sleep and wake-up algorithm to control the device to prioritize low-power mode with fewer channels and lower sampling rate. When a preset specific EEG feature is detected, all acquisition channels are woken up and switched to a high-precision mode with full channels and high sampling rate. When the acquisition task is completed or timed out, the corresponding acquisition channel is controlled to sleep and switch back to low-power mode.

8. The brain-computer interface terminal device according to claim 1, characterized in that, The main control integrated module has a built-in electrode configuration switching module. Based on the preset acquisition accuracy requirements and usage scenarios, it controls the working mode of the magnetic modular electrode and the number of acquisition points activated, so as to achieve synchronous matching between hardware electrode configuration and software acquisition logic.

9. The brain-computer interface terminal device according to claim 1, characterized in that, The main control integration module has a built-in intelligent recognition and automatic adaptation algorithm module for usage scenarios, which automatically identifies the user's current usage scenario based on the collected data and completes the fully automatic switching of the brain-computer interface terminal device configuration. During the operation of the brain-computer interface terminal device, the main control integration module continuously collects multi-dimensional data, including environmental electromagnetic interference data acquired by the dual-function ear clip module, EEG signal feature data acquired by the prefrontal cortex acquisition array, operating status data of the brain-computer interface terminal device, and user operation habit data. Through the scene intelligent recognition and automatic adaptation algorithm module, based on EEG signal feature data, device operating status data, and user operation habit data, and using a pre-trained scene classification model, it identifies the user's current usage scenario, including high-precision acquisition scenarios for medical research, long-term monitoring scenarios for home rehabilitation, real-time human-computer interaction control scenarios, and daily physiological state monitoring scenarios. After identifying the corresponding usage scenario, it matches the optimal device configuration scheme for the corresponding usage scenario, and adjusts the working mode and number of activated acquisition points of the magnetic modular electrodes, the sampling rate and working mode of the acquisition channel, the parameter configuration of the signal processing algorithm, and the power consumption management mode of the device in a coordinated manner, while simultaneously adjusting the acquisition logic and feature parsing strategy at the software level.

10. The brain-computer interface terminal device according to claim 1, characterized in that, The main control integration module has a built-in EEG intention intelligent prediction and closed-loop pre-linkage algorithm module, which predicts the user's operation intention based on the EEG pre-potential characteristics to realize the pre-linkage and low-latency closed-loop control of external devices. When the brain-computer interface terminal device is in human-computer interaction or rehabilitation training mode, the main control integration module continuously collects the user's EEG signals at a high sampling rate through the prefrontal cortex acquisition array, capturing the EEG preparatory potential characteristics related to the user's action intentions and operational decisions. The EEG intention intelligent prediction and closed-loop pre-linkage algorithm module is based on real-time acquired EEG signals. Through a pre-trained time-series prediction model, it identifies characteristic changes in the EEG preparatory potential and predicts the user's upcoming operation intention and corresponding control command. After predicting the user's operation intention, the main control integration module sends a pre-linkage command to the corresponding external device through the wireless communication unit, controlling the external device to enter the preparatory operation state. After the complete EEG intention recognition and confirmation is completed, an execution command is sent to control the external device to complete the corresponding action.

Images