Closed-loop nerve regulation method and device, medium and program product

Abstract

The invention provides a closed-loop nerve regulation method and device, a medium and a program product, and relates to the technical field of nerve regulation. The method comprises the steps that an electroencephalogram collection module collects electroencephalogram signals in real time and provides real-time feedback data for nerve regulation and control; the magnetic resonance imaging module provides high-precision brain structure and function positioning information; the three different stimulation modules (transcranial alternating current stimulation, low-field magnetic field stimulation and transcranial magnetic stimulation) can stimulate brain regions with different depths and ranges; the multi-mode navigation module carries out accurate positioning based on the magnetic resonance image; the electroencephalogram detection module analyzes the collected signals; and the central control module coordinates the work of each module. According to the multi-module cooperative working mode, accurate stimulation of different areas and different depths of the brain is achieved, meanwhile, the stimulation effect is ensured through real-time monitoring, and therefore the accuracy and effectiveness of nerve regulation and control are improved.

Description

Technical Field

[0001] This application relates to the field of neuromodulation technology, and in particular, to a closed-loop neuromodulation method, device, medium, and program product. Background Art

[0002] With the development of neuroscience and medical technology, neuromodulation technology plays an increasingly important role in the treatment of nervous system diseases. Neuromodulation regulates the activities of neurons by stimulating specific regions of the brain and has become an important means for treating various nervous system diseases such as depression and Parkinson's disease.

[0003] Currently, the commonly used neuromodulation devices in clinical practice mainly adopt a single stimulation method. For example, transcranial magnetic stimulation devices stimulate the cerebral cortex by generating pulsed magnetic fields, or transcranial direct current stimulation devices regulate the excitability of neurons through continuous current stimulation. When using these devices, doctors need to judge the stimulation position and parameters based on experience and evaluate the treatment effect by observing the clinical symptoms of patients.

[0004] However, in practical applications, the brain structure is complex and individual differences are significant. Due to the lack of real-time brain activity monitoring and precise navigation and positioning systems, it is difficult for doctors to accurately judge whether the stimulation acts on the target area, and thus it is difficult to dynamically adjust the stimulation parameters according to the real-time brain activity state of patients, resulting in unstable treatment effects. Summary of the Invention

[0005] This application provides a closed-loop neuromodulation method, device, medium, and program product, which are used to solve the problem in existing neuromodulation technologies that due to the lack of real-time brain activity monitoring and precise navigation and positioning systems, it is difficult to dynamically adjust the stimulation parameters according to the individual real-time brain activity, resulting in unstable treatment effects.

[0006] In a first aspect, this application provides a closed-loop neuromodulation device, including: an electroencephalogram acquisition module for collecting in real time the electrical signals generated by the activities of brain neurons through a plurality of electrodes attached to the scalp of a user; a magnetic resonance imaging module for obtaining high-resolution structural and functional images of the brain; a transcranial alternating current stimulation module for delivering alternating current with a set frequency and intensity to a set area of the brain through electrodes placed on the scalp; a low-field magnetic stimulation module for generating a low-intensity magnetic field acting on the brain; a transcranial magnetic stimulation module for generating a rapidly changing magnetic field to induce an induced current in brain tissue; a multi-modal navigation module for guiding the stimulation module to accurately apply the stimulation to a set target point according to the structural and functional characteristics of the brain; an electroencephalogram detection module for deeply analyzing and processing the original electroencephalogram signals acquired by the electroencephalogram acquisition module; and a central control module for being responsible for the coordinated management of each module.

[0007] By adopting the above technical solution, the electroencephalogram (EEG) acquisition module can acquire EEG signals in real time, providing real-time feedback data for neuromodulation; the magnetic resonance imaging (MRI) module provides high-precision brain structure and function localization information; three different stimulation modules (transcranial alternating current stimulation, low-field magnetic stimulation, and transcranial magnetic stimulation) can perform stimulation on brain regions with different depths and ranges; the multi-modal navigation module performs precise positioning based on magnetic resonance images; the EEG detection module analyzes the acquired signals; and the central control module coordinates the work of each module. This way of multi-module collaborative work realizes precise stimulation of different regions and depths of the brain, and at the same time ensures the stimulation effect through real-time monitoring, thereby improving the accuracy and effectiveness of neuromodulation.

[0008] Combined with some embodiments of the first aspect, in some embodiments, it further includes: a multi-modal power supply module for supplying stable power to each module of the closed-loop neuromodulation device.

[0009] By adopting the above technical solution, the multi-modal power supply module can provide corresponding power supply solutions according to the power demand characteristics of different modules. This dedicated power supply design ensures that each module can obtain stable and reliable power support during operation, avoiding abnormal device functions or treatment interruptions caused by unstable power supply.

[0010] Combined with some embodiments of the first aspect, in some embodiments, a closed-loop neuromodulation method includes: obtaining multiple EEG signal information through the EEG acquisition module and determining real-time brain activity state information in combination with the EEG detection module; obtaining brain structure and function imaging data through the MRI module; combining the brain activity state information and a preset neuromodulation strategy, and determining the stimulation parameter information of the transcranial alternating current stimulation module, low-field magnetic stimulation module, and transcranial magnetic stimulation module through a neuromodulation model, where the neuromodulation model is pre-constructed through deep learning from a set of brain activity state information annotated with the stimulation parameter information of the transcranial alternating current stimulation module, low-field magnetic stimulation module, and transcranial magnetic stimulation module; combining the preset neuromodulation target, brain structure and function imaging data, and determining the target position information to be stimulated through the multi-modal navigation module; and controlling the transcranial alternating current stimulation module, low-field magnetic stimulation module, and transcranial magnetic stimulation module to perform stimulation according to the preset neuromodulation strategy in combination with the target position information.

[0011] By adopting the above technical solutions, the electroencephalogram acquisition module and the detection module can obtain and analyze the brain state in real time; magnetic resonance imaging provides accurate anatomical and functional information; the neuromodulation model, based on deep learning methods, establishes an association between the stimulation parameters in historical data and the brain state, realizing the intelligent selection of parameters; the multimodal navigation module combines imaging data to ensure the accuracy of the stimulation position; the three stimulation modules perform collaborative stimulation according to the model output. This data-driven closed-loop regulation method realizes the personalization and precision of neuromodulation through the organic combination of real-time monitoring, intelligent analysis, and precise stimulation.

[0012] Combined with some embodiments of the first aspect, in some embodiments, after the step of determining the target position information to be stimulated through the multimodal navigation module by combining the preset neuromodulation target, brain structure, and functional imaging data, it further includes: performing a safety assessment on the target position information according to the preset safety standard to determine the safety coefficient; if the safety coefficient is greater than the set safety coefficient threshold, sending the target position information to the transcranial alternating current stimulation module, the low-field magnetic stimulation module, and the transcranial magnetic stimulation module.

[0013] By adopting the above technical solutions, after determining the target position, a safety assessment is first performed, the safety coefficient is calculated and compared with the preset threshold. Only when the safety coefficient meets the standard will the stimulation operation be executed. This safety assessment mechanism takes into account the particularity of the stimulation position, embeds the safety check into the treatment process, and objectively evaluates through the quantified safety coefficient. By adding this safety control link, the safety of the neuromodulation process is significantly improved, providing a more reliable guarantee for patients.

[0014] Combined with some embodiments of the first aspect, in some embodiments, after the step of controlling the transcranial alternating current stimulation module, the low-field magnetic stimulation module, and the transcranial magnetic stimulation module to perform stimulation according to the preset neuromodulation strategy by combining the target position information, it further includes: obtaining multiple activity state information of the brain in real time through the electroencephalogram detection module; combining the multiple activity state information, determining the stimulation effect score through the brain stimulation effect evaluation model, and sending it to the user terminal. The brain stimulation effect evaluation model is constructed in advance through deep learning based on multiple sets of brain activity state information marked with stimulation effect scores.

[0015] By adopting the above technical solutions, the brain activity state is monitored in real time, and the effect score is obtained through the evaluation model constructed by deep learning. The evaluation model is trained based on a large amount of labeled data and can accurately evaluate the stimulation effect. By real-time feedback of the evaluation results to the user terminal, doctors can timely understand the treatment progress. This real-time evaluation mechanism based on deep learning provides an objective and quantified method for evaluating the treatment effect, effectively improving the precision and controllability of neuromodulation.

[0016] In some embodiments in combination with some embodiments of the first aspect, after the step of determining a stimulation effect score through a brain stimulation effect evaluation model by combining the plurality of activity state information and sending it to the user terminal, the method further includes: if the stimulation effect score is lower than a set effect score threshold, sending a reminder to the user terminal.

[0017] By adopting the above technical solution, the device can automatically monitor the stimulation effect score and actively send a reminder when the score is lower than the threshold. This automatic early warning mechanism can detect the poor treatment effect in a timely manner without continuous manual monitoring. Through the timely reminder function, medical staff can quickly intervene to adjust the treatment plan, avoid the continuous progress of inefficient treatment, and improve the timeliness and effectiveness of treatment.

[0018] In some embodiments in combination with some embodiments of the first aspect, after the step of determining a stimulation effect score through a brain stimulation effect evaluation model by combining the plurality of activity state information, the method further includes: obtaining stimulation effect data of different functional regions of the brain, where the stimulation effect data at least includes the improvement of activity, the change in the frequency stability of electroencephalogram signals before and after stimulation, and the matching degree with a preset ideal electroencephalogram pattern; presenting the stimulation effect data to the user terminal in a visual form.

[0019] By adopting the above technical solution, the device comprehensively collects the stimulation effect data of different brain regions, including multiple dimensions such as the improvement of activity, the change in frequency stability, and the matching degree with the ideal pattern. These data are presented in a visual way, enabling doctors to intuitively understand the treatment effects of each brain region.

[0020] In a second aspect, the present application provides a closed-loop neuromodulation device, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, and the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the closed-loop neuromodulation device to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0021] In a third aspect, the present application provides a computer-readable storage medium, including instructions, which when running on a closed-loop neuromodulation device, enable the closed-loop neuromodulation device to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0022] In a fourth aspect, the present application provides a computer program product, which when running on a closed-loop neuromodulation device, enables the closed-loop neuromodulation device to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0023] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. By adopting the above technical solution, due to the adoption of a technical architecture of multi-module collaborative work, that is, real-time brain activity monitoring is realized through an electroencephalogram acquisition module and an electroencephalogram detection module, precise positioning is realized through magnetic resonance imaging and a multi-modal navigation module, and multi-modal collaborative stimulation is realized through three different types of stimulation modules under the overall coordination of a central control module, so the technical problem that it is difficult to accurately position and stimulate the target neuron group only relying on experience in the prior art is effectively solved. Furthermore, precise stimulation of different regions and depths of the brain is realized, and at the same time, the targeting and effectiveness of the stimulation are ensured through real-time monitoring.

[0024] 2. By adopting the above technical solution, since a neural regulation model based on deep learning is used to intelligently analyze electroencephalogram signals and image data, precise positioning is realized in combination with a multi-modal navigation module, and then stimulation is carried out through the synergistic effect of three stimulation modules, so the technical problem that fixed stimulation parameters in the prior art are difficult to adapt to the dynamic change requirements of patients is effectively solved. Furthermore, personalized precise regulation based on the real-time brain activity state is realized, and the adaptability and treatment effect of neural regulation are significantly improved.

[0025] 3. By adopting the above technical solution, since a brain stimulation effect evaluation model based on deep learning is used in combination with the real-time monitoring function of an electroencephalogram detection module, the technical problem that it is difficult to objectively evaluate the treatment effect in the prior art is effectively solved. Furthermore, real-time and quantitative evaluation of the neural regulation effect is realized, providing a reliable basis for the timely adjustment of the treatment plan and ensuring the accuracy and controllability of the treatment process. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a module connection diagram of an embodiment of the present invention; Figure 2 is a system structure diagram of a closed-loop neural regulation device in an embodiment of the present application; Figure 3 is a schematic diagram of the application of a neural regulation device on the head in an embodiment of the present application; Figure 4 is a flowchart of a closed-loop neural regulation method in an embodiment of the present application; Figure 5 is a schematic diagram of an application scenario of a closed-loop neural regulation method; Figure 6 is a schematic diagram of the structure of an entity device of a closed-loop neural regulation device in an embodiment of the present application.

[0027] Description of the reference numerals: 1. Electroencephalogram acquisition module; 2. Magnetic resonance imaging module; 3. Transcranial alternating current stimulation module; 4. Low-field magnetic stimulation module; 5. Transcranial magnetic stimulation module; 6. Multimodal navigation module; 7. Electroencephalogram detection module; 8. Central control module. Detailed implementation manners

[0028] The terms used in the following embodiments of this application are only for the purpose of describing specific embodiments and are not intended to limit this application. As used in the description of this application and the appended claims, the singular forms "a", "an", "the", "above", "said", "this" are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term " / and" used in this application refers to and includes any or all possible combinations of one or more of the listed items.

[0029] Hereinafter, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as implying or suggesting relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of this application, unless otherwise stated, the meaning of "a plurality" is two or more.

[0030] The embodiments of this application provide a closed-loop neuromodulation device, as Figure 1 shown, Figure 1 is the module connection diagram of the embodiments of this application, including: The electroencephalogram acquisition module 1 is used to collect the electrical signals generated by the activities of brain neurons in real time through a plurality of electrodes attached to the user's scalp; The magnetic resonance imaging module 2 is used to obtain high-resolution structural and functional images of the brain; The transcranial alternating current stimulation module 3 is used to deliver alternating current with a set frequency and intensity to a set area of the brain through electrodes placed on the scalp; The low-field magnetic stimulation module 4 is used to generate a low-intensity magnetic field acting on the brain; The transcranial magnetic stimulation module 5 is used to generate a rapidly changing magnetic field to induce an induced current in brain tissue; The multimodal navigation module 6 is used to guide the stimulation module to accurately apply the stimulation to a set target according to the structural and functional characteristics of the brain; The electroencephalogram detection module 7 performs in-depth analysis and processing on the original electroencephalogram signals acquired by the electroencephalogram acquisition module 1; The central control module 8 is used to be responsible for the coordinated management of each module.

[0031] In this way, the electroencephalogram acquisition module 1 can collect in real time the electrical signals generated by the activities of brain neurons, enabling non-invasive and convenient acquisition of the electrophysiological information of brain nerve activities. The magnetic resonance imaging module 2 can clearly present the anatomical structure of the brain, including the morphology and positional relationships of gray matter, white matter, ventricles, etc., and can also reflect the functional activities of the brain, such as the activation of different brain regions during the execution of specific tasks. By providing these detailed imaging data, doctors can more intuitively understand the structural and functional states of the brain, providing a strong basis for accurately locating the lesion area and determining the neuromodulation target. The transcranial alternating current stimulation module 3 can regulate the excitability of brain neurons. By changing the frequency and intensity of the current, it is possible to enhance or inhibit the activities of neurons in different brain regions. The low-field magnetic stimulation module 4 can penetrate the skull and act on brain tissues, affecting the membrane potential of neurons and the release of neurotransmitters, thereby regulating the nerve activities of the brain. The advantage of this module is that its stimulation is relatively mild and has a high safety level, and it can be used for a long time and multiple times. The induced current generated by the rapidly changing magnetic field in brain tissues can directly stimulate neurons, causing the excitation or inhibition of neurons. This stimulation method of the transcranial magnetic stimulation module 5 has the advantages of non-invasiveness and simple operation, and can quickly and effectively regulate the nerve activities of the cerebral cortex. With the brain structure and function images obtained by the magnetic resonance imaging module 2, the multi-modal navigation module 6 can accurately calculate the three-dimensional coordinates of the stimulation target and guide each stimulation module to accurately act on the target area. This greatly improves the accuracy of neuromodulation, avoids unnecessary stimulation of the surrounding normal brain tissues, reduces the occurrence of side effects, and at the same time improves the treatment effect, making neuromodulation treatment safer and more effective. The electroencephalogram detection module 7 can remove the noise and interference in the original electroencephalogram signals and extract valuable neuroelectrophysiological features, such as electroencephalogram rhythms, event-related potentials, etc. By analyzing these features, it is possible to more accurately judge the activity state of the brain, evaluate the effect of neuromodulation, and provide data support for subsequent adjustment of stimulation parameters, realizing the feedback regulation function of the closed-loop neuromodulation system. The central control module 8 can, according to the preset programs and instructions, orderly control the startup, stop, and operation parameters of each module. For example, after obtaining the brain structure and function images, the central control module 8 will, according to the imaging data and the preset neuromodulation strategy, coordinate the multi-modal navigation module 6 to determine the target position, and then control the transcranial alternating current stimulation module 3, the low-field magnetic stimulation module 4, and the transcranial magnetic stimulation module 5 to perform stimulation according to the set parameters, while commanding the electroencephalogram detection module 7 to monitor and analyze the stimulation effect to ensure the efficient and stable operation of the entire neuromodulation process.

[0032] In some embodiments, the closed-loop neuromodulation device further includes a multi-mode power supply module for supplying stable power to each module of the closed-loop neuromodulation device. Specifically, the multi-mode power supply module integrates multiple power supply methods, such as being compatible with mains power supply, battery power supply, and wireless charging modes. Mains power supply can ensure a continuous and stable power source when the device is used in a fixed location, providing guarantee for the long-term stable operation of the device. When the device needs to be used in a mobile scenario, the battery power supply mode comes in handy. For example, when the patient goes out, the battery power supply can ensure the device works continuously without affecting the neuromodulation treatment process. The wireless charging mode further improves the convenience of device use. The user does not need to plug and unplug the charging cable tediously. Just place the device on the wireless charging device to charge, reducing the damage risk caused by frequent plugging and unplugging of the interface and improving the durability of the device.

[0033] The following describes the hardware closed-loop system structure of the closed-loop neuromodulation device provided in this embodiment. Please refer to Figure 2 , which is a system structure diagram of the closed-loop neuromodulation device in the embodiment of the present application.

[0034] Figure 2 In, at the top of the figure is the "computer application software", which serves as the control core of the entire system, responsible for unified management and data processing of each hardware module, and coordinating the overall operation of the system. The computer application software realizes its functions by connecting the four hardware modules below. From left to right, they are the Low-Field Magnetic Stimulation module (LFMS), the Transcranial Alternating Current Stimulation module (tACS), the Transcranial Magnetic Stimulation module (TMS), and the Electroencephalogram Acquisition module (EEG). Each stimulation module contains "stimulation" related components (such as stimulation coils, stimulation electrodes) and "synchronous output" components. These modules can perform corresponding stimulations on the human body and output synchronous signals. The electroencephalogram acquisition module is provided with a "DC input channel" and an "electroencephalogram channel" for inputting direct current and acquiring electroencephalogram signals. The computer application software is connected to the LFMS, tACS, and TMS modules through the "USB to serial port" method, and is connected to the EEG module through the "network port or USB to EEG" method. This connection method enables the computer application software to perform data transmission and instruction interaction with each hardware module, realizing closed-loop control.

[0035] For easy understanding, the following introduces the schematic diagram of the application device of the closed-loop neuromodulation device. Please refer to Figure 3 , Figure 3Shown is a schematic diagram of the application of a neuromodulation device on the head. The figure shows a head-mounted device with multiple electrodes and coils distributed on it. The head-mounted device is marked with "EEG", representing electroencephalogram acquisition electrodes for collecting electrical signals generated by the brain; "rTMS" represents repetitive transcranial magnetic stimulation coils, which are usually used to apply magnetic stimulation to specific regions of the brain; "tACS" represents transcranial alternating current stimulation electrodes, which can apply weak alternating current stimulation to the brain; "LFMS" represents a low-field magnetic stimulation device, which is also used to perform magnetic stimulation on the brain. In addition, there is a label of "synchronous anti-interference" in the figure, indicating that when the system performs various stimulations and signal acquisitions, corresponding measures are taken to reduce interference between different devices to ensure the accuracy of signal acquisition and stimulation.

[0036] For ease of understanding, the method flow of the closed-loop neuromodulation device provided in this embodiment will be described below. Please refer to Figure 4 , which is a schematic flow diagram of a closed-loop neuromodulation method in an embodiment of the present application.

[0037] S401. Obtain multiple electroencephalogram signal information through an electroencephalogram acquisition module, and combine it with an electroencephalogram detection module to determine real-time brain activity state information; The electroencephalogram acquisition module of the closed-loop neuromodulation device is equipped with multiple carefully designed electrodes, which are precisely attached to the user's scalp. The design is based on the neuroelectrophysiological distribution characteristics of the brain scalp. Electrodes at different positions can collect electrical signals generated by the activities of neurons in the corresponding brain regions. When brain neurons are active, extremely weak electrical signals are generated, and these signals are conducted out from the inside of the brain through tissues such as the skull and scalp. The electroencephalogram acquisition module can capture these weak signals in real time with its highly sensitive sensors. During the acquisition process, to ensure the accuracy and stability of the signals, the module will adopt a series of advanced technologies, including using special filtering technologies to reduce external electromagnetic interference, such as mobile phone signals and electromagnetic radiation generated by electrical appliances commonly found in life. These interferences may be mixed into the electroencephalogram signals and affect the accuracy of subsequent analysis. At the same time, by optimizing the contact method between the electrodes and the scalp, the signal acquisition efficiency is improved to ensure that the collected electroencephalogram signals can truly reflect the activities of brain neurons.

[0038] The collected original electroencephalogram signals are transmitted to the electroencephalogram detection module. The electroencephalogram detection module will first preprocess these original signals and use digital filtering algorithms to further remove the noise in the signals. This noise may come from the body's own physiological activities, such as electromyographic interference generated by muscle movement and electrocardiographic interference caused by the heartbeat. Through these filtering operations, the useful electroencephalogram signal components in the original signals can be highlighted.

[0039] Next, the electroencephalogram (EEG) detection module will extract features from the preprocessed signals. Brain activities generate EEG signals with different frequencies, amplitudes, and rhythms in different states. Common EEG rhythms include alpha waves, beta waves, theta waves, and delta waves, etc. Different EEG rhythms correspond to different brain activity states. For example, alpha waves are more obvious when a person is in a relaxed and eyes-closed state, while beta waves dominate when a person is in a tense, excited, or focused state. The EEG detection module determines the current brain activity state by analyzing features such as the frequency components and power spectra of the signals.

[0040] S402. Obtain brain structure and functional imaging data through the magnetic resonance imaging module; At the hardware level, this module is equipped with a high-field superconducting magnet that can generate a stable and uniform magnetic field, and its field strength is usually between 1.5T and 3.0T, and even higher-field magnets are gradually applied to medical devices. The role of the strong magnetic field is to make the hydrogen nuclei in the brain (mainly from water molecules) produce a specific spin arrangement, laying the foundation for subsequent imaging. At the same time, the device is also equipped with a gradient coil system for generating small magnetic field gradient changes on the basis of the main magnetic field. These gradient changes can accurately locate the brain tissue signals at different levels and positions, thereby realizing the precise encoding of the three-dimensional space of the brain. The radiofrequency transmission coil is responsible for transmitting radiofrequency pulses with a specific frequency to the brain. When the frequency of the radiofrequency pulse is consistent with the precession frequency of the hydrogen nuclei, a resonance phenomenon will occur, and the hydrogen nuclei absorb energy and change their spin states. The radiofrequency receiving coil is used to receive the energy signals released by the hydrogen nuclei during the relaxation process, and these signals contain rich information about the brain tissue.

[0041] During the data acquisition process, in order to obtain high-quality imaging data, a variety of imaging sequences are used. For brain structure imaging, common sequences include T1-weighted imaging (T1WI) and T2-weighted imaging (T2WI). T1WI can clearly show the contrast differences between different tissues such as gray matter, white matter, and cerebrospinal fluid in the brain. Gray matter shows medium signal intensity on T1WI images, white matter has a higher signal, and cerebrospinal fluid has a lower signal. Through this contrast, doctors can directly observe the anatomical structure of the brain, such as the morphology and positional relationships of gyri, sulci, ventricles, etc., to help determine the normal structure of the brain and whether there are lesions, such as the location and scope of abnormal areas like tumors and cerebral infarctions. T2WI is more sensitive to the water content in tissues and has advantages in showing lesions such as brain edema and inflammation in the brain, and can provide more detailed structural references for the selection of neuromodulation targets.

[0042] In the aspect of functional imaging, blood oxygenation level-dependent (BOLD) imaging technology is mainly adopted. This technology is based on the principle of blood oxygen metabolism changes in local brain tissue during brain neuron activities. When the neuron activities in a certain area of the brain increase, the blood flow in the local brain tissue increases, but the increase in oxygen consumption is relatively small, resulting in an increase in local blood oxygen saturation and a decrease in the content of deoxyhemoglobin. Since deoxyhemoglobin has paramagnetism, it will cause local magnetic field inhomogeneity, thus affecting the magnetic resonance signal. BOLD imaging reflects the brain's functional activities by detecting the change in magnetic resonance signal intensity caused by this blood oxygen saturation change. For example, when performing cognitive tasks (such as memory tests, language processing, etc.) or sensory stimuli (such as visual, auditory stimuli), the corresponding functional areas of the brain will be activated. BOLD imaging can capture the signal changes in these activated areas in real time and generate brain function maps. These function maps can help doctors understand the functional distribution of different areas of the brain and the impact of lesions on brain functions, providing a functional basis for precise neuromodulation.

[0043] The magnetic resonance imaging module of the closed-loop neuromodulation device adopts advanced superconducting magnet technology, which can generate a high-field and uniform magnetic field to ensure precise imaging of the brain. During the imaging process, to improve the image resolution and quality, the device will automatically adjust the scanning parameters according to individual differences such as the patient's body shape and brain size.

[0044] S403. Combine the brain activity state information and the preset neuromodulation strategy, and determine the stimulation parameter information of the transcranial alternating current stimulation module, the low-field magnetic stimulation module, and the transcranial magnetic stimulation module through a neuromodulation model. This neuromodulation model is constructed in advance through deep learning using a set of brain activity state information annotated with the stimulation parameter information of multiple transcranial alternating current stimulation modules, low-field magnetic stimulation modules, and transcranial magnetic stimulation modules. After obtaining the brain activity state information and the preset neuromodulation strategy, the closed-loop neuromodulation device uses a neuromodulation model to determine the stimulation parameter information of the three stimulation modules. This process relies on deep learning technology. Through learning a large amount of labeled data, the model can accurately find the most suitable combination of stimulation parameters for the current brain state. The specific model construction process is as follows: First, a large amount of brain activity state information sets with labels of the stimulation parameter information of the transcranial alternating current stimulation module, low-field magnetic stimulation module, and transcranial magnetic stimulation module need to be collected. These data come from a large number of clinical trials and research, covering patients of different age groups, different disease types (such as neurological diseases like depression, Parkinson's disease, epilepsy, etc.), and different disease severity levels. During the data collection process, the experimental conditions are strictly controlled to ensure the accuracy and reliability of the data. For example, for transcranial alternating current stimulation, the current intensity of each stimulation (usually between 0 - 2 mA), frequency (such as different frequency settings from 1 Hz - 100 Hz), waveform (sine wave, square wave, etc.), and stimulation duration (ranging from a few minutes to dozens of minutes) are detailedly recorded; for low-field magnetic stimulation, parameters such as magnetic field intensity (generally in the range of 0.01 T - 0.1 T), magnetic field direction, stimulation mode (continuous stimulation or pulsed stimulation) are recorded; for transcranial magnetic stimulation, information such as stimulation frequency (such as below 1 Hz for low frequency and above 5 Hz for high frequency), pulse intensity (expressed as a percentage of the maximum stimulation output), and stimulation site are recorded. At the same time, through various monitoring means, such as electroencephalogram monitoring, functional magnetic resonance imaging monitoring, etc., the brain activity state information before and after stimulation is synchronously obtained, including the frequency change of electroencephalogram signals, power spectrum change, activation or inhibition of brain functional regions, etc., and this information is accurately labeled with the corresponding stimulation parameters.

[0045] Next, deep learning algorithms such as long short-term memory networks can be selected. It can effectively process time series data, capture the changing characteristics of brain activity states over time, and the complex dependence relationships between brain activity states and stimulation parameters. The input layer of the model receives brain activity state information (such as preprocessed electroencephalogram signal feature vectors, brain functional area activation data extracted by functional magnetic resonance imaging, etc.) and preset neuromodulation strategies (such as treatment goals for different diseases, stimulation priorities, etc.). Through complex nonlinear transformations in multiple hidden layers, deep feature extraction and analysis of the input data are performed. In the hidden layer, neurons transmit information through weight connections, and these weights are continuously adjusted and optimized during the training process so that the model can learn the potential laws in the data. During the training process, the backpropagation algorithm is used to calculate the error between the model prediction results and the actual labeled data, and the weights of the model are continuously adjusted through optimization algorithms such as gradient descent to gradually reduce the error. After repeated training with a large amount of data, the model gradually learns the mapping relationship between brain activity states and stimulation parameters, and thus can accurately output the stimulation parameter information of transcranial alternating current stimulation modules, low-field magnetic stimulation modules, and transcranial magnetic stimulation modules according to the input brain activity state information and neuromodulation strategies, such as the optimal frequency, intensity, and duration of transcranial alternating current stimulation, the appropriate magnetic field intensity and stimulation mode of low-field magnetic stimulation, and the frequency, intensity, and number of stimulations of transcranial magnetic stimulation.

[0046] S404. Combine the preset neuromodulation goals, brain structure, and functional imaging data, and determine the target position information to be stimulated through the multi-modal navigation module; When determining the target position information to be stimulated, the multi-modal navigation module of the closed-loop neuromodulation device integrates a variety of advanced technologies. From a hardware perspective, the multi-modal navigation module integrates an optical tracking system, an electromagnetic positioning component, and a three-dimensional reconstruction algorithm based on magnetic resonance imaging data. The optical tracking system installs optical markers on the patient's head and the stimulation device, and uses an infrared camera to capture the position and attitude changes of the markers in real time, which can provide sub-millimeter-level accuracy and track the minute movements of the patient's head in real time. The electromagnetic positioning component emits and receives low-frequency electromagnetic field signals, which are not affected by obstacles. Even when there is partial coverage on the patient's head, it can accurately measure the relative position relationship between the head and the stimulation device, providing additional guarantee for navigation.

[0047] Using the high-resolution brain structure and functional imaging data obtained by magnetic resonance imaging, the module first performs a fine three-dimensional reconstruction of the brain. During the reconstruction process, different tissues such as gray matter, white matter, and ventricles of the brain are accurately distinguished through image segmentation technology, and at the same time, combined with the functional imaging data, the positions and ranges of each functional area are marked. For example, for the motor functional area, its specific position in the brain is determined according to the BOLD imaging technology and clearly presented in the three-dimensional model.

[0048] In terms of software algorithms, the multi-modal navigation module adopts spatial registration technology. Due to individual differences in the morphology and structure of the brains of different patients, the spatial registration technology matches and aligns the brain image data of the patient with a standard brain template. By calculating the spatial transformation relationship between the feature points in the image data (such as specific anatomical landmark points, boundary points of functional regions, etc.), the brain image of the patient is mapped into the standard space, enabling the comparison and analysis of the brain structures and functional regions of different patients under a unified coordinate system. In this way, doctors can more accurately determine the target position and avoid positioning errors caused by individual differences. To further improve the positioning accuracy, the multi-modal navigation module also has a real-time update and correction function. During the treatment process, the patient's head may move slightly. The optical tracking system and the electromagnetic positioning component will monitor these changes in real time and transmit the data to the multi-modal navigation module. The module dynamically adjusts and corrects the target position based on this real-time data to ensure that the stimulation always accurately acts on the target area.

[0049] In some embodiments, the brain structure is complex and fragile. If neuromodulatory stimulation acts on an inappropriate location, it may trigger serious adverse reactions, affecting the treatment effect and even endangering the health of the patient. Therefore, before this step, it is necessary to conduct a safety assessment of the target location information before implementing neuromodulatory stimulation. Preset safety criteria can be formulated based on a large number of clinical studies, anatomical data, and past neuromodulatory practice experience. For example, referring to anatomical information such as the functional importance of different brain regions, blood vessel distribution, and nerve fiber orientation, the stimulation safety boundaries of different brain regions are determined. At the same time, combining the probabilities and severities of adverse reactions under different stimulation parameters in clinical studies, corresponding safety thresholds are formulated. The calculation of the safety factor takes multiple factors into comprehensive consideration. Firstly, it is the distance between the target and surrounding important structures (such as large blood vessels, key nerve nuclei). Using magnetic resonance imaging data, the spatial distance between the target and these important structures is accurately measured through image analysis algorithms, and different weights are assigned according to the distance. For example, the closer to a large blood vessel, the higher the weight and the greater the impact on the safety factor. Secondly, consider the potential impact of stimulation parameters (such as current intensity, magnetic field intensity, stimulation frequency, etc.) on surrounding tissues. By establishing a biophysical model, the distribution of electric and magnetic fields in brain tissue under different stimulation parameters is simulated to evaluate the degree of stimulation of surrounding tissues. These factors are comprehensively calculated to obtain a quantified safety factor. Setting the safety factor threshold is to judge whether the target location is safe. The determination of this threshold is also based on clinical experience and research data to balance the treatment effect and safety. When the calculated safety factor is greater than the set safety factor threshold, it indicates that the target location is relatively safe, and the target location information can be sent to the transcranial alternating current stimulation module, low-field magnetic stimulation module, and transcranial magnetic stimulation module to initiate the stimulation operation. If the safety factor does not reach the threshold, the stimulation is paused, and the doctor is prompted to re-evaluate the target location or adjust the stimulation plan, such as changing the target location, adjusting the stimulation parameters, etc., to increase the safety factor and ensure the safety of the treatment.

[0050] In addition, artificial intelligence-assisted positioning technology can also be introduced. Specifically, a large amount of brain imaging data can be collected in advance. These data come from a wide range of sources, including magnetic resonance imaging (MRI), positron emission tomography (PET), and other imaging examination results from different medical institutions. At the same time, the corresponding determined target position information is collected. This information has been strictly evaluated and determined by professional doctors and has high accuracy and reliability. The data covers the situations of patients of different ages, genders, and disease types (such as Parkinson's disease, epilepsy, depression, etc.) to ensure that the model can learn more comprehensive features. Preprocessing the collected brain imaging data is a crucial step. First, image enhancement operations are performed. By adjusting parameters such as the contrast and brightness of the image, the brain structure in the image becomes clearer, facilitating subsequent feature extraction. Then, normalization processing is carried out to unify the imaging data obtained under different devices and scanning conditions to the same scale and range, eliminating data inconsistencies. According to the characteristics of the task and the nature of the data, a suitable deep learning model such as a convolutional neural network (CNN) is selected. In the target positioning task, CNN is widely used because of its powerful ability in image feature extraction. Through the combination of convolutional layers, pooling layers, and fully connected layers, CNN can automatically extract key features in brain images and learn the mapping relationship between imaging data and target positions. When building a deep learning model, it is necessary to determine the architecture and parameters of the model. Multiple convolutional layers and pooling layers can be set for feature extraction, and then the extracted features are mapped to the prediction space of the target position through fully connected layers. Then, the preprocessed brain imaging data and target position information are divided into a training set, a validation set, and a test set. The training set is used for parameter learning of the model, the validation set is used to adjust the hyperparameters of the model (such as learning rate, batch size, etc.) during the training process, and the test set is used to evaluate the final performance of the model.

[0051] In practical applications, doctors input the patient's brain imaging data and preset neuroregulation goals (such as treating a certain disease, improving a certain symptom, etc.) into the trained artificial intelligence model. After receiving the input data, the model quickly performs calculations and analyzes to generate multiple potential target position suggestions. These suggestions are based on a large amount of data and patterns learned by the model, considering factors such as the anatomical structure of the brain, disease characteristics, and neuroregulation goals. At the same time, the model will give a confidence level for each suggestion, which reflects the reliability assessment of the model for this suggestion. The calculation of the confidence level is usually based on the prediction probability or uncertainty estimation of the model. The variance of the prediction results can be calculated by sampling the prediction results of the model multiple times through a set method to estimate the uncertainty. Doctors refer to the target position suggestions and confidence levels generated by the model and comprehensively evaluate these suggestions in combination with their own clinical experience and professional knowledge. Doctors can view the brain imaging data and target position suggestions on the visualization interface to intuitively understand the rationality and feasibility of each suggestion.

[0052] S405. Combine the target position information and control the transcranial alternating current stimulation module, low-field magnetic stimulation module, and transcranial magnetic stimulation module to perform stimulation according to a preset neuromodulation strategy.

[0053] After determining the target position information, the closed-loop neuromodulation device accurately controls the transcranial alternating current stimulation module, low-field magnetic stimulation module, and transcranial magnetic stimulation module to perform collaborative stimulation according to a preset neuromodulation strategy.

[0054] For the transcranial alternating current stimulation module, according to the stimulation parameters determined by the preset neuromodulation strategy and neuromodulation model, such as current intensity, frequency, and waveform, etc., alternating current is delivered to the set area of the brain through scalp electrodes. In actual operation, in order to ensure the uniform distribution of current in the target area, electrode array optimization technology is adopted. By optimizing the shape, size, and layout of the electrodes, the current can act more accurately on the target, reducing the impact on the surrounding normal brain tissue. At the same time, using feedback control technology, the contact resistance between the electrode and the scalp is monitored in real time. When the resistance changes, the output current is automatically adjusted to ensure the stability and consistency of the stimulation.

[0055] After receiving the stimulation instruction, the low-field magnetic stimulation module generates a low-intensity magnetic field to act on the brain according to parameters such as the preset magnetic field intensity and stimulation mode (continuous or pulsed). In order to improve the focusing and penetration of the magnetic field, magnetic focusing technology is adopted. Through a specially designed magnetic coil structure and magnetic field distribution algorithm, the magnetic field can act more concentratedly on the target area, enhancing the stimulation effect. In addition, in order to ensure the safety of the patient, a magnetic field intensity monitoring device is integrated in the module to monitor the magnetic field intensity in real time to prevent excessive magnetic field intensity from causing harm to the patient.

[0056] The transcranial magnetic stimulation module generates a rapidly changing magnetic field according to parameters such as the preset stimulation frequency, pulse intensity, and number of stimulations, so that the brain tissue generates an induced current, thereby stimulating neurons. In terms of technical implementation, a high-power pulse generator and an efficient magnetic coil design are adopted, which can quickly generate a high-intensity magnetic field change. At the same time, in order to reduce the discomfort during the stimulation process, adaptive pulse adjustment technology is adopted. According to the real-time feedback of the patient and the physiological response of the brain, the intensity and frequency of the pulse are automatically adjusted to make the stimulation process more comfortable.

[0057] In the embodiments of the present application, the electroencephalogram acquisition module and the electroencephalogram detection module are used to obtain and analyze the brain activity state information in real time. The magnetic resonance imaging module provides accurate brain structure and functional imaging data. The neural regulation model intelligently determines the stimulation parameters. The multi-modal navigation module accurately determines the target position. And the multi-module collaboratively controls the stimulation module to perform stimulation, realizing precise neural regulation of different regions and depths of the brain. This not only effectively solves the problems of inaccurate positioning, difficult adaptation of stimulation parameters to individual dynamic changes, and difficult objective evaluation of treatment effects in existing neural regulation technologies, but also significantly improves the accuracy, effectiveness, adaptability, and controllability of neural regulation, providing a more reliable and personalized solution for the treatment of nervous system diseases.

[0058] In some embodiments, after step S405, the original electroencephalogram signals acquired by the acquisition module contain rich information about brain neuron activities, but are also mixed with various noises, such as myoelectric interference, electrocardiogram interference from the human body itself, and external electromagnetic interference, etc. The electroencephalogram detection module first uses digital filtering algorithms, such as band-pass filtering, wavelet filtering and other technologies, to remove these noises and retain the useful electroencephalogram signal components. Then, through the feature extraction algorithm, the preprocessed signals are analyzed. The system evaluates the stimulation effect according to the pre-set brain stimulation effect evaluation model. In the training stage of the brain stimulation effect evaluation model, a large number of brain activity state information sets with stimulation effect score annotations can be collected in advance. These data cover the brain responses of different patients under different stimulation parameters. The brain activity state information obtained by the electroencephalogram detection module, such as electroencephalogram signal feature vectors, brain functional area activation data, etc., is used as the model input. The model undergoes complex operations of multiple layers of neurons to learn the mapping relationship between the brain activity state and the stimulation effect score. During training, the backpropagation algorithm is used to adjust the model parameters to minimize the error between the model predicted score and the actual annotated score. After training, in actual applications, the model quickly outputs the stimulation effect score according to the input brain activity state information. Then, through network communication technologies, such as Wi-Fi, Bluetooth or mobile networks, the score is sent to the user terminal device, such as a doctor's computer, mobile phone APP, etc. On the user terminal device, the effect score threshold is pre-set. When the stimulation effect score is received, the system automatically compares it with the threshold. If the score is lower than the threshold, the user terminal device issues a reminder through the built-in reminder mechanism. For example, on the mobile phone APP, it notifies the medical staff that the current stimulation effect is poor through vibration, pop-up window or sound reminder, etc.; on the computer side, it reminds the doctor through screen pop-up window, system prompt sound and other forms, so as to adjust the treatment plan in time.

[0059] It is also possible to determine the locations of different functional regions by combining functional magnetic resonance imaging data or pre-established brain function maps. By comparing the activation levels of brain functional regions before and after stimulation during the execution of specific tasks, the improvement of activities can be evaluated. For example, for cognitive functional regions, let the patient perform tasks such as memory tests and attention tests, and at the same time monitor the changes in the electroencephalogram (EEG) activities in this region. If the performance of task completion improves after stimulation and the EEG activities in the corresponding region increase, it indicates an improvement in activities. Perform spectral analysis on the EEG signals of different functional regions before and after stimulation, and calculate the power spectral density in specific frequency bands. Based on a large number of clinical studies and EEG data of the normal population, establish ideal EEG patterns under different functional states. Compare the EEG signals of each functional region after stimulation with the ideal pattern, and evaluate the matching degree by calculating the signal similarity, such as using methods like the Pearson correlation coefficient. The higher the matching degree, the closer the stimulation effect is to the expectation. Then, use data visualization technology to present the obtained stimulation effect data to the user side in an intuitive and easy-to-understand manner. For the improvement of activities, a bar chart can be used to show the comparison of task completion performance of different functional regions before and after stimulation; for the change in frequency stability, a line chart can be used to show the change trend of the EEG frequency stability index of each region at different time points; for the matching degree with the preset ideal EEG pattern, a radar chart can be used to show the matching degree of each functional region in different dimensions. Integrate these visualization charts into the user-side interface to facilitate doctors to view and analyze intuitively, providing a basis for adjusting the neuromodulation strategy.

[0060] The following introduces the schematic diagram of the application scenario of the application. Please refer to Figure 5 , Figure 5 which is a schematic diagram of an application scenario of the closed-loop neuromodulation method.

[0061] In Figure 5In it, there are multiple function options above the screenshot of the software operation interface, such as "Home", "Patient", "Program", etc., which facilitate users to perform operations of different functions. In the middle part of the interface, there are parameter setting areas related to TMS, LFMS, and tACS respectively. For example, in the TMS area, parameters such as program, stimulation intensity, and frequency can be set; the LFMS area also has similar parameter setting items, including stimulation intensity, frequency, etc. There are also operation buttons for starting and stopping stimulation below, which facilitate users to start or stop stimulation. After the EEG collects brain signals, it will perform synchronous output and transmit the signals to the software system. The software system controls the TMS, LFMS, and tACS devices to enter the collaborative working mode according to the preset program and algorithm, combined with the brain state reflected by the EEG signals. The collaborative working mode can be selected according to the actual situation, including but not limited to the collaborative working modes in the three cases of TMS+LFMS, TMS + tACS, and TMS+LFMS+tACS. After the user selects a specific collaborative working mode, the TMS, LFMS, and tACS devices will send stimulation signals to multiple target points of the brain according to the parameters set by the software, realizing precise regulation of the brain to achieve the purpose of treating related neurological diseases or improving brain function.

[0062] The closed-loop neuromodulation device in the embodiment of the present application will be described below from the perspective of hardware processing. Please refer to Figure 6 , which is a schematic structural diagram of an entity device of the closed-loop neuromodulation device in the embodiment of the present application.

[0063] It should be noted that Figure 6 The structure of the closed-loop neuromodulation device shown is only an example and should not bring any limitation to the functions and usage scope of the embodiments of the present application.

[0064] As Figure 6 shown, the closed-loop neuromodulation device includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 602 or the program loaded from the storage part 608 into the random access memory (RAM) 606, such as executing the method described in the above embodiment. In the RAM 606, various programs and data required for system operation are also stored. The CPU 601, ROM 602, and RAM 606 are connected to each other through a bus 604. The input / output (I / O) interface 605 is also connected to the bus 604.

[0065] The following components are connected to the I / O interface 605: an input section 606 including an audio input device, a button switch, etc.; an output section 607 including a liquid crystal display (LCD), an audio output device, an indicator light, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. The drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed so that a computer program read from it can be installed into the storage section 608 as needed.

[0066] Specifically, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication section 609, and / or installed from the removable medium 611. When the computer program is executed by the central processing unit (CPU) 601, various functions defined in the present application are executed.

[0067] It should be noted that specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer-readable storage medium can be any tangible medium that contains or stores a program, and the program can be used by or combined with an instruction execution system, device, or component.

[0068] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present application. Among them, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the above-mentioned module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings.

[0069] Specifically, the closed-loop neuromodulation device of this embodiment includes a processor and a memory, and a computer program is stored on the memory. When the computer program is executed by the processor, the closed-loop neuromodulation method provided in the above embodiment is implemented.

[0070] On the other hand, the present application also provides a computer-readable storage medium, which may be included in the closed-loop neuromodulation device described in the above embodiment; or it may exist separately without being assembled into the closed-loop neuromodulation device. The above storage medium carries one or more computer programs. When the one or more computer programs are executed by a processor of the closed-loop neuromodulation device, the closed-loop neuromodulation device is enabled to implement the pending method provided in the above embodiment.

[0071] As mentioned above, the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the various embodiments of the present application.

[0072] As used in the above embodiments, depending on the context, the term "when..." may be interpreted to mean "if..." or "after..." or "in response to determining..." or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if detecting (the stated condition or event)" may be interpreted to mean "if determining..." or "in response to determining..." or "when detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)".

[0073] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by relevant hardware instructed by a computer program. This program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above method embodiments. The aforementioned storage medium includes various media that can store program codes, such as ROM, random access memory (RAM), magnetic disks, or optical discs.

Claims

1. A closed-loop neuromodulation device, comprising: An electroencephalogram (EEG) acquisition module for collecting, in real time, electrical signals generated by the activities of brain neurons through a plurality of electrodes attached to the user's scalp; A magnetic resonance imaging (MRI) module for obtaining high-resolution structural and functional images of the brain; A transcranial alternating current stimulation module for delivering an alternating current with a set frequency and intensity to a set area of the brain through electrodes placed on the scalp; A low-field magnetic stimulation module for generating a low-intensity magnetic field acting on the brain; A transcranial magnetic stimulation module for generating a rapidly changing magnetic field to induce an induced current in brain tissue; A multimodal navigation module for guiding the stimulation module to accurately apply stimulation to a set target according to the structural and functional characteristics of the brain; An EEG detection module for performing in-depth analysis and processing on the raw EEG signals obtained by the EEG acquisition module; A central control module for responsible for the coordinated management of each module.

2. The closed-loop neuromodulation device according to claim 1, wherein It further includes: A multimodal power supply module for supplying stable power to each module of the closed-loop neuromodulation device.

3. A closed-loop neuromodulation method, applied to a closed-loop neuromodulation device, characterized in that, The method includes: Obtaining multiple EEG signal information through the EEG acquisition module and determining real-time brain activity state information in combination with the EEG detection module; Obtaining brain structural and functional image data through the MRI module; Combining the brain activity state information and a preset neuromodulation strategy, and determining the stimulation parameter information of the transcranial alternating current stimulation module, the low-field magnetic stimulation module, and the transcranial magnetic stimulation module through a neuromodulation model, where the neuromodulation model is pre-constructed through deep learning from a set of brain activity state information annotated with the stimulation parameter information of the transcranial alternating current stimulation module, the low-field magnetic stimulation module, and the transcranial magnetic stimulation module; Combining a preset neuromodulation target, brain structural and functional image data, and determining the target position information to be stimulated through the multimodal navigation module; Combining the target position information and controlling the transcranial alternating current stimulation module, the low-field magnetic stimulation module, and the transcranial magnetic stimulation module to perform stimulation according to a preset neuromodulation strategy.

4. The method according to claim 3, characterized in that, After the step of combining a preset neuromodulation target, brain structural and functional image data, and determining the target position information to be stimulated through the multimodal navigation module, it further includes: Performing a safety assessment on the target position information according to a preset safety standard to determine a safety factor; If the safety factor is greater than a set safety factor threshold, sending the target position information to the transcranial alternating current stimulation module, the low-field magnetic stimulation module, and the transcranial magnetic stimulation module.

5. The method according to claim 3, wherein After the step of combining the target position information and controlling the transcranial alternating current stimulation module, the low-field magnetic stimulation module, and the transcranial magnetic stimulation module to perform stimulation according to a preset neuromodulation strategy, it further includes: Obtaining multiple activity state information of the brain in real time through the EEG detection module; Combining the multiple activity state information, determining a stimulation effect score through a brain stimulation effect evaluation model, and sending it to the user terminal, where the brain stimulation effect evaluation model is pre-constructed through deep learning from a set of brain activity state information annotated with stimulation effect scores.

6. The method according to claim 5, characterized in that, After the step of combining the multiple activity state information, determining a stimulation effect score through a brain stimulation effect evaluation model, and sending it to the user terminal, it further includes: If the stimulation effect score is lower than the set effect score threshold, a reminder is sent to the user terminal.

7. The method according to claim 5, wherein After the step of determining the stimulation effect score through the brain stimulation effect evaluation model in combination with the multiple activity state information, the method further includes: Obtaining the stimulation effect data of different functional regions of the brain, where the stimulation effect data at least includes the improvement of activity, the change in the frequency stability of the electroencephalogram signal before and after stimulation, and the degree of matching with the preset ideal electroencephalogram pattern; Presenting the stimulation effect data to the user terminal in a visual form.

8. A closed-loop neuromodulation device, characterized in that, The closed-loop neuromodulation device includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to cause the closed-loop neuromodulation device to execute the method according to any one of claims 1-7.

9. A computer-readable storage medium, comprising instructions, characterized in that, When the instruction runs on the closed-loop neuromodulation device, it causes the closed-loop neuromodulation device to execute the method according to any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product runs on the closed-loop neuromodulation device, it causes the closed-loop neuromodulation device to execute the method according to any one of claims 1-7.

Images