A closed-loop neuromodulation method, system, device, medium and product based on rehabilitation training

Abstract

The application discloses a closed-loop neural regulation method, system, device, medium and product based on rehabilitation training, and relates to the field of medical rehabilitation. The method comprises the following steps: after acquiring a motion signal of a target person, sequentially performing smoothing and feature extraction to obtain a motion change characteristic value; if the value is smaller than a minimum motion threshold value, determining invalidity and continuing acquisition; otherwise, taking the value as a motion characteristic sample; based on a historical motion cache area, determining a dynamic trigger threshold value by using a statistical distribution quantile method; comparing the motion characteristic sample with the dynamic trigger threshold value; if the motion characteristic sample is smaller, updating the cache area; otherwise, calculating a time interval between a current time and a last stimulation trigger time; if the interval is smaller than a minimum interval threshold value, not stimulating and updating the historical cache area; otherwise, taking the current time as the stimulation trigger time, issuing a stimulation trigger instruction, and controlling a neural stimulation device to stimulate. The method improves the stability and adaptability of stimulation triggering.

Description

Technical Field

[0001] This application relates to the field of medical rehabilitation, and in particular to a closed-loop neuromodulation method, system, device, medium, and product based on rehabilitation training. Background Technology

[0002] Stroke is one of the leading causes of disability among adults. Achieving more precise, real-time, and individualized motor control during stroke rehabilitation to improve upper limb motor function recovery is a critical technical challenge in the field of stroke rehabilitation. Vagus nerve stimulation (VNS), a neuromodulation technique capable of regulating central nervous system activity, has been increasingly applied in neurorehabilitation in recent years. However, current common VNS stimulation methods still have certain shortcomings. On the one hand, VNS stimulation often relies on therapists observing and manually triggering the target individual's movements during training, increasing the clinical workload and making it difficult to apply to home or remote rehabilitation scenarios. On the other hand, periodic stimulation based on fixed time intervals is difficult to adjust according to the target individual's real-time motor performance, failing to ensure effective pairing between stimulation and high-quality motor movements. Furthermore, some automatic triggering schemes based on fixed thresholds are difficult to adapt to differences in motor ability among different target individuals, and even among the same target individual at different training stages, easily leading to problems such as overly frequent, sparse, or non-optimal triggering, thus affecting the stimulation effect and safety.

[0003] Existing methods lack unified and reliable theoretical guidance for selecting stimulus triggering timing. Although existing research indicates that stimuli should be matched with the target individual's motor behavior as closely as possible, effective methods for objectively assessing movement quality and determining appropriate stimulus triggering times during continuous training are still lacking. Furthermore, existing methods lack individualized adjustment capabilities in stimulus triggering strategies. Due to significant differences in motor ability, injury severity, and performance during training among different target individuals, stimulus triggering methods based on fixed rules or experience are difficult to adapt to these differences, easily leading to unstable stimulus triggering or pairing with suboptimal motor movements. Existing methods also fall short in terms of automation; most stimulus methods rely on manual judgment or simple rules, making it difficult to achieve long-term, stable operation in applications such as home rehabilitation or remote rehabilitation.

[0004] Therefore, based on the above problems, there is an urgent need to provide a closed-loop neuromodulation method based on rehabilitation training, which can improve the stability and adaptability of stimulus triggering. Summary of the Invention

[0005] The purpose of this application is to provide a closed-loop neuromodulation method, system, device, medium, and product based on rehabilitation training, which can improve the stability and adaptability of stimulus triggering.

[0006] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a closed-loop neuromodulation method based on rehabilitation training, comprising: Acquire the motion signals generated by the target individuals during rehabilitation training; The motion signal is smoothed and its features are extracted sequentially to obtain motion change feature values; The motion change feature value is compared with the minimum motion threshold. If the motion change feature value is less than the minimum motion threshold, the motion change feature value is determined to be invalid, and motion signal acquisition continues. If the motion change feature value is greater than or equal to the minimum motion threshold, the motion change feature value is used as a motion feature sample. A historical motion buffer is constructed based on historical motion feature samples, and a dynamic trigger threshold is determined based on the threshold method of statistical distribution quantiles. The motion feature sample is compared with the dynamic trigger threshold; if the motion feature sample is less than the dynamic trigger threshold, the historical motion buffer is updated based on the motion feature sample; if the motion feature sample is greater than or equal to the dynamic trigger threshold, the time interval between the current moment and the last stimulus trigger moment is determined. The time interval is compared with the minimum interval threshold; if the time interval is less than the minimum interval threshold, no neural stimulation is applied to the target person, and the historical motion buffer is updated based on the motion feature samples; if the time interval is greater than or equal to the minimum interval threshold, the current moment is used as the stimulation trigger moment. The stimulus triggering command is determined based on the timing of the stimulus triggering; and the nerve stimulation device is controlled to provide nerve stimulation to the target person based on the stimulus triggering command.

[0007] Secondly, this application provides a closed-loop neuromodulation system based on rehabilitation training, comprising: The motion signal acquisition module is used to acquire motion signals generated by the target person during rehabilitation training. The motion change feature value determination module is used to perform smoothing and feature extraction on the motion signal sequentially to obtain motion change feature values; The motion feature sample determination module is used to compare motion change feature values ​​with a minimum motion threshold. If the motion change feature value is less than the minimum motion threshold, the motion change feature value is determined to be invalid, and motion signal acquisition continues. If the motion change feature value is greater than or equal to the minimum motion threshold, the motion change feature value is used as a motion feature sample. The dynamic trigger threshold determination module is used to construct a historical motion buffer based on historical motion feature samples and determine the dynamic trigger threshold based on the threshold method of statistical distribution quantiles. The time interval determination module is used to compare motion feature samples with dynamic trigger thresholds; if the motion feature samples are less than the dynamic trigger thresholds, the historical motion buffer is updated based on the motion feature samples; if the motion feature samples are greater than or equal to the dynamic trigger thresholds, the time interval between the current moment and the last stimulus trigger moment is determined. The stimulus trigger time determination module is used to compare the time interval with the minimum interval threshold. If the time interval is less than the minimum interval threshold, the target person will not be given neural stimulation, and the historical motion buffer will be updated based on the motion feature samples. If the time interval is greater than or equal to the minimum interval threshold, the current time will be used as the stimulus trigger time. The neurostimulation module is used to determine the stimulation trigger command based on the stimulation trigger time; and to control the neurostimulation device to perform neurostimulation on the target person according to the stimulation trigger command.

[0008] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described closed-loop neuromodulation method based on rehabilitation training.

[0009] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described closed-loop neuromodulation method based on rehabilitation training.

[0010] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described closed-loop neuromodulation method based on rehabilitation training.

[0011] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a closed-loop neuromodulation method, system, device, medium, and product based on rehabilitation training. It acquires continuous motion signals and obtains motion change feature values ​​based on smoothing and feature extraction. The motion change feature values ​​are compared with a minimum motion threshold to determine motion feature samples, filtering out invalid motion signals and ensuring the quality of the motion feature samples. A historical motion buffer is constructed based on the target person's historical motion feature samples to determine a dynamic trigger threshold. This application can dynamically determine the threshold based on the statistical distribution of recent motion performance, achieving dynamic evaluation of motion signals. The dynamic trigger threshold is adaptively adjusted according to changes in the target person's training status. The time interval is compared with a minimum interval threshold to determine the stimulus trigger time. Under the premise of ensuring a reasonable stimulus trigger interval, effective pairing of neural stimulation and the target person's motion signals is achieved, thereby improving the stability and individualization level of stimulus triggering. In summary, this application forms a complete closed-loop stimulus triggering mechanism that does not rely on manual judgment or static rules, realizing dynamic evaluation based on motion signals, adaptive updating of stimulus triggering conditions, and reasonable control of stimulus timing. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart illustrating a closed-loop neuromodulation method based on rehabilitation training in one embodiment of this application. Figure 2 This is a schematic diagram illustrating the process of determining motion change characteristic values ​​in one embodiment of this application; Figure 3 This is a schematic diagram of the dynamic trigger threshold determination process in one embodiment of this application. Detailed Implementation

[0014] 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, and 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.

[0015] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0016] The basic principle of VNS (Vacuum Nerve Stimulation) lies in regulating the neural activity of the central nervous system and promoting neural plasticity by applying electrical stimulation to the vagus nerve. In motor function rehabilitation after nerve injury, VNS is usually combined with active motor training to enhance the reorganization of neural networks during training. However, existing relatively mature VNS-assisted rehabilitation programs still have certain limitations, making it difficult to fully realize their potential advantages in practical applications. First, existing programs lack unified and reliable theoretical guidance on the selection of stimulus triggering timing. Although existing research has shown that stimuli should be matched with the target person's motor behavior as much as possible, there is still a lack of effective methods to objectively assess the quality of movement and determine the appropriate stimulus triggering time during continuous training. Second, existing programs lack individualized adjustment capabilities in stimulus triggering strategies. Due to significant differences in motor ability, injury degree, and performance during training among different target individuals, stimulus triggering methods based on fixed rules or experience are difficult to adapt to these differences, easily leading to unstable stimulus triggering or matching with non-optimal motor movements. Finally, existing programs are still insufficient in terms of automation. Most stimulation methods rely on manual judgment or simple rules, making it difficult to achieve long-term, stable operation in application scenarios such as home rehabilitation or remote rehabilitation.

[0017] In addition, existing stimulus triggering schemes can be categorized as follows: 1. The therapist observes the target individual's motor performance during training and manually triggers the stimulus when the individual performs a good movement. This approach can ensure the timing of stimulus and movement to a certain extent, but it is highly dependent on the therapist and is difficult to scale up and automate.

[0018] 2. Stimuli are applied to the target person at preset time intervals, regardless of the target person's real-time movement status. This method is simple to implement and has high stability, but it cannot distinguish the quality of different movements and is prone to pairing stimuli with low-quality or non-target movements.

[0019] 3. A triggering strategy based on a fixed threshold is adopted, which determines whether to trigger a stimulus by comparing the motion signal with a preset threshold. However, since the threshold setting is difficult to take into account the motion performance of different target individuals and different training stages, this type of scheme has poor stability in practical applications.

[0020] In summary, existing technologies still have shortcomings in terms of stimulus triggering timing, individualized regulation capabilities, and automated applications, and require further improvement. Therefore, this application proposes a closed-loop neuromodulation method based on rehabilitation training, which enables a complete closed-loop stimulus triggering mechanism that does not rely on manual judgment or static rules, and achieves dynamic evaluation based on motion signals, adaptive updating of stimulus triggering conditions, and rational control of stimulus timing.

[0021] In one exemplary embodiment, such as Figure 1 As shown, a closed-loop neuromodulation method based on rehabilitation training is provided, including the following S1 to S7. Wherein: S1: Acquire the motion signals generated by the target person during rehabilitation training.

[0022] Specifically, the motion signal can be any one or more of angle, displacement, velocity, or acceleration signals, with the aim of acquiring characteristic information that reflects changes in motion performance. In this application, motion signals generated by the target person during rehabilitation training are continuously acquired at a frequency of 60Hz.

[0023] S2: The motion signal is smoothed and its features are extracted sequentially to obtain motion change feature values.

[0024] The process of determining the characteristic values ​​of motion change is as follows: Figure 2 As shown, S2 specifically includes: S21: Smooth the motion signal to obtain a smooth signal.

[0025] S22: Determine the gradient of the smoothed signal within a preset time window, and determine the rate of change sequence of the smoothed signal.

[0026] Calculate the gradient of the smoothed signal at each sampling point within the most recent time window at the current moment to obtain the smoothed signal rate of change sequence.

[0027] S23: Take the mean of the smoothed signal rate of change sequence to obtain the motion change characteristic value.

[0028] The motion change characteristic value is the average rate of change within a preset time window T, used to characterize the strength of the motion change amplitude within this event window. The preset time window T is 300 ms. The motion change characteristic value reflects the average level of change during a complete motion change process, and the calculation formula for the motion change characteristic value is as follows: ; in, for Motion change characteristic value at time, for The smoothed signal rate of change at time T, where T is the preset time window. It is a smoothed signal.

[0029] S3: Compare the motion change feature value with the minimum motion threshold; if the motion change feature value is less than the minimum motion threshold, the motion change feature value is determined to be invalid, and motion signal acquisition continues; if the motion change feature value is greater than or equal to the minimum motion threshold, the motion change feature value is used as a motion feature sample.

[0030] Specifically, to avoid misinterpreting resting states or sensor noise as valid movement, a minimum movement threshold is set. The process for determining the minimum movement threshold is as follows: acquire the resting movement signals of the target person at multiple time periods; determine the maximum value of the resting movement signal in each time period, and average the multiple maximum values ​​to obtain the minimum movement threshold. Here, the resting movement signal refers to the movement signal generated by the target person in a resting state.

[0031] When the motion change feature value is less than the minimum motion threshold, the motion change feature value is determined to be an invalid sample and will not participate in the subsequent determination process of the dynamic trigger threshold; when the motion change feature value is greater than or equal to the minimum motion threshold, the motion change feature value will be used as a motion feature sample.

[0032] S4: Construct a historical motion buffer based on historical motion feature samples, and determine the dynamic trigger threshold based on the threshold method of statistical distribution quantiles.

[0033] S4 specifically includes: S41: Sort the historical motion feature samples in the historical motion buffer according to a preset order to obtain a motion feature sample sequence.

[0034] Specifically, let the set of motion feature samples within the historical motion buffer be... ,in, For the first One motion feature sample, , The total number of motion feature samples. This is the upper limit of the historical motion buffer capacity.

[0035] The sequence is obtained by sorting the values ​​from smallest to largest. ,in, For the sorted number A number of motion feature samples.

[0036] S42: Using a pre-set quantile, select a dynamic trigger threshold in the motion feature sample sequence.

[0037] In this application, the quantile is set as Then the dynamic trigger threshold Defined as the sequence obtained by sorting, located at Motion feature samples at quantile locations, i.e. .

[0038] S5: Compare the motion feature sample with the dynamic trigger threshold; if the motion feature sample is less than the dynamic trigger threshold, update the historical motion buffer based on the motion feature sample; if the motion feature sample is greater than or equal to the dynamic trigger threshold, determine the time interval between the current moment and the last stimulus trigger moment.

[0039] Specifically, such as Figure 3 As shown, the size of motion feature samples is evaluated based on a dynamic trigger threshold, and the historical motion buffer is updated according to the evaluation results. When updating the historical motion buffer, motion feature samples are added to the historical motion buffer, and the dynamic trigger threshold is updated based on a threshold method using statistical distribution quantiles. That is, whenever a new motion feature sample is input into the historical motion buffer, this application regenerates the stimulus trigger condition (dynamic trigger threshold) so that the dynamic trigger threshold is dynamically adjusted according to recent motion feature samples. For example, selecting the dynamic trigger threshold is equivalent to selecting motion feature samples that are in the top 5% of historical motion feature samples.

[0040] When the capacity of the historical motion buffer reaches its limit, the earliest added historical motion feature sample is removed. That is, this application uses a first-in, first-out sliding window update rule for the historical motion buffer. When a new motion feature sample is generated, it is added to the end of the historical motion buffer. When the capacity of the historical motion buffer reaches its limit N, the earliest motion feature sample added to the historical motion buffer is automatically removed to ensure that the historical motion buffer stores the N most recent valid motion feature samples. N can be 3000 to balance covering short training segments with preventing fatigue from overly long periods.

[0041] S6: Compare the time interval with the minimum interval threshold; if the time interval is less than the minimum interval threshold, do not provide neural stimulation to the target person, and update the historical motion buffer based on the motion feature samples; if the time interval is greater than or equal to the minimum interval threshold, use the current moment as the stimulus trigger moment.

[0042] Specifically, when the stimulus triggering condition is met (motion feature sample greater than or equal to the dynamic triggering threshold), the time interval between the current moment and the previous stimulus triggering moment is determined. If the minimum interval threshold is met, i.e. If the current moment is used as the stimulus trigger moment, otherwise the stimulus is not triggered and the historical motion cache continues to be dynamically updated. For time intervals, This is the minimum interval threshold.

[0043] S7: Determine the stimulus triggering command based on the stimulus triggering time; and control the nerve stimulation device to perform nerve stimulation on the target person based on the stimulus triggering command.

[0044] This application does not limit the specific type of motion signal, sampling frequency, or time window length.

[0045] This application modulates the neural activity of a target individual to alter their subsequent motor performance. The altered motor performance is then collected in real time (motor signals) and used to update the historical motor buffer and dynamic trigger threshold. By allowing the stimulus execution results to influence the acquisition and evaluation of the next round of motor signals, this application achieves an adaptive feedback control mechanism based on motor behavior, forming a complete closed-loop neural stimulation control process.

[0046] This application breaks through the traditional method of triggering vagus nerve stimulation in rehabilitation training with fixed time or fixed threshold. It no longer relies on manual judgment or preset static stimulation conditions. Instead, it continuously analyzes the motion signals generated by the target person during rehabilitation training to construct a historical motion buffer reflecting recent motor performance. Based on the degree of improvement of current motor performance relative to recent performance, it adaptively determines the dynamic trigger threshold, thereby achieving effective pairing between neural stimulation and high-quality motor movements.

[0047] This application not only focuses on the technical issue of "when to trigger stimulation," but also comprehensively considers stimulation safety and clinical applicability. By introducing a minimum interval threshold, it avoids the potential risks associated with excessively frequent stimulation. While ensuring a reasonable stimulation sequence, it achieves automation, dynamism, and improved individual adaptability in the neural stimulation triggering process. This application has excellent clinical-friendly characteristics; its input is a conventional motor signal, and the overall processing flow can run in real time during rehabilitation training, reducing reliance on continuous professional intervention and showing promising application prospects for home-based and remote rehabilitation.

[0048] Based on the same inventive concept, this application also provides a closed-loop neuromodulation system for implementing a closed-loop neuromodulation method based on rehabilitation training. The solution provided by this system is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the closed-loop neuromodulation system based on rehabilitation training provided below can be found in the limitations of the closed-loop neuromodulation method based on rehabilitation training described above, and will not be repeated here.

[0049] In one exemplary embodiment, a closed-loop neuromodulation system based on rehabilitation training is provided, comprising: The motion signal acquisition module is used to acquire motion signals generated by the target person during rehabilitation training. The motion change feature value determination module is used to perform smoothing and feature extraction on the motion signal sequentially to obtain motion change feature values; The motion feature sample determination module is used to compare motion change feature values ​​with a minimum motion threshold. If the motion change feature value is less than the minimum motion threshold, the motion change feature value is determined to be invalid, and motion signal acquisition continues. If the motion change feature value is greater than or equal to the minimum motion threshold, the motion change feature value is used as a motion feature sample. The dynamic trigger threshold determination module is used to construct a historical motion buffer based on historical motion feature samples and determine the dynamic trigger threshold based on the threshold method of statistical distribution quantiles. The time interval determination module is used to compare motion feature samples with dynamic trigger thresholds; if the motion feature samples are less than the dynamic trigger thresholds, the historical motion buffer is updated based on the motion feature samples; if the motion feature samples are greater than or equal to the dynamic trigger thresholds, the time interval between the current moment and the last stimulus trigger moment is determined. The stimulus trigger time determination module is used to compare the time interval with the minimum interval threshold. If the time interval is less than the minimum interval threshold, the target person will not be given neural stimulation, and the historical motion buffer will be updated based on the motion feature samples. If the time interval is greater than or equal to the minimum interval threshold, the current time will be used as the stimulus trigger time. The neurostimulation module is used to determine the stimulation trigger command based on the stimulation trigger time; and to control the neurostimulation device to perform neurostimulation on the target person according to the stimulation trigger command.

[0050] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal. The computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is connected to the system bus via the I / O interfaces. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device stores closed-loop neuromodulation data based on rehabilitation training. The I / O interfaces of the computer device are used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a closed-loop neuromodulation method based on rehabilitation training.

[0051] In one exemplary embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps in the embodiment of the closed-loop neuromodulation method based on rehabilitation training.

[0052] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the embodiment of the closed-loop neuromodulation method based on rehabilitation training.

[0053] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in an embodiment of a closed-loop neuromodulation method based on rehabilitation training.

[0054] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0055] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0056] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, etc., and are not limited to these.

[0057] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0058] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A closed-loop neural modulation method based on rehabilitation training, characterized in that, The method includes: Acquire the motion signals generated by the target individuals during rehabilitation training; The motion signal is smoothed and its features are extracted sequentially to obtain motion change feature values; The motion change feature value is compared with the minimum motion threshold. If the motion change feature value is less than the minimum motion threshold, the motion change feature value is determined to be invalid, and motion signal acquisition continues. If the motion change feature value is greater than or equal to the minimum motion threshold, the motion change feature value is used as a motion feature sample. A historical motion buffer is constructed based on historical motion feature samples, and a dynamic trigger threshold is determined based on the threshold method of statistical distribution quantiles. The motion feature sample is compared with the dynamic trigger threshold; if the motion feature sample is less than the dynamic trigger threshold, the historical motion buffer is updated based on the motion feature sample; if the motion feature sample is greater than or equal to the dynamic trigger threshold, the time interval between the current moment and the last stimulus trigger moment is determined. The time interval is compared with the minimum interval threshold; if the time interval is less than the minimum interval threshold, no neural stimulation is applied to the target person, and the historical motion buffer is updated based on the motion feature samples; if the time interval is greater than or equal to the minimum interval threshold, the current moment is used as the stimulation trigger moment. The stimulus triggering command is determined based on the timing of the stimulus triggering; and the nerve stimulation device is controlled to provide nerve stimulation to the target person based on the stimulus triggering command.

2. The closed-loop neuromodulation method based on rehabilitation training according to claim 1, characterized in that, The motion signals include: angle signals, displacement signals, velocity signals, and acceleration signals.

3. The closed-loop neuromodulation method based on rehabilitation training according to claim 1, characterized in that, The process of sequentially smoothing and extracting features from the motion signal to obtain motion change feature values ​​specifically includes: The motion signal is smoothed to obtain a smoothed signal; Within a preset time window, determine the gradient of the smoothed signal and the sequence of the rate of change of the smoothed signal; The motion change characteristic value is obtained by averaging the change rate sequence of the smooth signal.

4. The closed-loop neural modulation method based on rehabilitation training according to claim 1, characterized in that, The process of determining the minimum motion threshold includes: Acquire resting motion signals of the target person at multiple time periods; the resting motion signals are the motion signals generated by the target person in a resting state; Determine the maximum value of the resting motion signal in each time period, and take the average of multiple maximum values ​​to obtain the minimum motion threshold.

5. The closed-loop neuromodulation method based on rehabilitation training according to claim 4, characterized in that, The process of constructing a historical motion buffer based on historical motion feature samples and determining a dynamic trigger threshold based on a threshold method using statistical distribution quantiles specifically includes: The historical motion feature samples in the historical motion buffer are sorted in a preset order to obtain a motion feature sample sequence; Using pre-set quantiles, a dynamic trigger threshold is selected from the motion feature sample sequence.

6. The closed-loop neuromodulation method based on rehabilitation training according to claim 4, characterized in that, If the motion feature sample is less than the dynamic trigger threshold, the historical motion cache is updated based on the motion feature sample, specifically including: Motion feature samples are added to the historical motion cache, and the dynamic trigger threshold is updated based on the threshold method of statistical distribution quantiles. When the capacity of the historical motion buffer reaches its limit... Remove the earliest added historical motion feature samples.

7. A closed-loop neuromodulation system based on rehabilitation training, used to implement the closed-loop neuromodulation method based on rehabilitation training as described in any one of claims 1-6, characterized in that, The system includes: The motion signal acquisition module is used to acquire motion signals generated by the target person during rehabilitation training. The motion change feature value determination module is used to perform smoothing and feature extraction on the motion signal sequentially to obtain motion change feature values; The motion feature sample determination module is used to compare motion change feature values ​​with a minimum motion threshold. If the motion change feature value is less than the minimum motion threshold, the motion change feature value is determined to be invalid, and motion signal acquisition continues. If the motion change feature value is greater than or equal to the minimum motion threshold, the motion change feature value is used as a motion feature sample. The dynamic trigger threshold determination module is used to construct a historical motion buffer based on historical motion feature samples and determine the dynamic trigger threshold based on the threshold method of statistical distribution quantiles. The time interval determination module is used to compare motion feature samples with dynamic trigger thresholds; if the motion feature samples are less than the dynamic trigger thresholds, the historical motion buffer is updated based on the motion feature samples; if the motion feature samples are greater than or equal to the dynamic trigger thresholds, the time interval between the current moment and the last stimulus trigger moment is determined. The stimulus trigger time determination module is used to compare the time interval with the minimum interval threshold. If the time interval is less than the minimum interval threshold, the target person will not be given neural stimulation, and the historical motion buffer will be updated based on the motion feature samples. If the time interval is greater than or equal to the minimum interval threshold, the current time will be used as the stimulus trigger time. The neurostimulation module is used to determine the stimulation trigger command based on the stimulation trigger time; and to control the neurostimulation device to perform neurostimulation on the target person according to the stimulation trigger command.

8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the closed-loop neuromodulation method based on rehabilitation training as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the closed-loop neuromodulation method based on rehabilitation training as described in any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the closed-loop neuromodulation method based on rehabilitation training as described in any one of claims 1-6.

Images