An intelligent auxiliary customization parameter setting method and system

Abstract

The application relates to an intelligent auxiliary customized parameter setting method and system, the method comprising the following steps: S1, setting a stimulation device at a target position; the stimulation device is used for applying a stimulation signal to the target position based on a stimulation parameter; S2, collecting a feedback signal in real time and extracting a feedback feature; obtaining a customized stimulation parameter based on the feedback feature; repeatedly executing the step until an iteration stop condition or an optimization stop condition is met; and sending the customized stimulation parameter to the stimulation device. The optimization process is limited within a logical range, random optimization is performed within the logical space range constructed by the stimulation parameter sample, the customized parameter setting mode can be searched between the degrees of freedom and prior experience, and the customized efficiency is improved.

Description

[Technical Field]

[0001] This invention belongs to the field of smart healthcare technology, and in particular relates to an intelligent assisted customized parameter setting method and system. [Background Technology]

[0002] With the increasing aging population and growing emphasis on health, the need to improve medical technology and patient experience through next-generation information technology has become increasingly urgent, giving rise to the concept of "smart healthcare." Smart healthcare enables the digitalization, intelligentization, and automation of medical management and services, and has a wide range of application scenarios.

[0003] The rapid development and maturation of technologies such as cloud computing, IoT, mobile internet, big data, AI, 5G networks, and blockchain are empowering the digitalization and intelligentization of healthcare, laying the technological foundation for smart healthcare. The development of mobile internet technology allows healthcare services to be accessed anytime, anywhere through smart devices, improving the convenience of these services. IoT technology enables medical devices and sensors to connect, achieving remote monitoring and real-time data collection, providing fundamental data support for smart healthcare. Big data technology, through the analysis and mining of massive amounts of medical data, provides data support for disease prevention, diagnosis, and treatment, improving the accuracy of healthcare services. Artificial intelligence technology is widely used in smart healthcare, including medical image analysis, assisted diagnosis, case analysis, and decision support, improving the intelligence level of healthcare services. The high speed and low latency of 5G networks provide technical support for telemedicine, surgical robots, and remote medical education, making healthcare services more efficient. The development and application of these technologies have provided strong technical support for smart healthcare, promoting the digitalization, intelligentization, and networking of medical services, and greatly improving the quality and efficiency of medical services. These technologies enable the effective utilization of remote computing power and extensive data during assisted procedures.

[0004] Neurostimulation therapy can be used to treat a variety of diseases. For example, deep brain stimulation (DBS) is an effective treatment for Parkinson's disease, essential tremor, dystonia, and obsessive-compulsive disorder; vagus nerve stimulation (VNS) can be used to treat epilepsy and suppress seizures; similarly, there are spinal cord stimulation and sacral nerve stimulation therapies. When applying stimulation, external devices are needed to adjust stimulation parameters to achieve different effects. For example, adjusting the polarity of the contact points changes the stimulation location, and modifying the amplitude, pulse width, and frequency changes the range of influence. Customized parameter setting is a crucial step in neurostimulation, involving adjusting stimulation parameters according to the patient's specific condition and needs to achieve the best results. Current parameter setting methods are often based on prior experience or completely random trials based on prior experience, resulting in poor user experience, low optimization efficiency, and low customization efficiency. Improving the efficiency of customized parameter setting is a technical problem to be solved. To address the problems in the prior art, this invention limits the optimization process to a logical scope, performing random optimization within the logical space constructed by the stimulus parameter samples. This allows for the search of customized parameter setting methods between degrees of freedom and prior experience, thereby improving the efficiency of customized parameter setting. [Summary of the Invention]

[0005] To address the aforementioned problems in the prior art, this invention proposes an intelligent assisted customized parameter setting method and system, the method comprising:

[0006] Step S1: Set the stimulation device at the target location; the stimulation device is used to apply a stimulation signal to the target location based on stimulation parameters;

[0007] Step S2: Acquire feedback signals in real time and extract feedback features; obtain customized stimulation parameters based on feedback features; repeat this step until the iteration cutoff condition or optimization cutoff condition is met; send the customized stimulation parameters to the stimulation device.

[0008] The process of obtaining customized stimulation parameters based on feedback features specifically includes the following steps:

[0009] Step S21: Obtain N stimulus parameter samples; specifically: obtain N stimulus parameter samples that match the stimulus object attributes based on the stimulus object attributes; wherein: the stimulus parameter samples contain K types of stimulus parameters; n is the stimulus parameter sample number, k is the stimulus parameter type number; k = 1 to K; n = 1 to N;

[0010] Step S22: Select 3 stimulus parameter samples from the stimulus parameter samples; determine whether the iteration cutoff condition is met. If yes, end step S2; otherwise, proceed to the next step.

[0011] The iteration cutoff condition is: the number of times the three collected parameter samples have been repeatedly taken exceeds the selection count threshold;

[0012] The step of selecting 3 stimulus parameter samples from N stimulus parameter samples specifically means: selecting 3 most similar stimulus parameter samples from N stimulus parameter samples;

[0013] Step S23: Obtain any two combinations of unprocessed stimulus parameter types, and the two stimulus parameter types in the combination are called the first parameter type k1 and the second parameter type k2, respectively; construct a two-dimensional coordinate plane using the two stimulus parameter types;

[0014] Step S24: Obtain the stimulus parameters of the first parameter type k1 and the second parameter type k2 from the three stimulus parameter samples respectively to construct the corresponding stimulus parameter combinations. The stimulus parameter combinations corresponding to the three stimulus parameter samples are respectively called the first combination U1 = (u1 k1 u1 k2 ), the second combination U2=(u2 k1 u2 k2 ), the third combination U3=(u3 k1 u3 k2 Place the three stimulus parameters into a two-dimensional coordinate plane; in a fully random manner, randomly select a point within the triangular plane constructed by the three stimulus parameters, and use the horizontal and vertical coordinate values ​​corresponding to this random point as the current stimulus parameter values ​​for the first parameter type and the second parameter type, respectively; obviously, the fully random method here is also fully random within the logical space, rather than random trial;

[0015] The step of randomly selecting a point within a triangular plane constructed from the three stimulus parameters involves: generating two random numbers r1 and r2 between [0, 1]; and calculating the random point Ux = (ux) within the triangular plane using the following formulas (1) and (2). k1 ux k2 );

[0016] ux k1 =u1 k1 +r1(u2 k1 -u1 k1 )+r2(u3 k1 -u1 k1 (1);

[0017] ux k2 =u1 k2 +r1(u2 k2 -u1 k2 )+r2(u3 k2 -u1 k2 (2);

[0018] Step S25: Determine if there are still unprocessed stimulus parameter types. If yes, return to step S23; otherwise, proceed to the next step.

[0019] Step S26: Apply a stimulus signal to the target location based on the current stimulus parameter values ​​for all stimulus parameter types; collect feedback signals in real time and extract feedback features; determine the degree of matching between the feedback features and the target feedback features; if the degree of matching is greater than that of the quasi-stimulation parameter, replace the quasi-stimulation parameter with the current stimulus parameter;

[0020] Step S27: Determine whether the optimization cutoff condition is met. If so, use the quasi-stimulation parameter as the customized stimulation parameter and send the customized stimulation parameter to the stimulation device; otherwise, return to step S22.

[0021] The optimization cutoff condition is that the stimulus target is reached or the feedback characteristics meet the target feedback characteristic requirements.

[0022] Furthermore, the stimulation device includes stimulation electrodes, a stimulator, and a stimulation controller.

[0023] Furthermore, the stimulating electrodes are placed on the skin and connected to the stimulator via wires.

[0024] Furthermore, N≥9.

[0025] Furthermore, the stimulation controller is connected to the stimulator via wired or wireless means, thereby enabling control of the stimulator and its stimulation electrodes.

[0026] Furthermore, in the wireless mode, communication between the stimulator and the stimulation controller is achieved through radio frequency transmission.

[0027] An intelligent assisted customized parameter setting system is provided, wherein the intelligent assisted customized parameter setting system is used to implement the above-mentioned intelligent assisted customized parameter setting method.

[0028] An intelligent auxiliary customized parameter setting device is provided, which is used to implement the above-mentioned intelligent auxiliary customized parameter setting method.

[0029] An intelligent assisted customized parameter setting platform is provided, which is used to implement the above-mentioned intelligent assisted customized parameter setting method.

[0030] A smart auxiliary customized parameter setting server, characterized in that the smart auxiliary customized parameter setting server is used to implement the above-mentioned smart auxiliary customized parameter setting method.

[0031] The beneficial effects of this invention include:

[0032] (1) Limit the optimization method to three similar stimulus parameter samples or cluster range, limit the optimization process to the logical range, and perform random optimization within the logical space constructed by the stimulus parameter samples. This allows for searching customized parameter setting methods between degrees of freedom and prior experience, improving the efficiency of customized parameter setting while exploring the optimization space to achieve the goal of optimal customization.

[0033] (2) By guiding the stimulation parameters to be randomly optimized within any two types of dimensions through a semi-random or semi-random plus fully random method, and semi-randomly optimized between any two types of dimensions, the stimulation parameters are subject to random constraints between the types of stimulation parameters during the optimization process. While maintaining the use of known information, it can also explore the numerical space that has not been fully explored in other stimulation parameter spaces, improve the efficiency of customized parameter setting, effectively assist in improving the stimulation method of the stimulation device, and improve its applicability in different scenarios. [Attached Image Description]

[0034] The accompanying drawings, which are provided to further illustrate the invention and form part of this application, are not intended to unduly limit the invention. In the drawings:

[0035] Figure 1 This is a schematic diagram of the intelligent auxiliary customized parameter setting method provided by the present invention.

Detailed Implementation Methods

[0036] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. The illustrative embodiments and descriptions are only used to explain the present invention and are not intended to limit the present invention.

[0037] This invention proposes an intelligent assisted customized parameter setting method and system, as shown in the appendix. Figure 1 As shown, the method includes the following steps:

[0038] Step S1: Set the stimulation device at the target location; the stimulation device applies a stimulation signal to the target location based on stimulation parameters; wherein; the stimulation device includes stimulation electrodes, a stimulator, and a stimulation controller;

[0039] Preferred: The target location belongs to or is associated with the stimulus object;

[0040] Preferably: the stimulating electrode is placed on the skin and connected to the stimulator via a wire; wherein: the electrode is used to transmit a stimulating current to excitable tissue, causing muscle contraction; the stimulator is used to generate a current to the electrode; and the stimulation controller adjusts the stimulation parameters according to the patient's instructions and feedback signals.

[0041] Preferably, the stimulation controller is connected to the stimulator via wired or wireless means to control the stimulator and its electrodes; in the wireless mode, communication between the stimulator and the stimulation controller is achieved through radio frequency transmission.

[0042] Preferably, the electrodes are in one or more groups; each group includes a stimulating electrode (cathode) placed on the motor point of the peripheral nerve or muscle to be stimulated; and a reference electrode (anode) placed on tissue that is far from the stimulating electrode and difficult to excite, or placed around the stimulating electrode, so as to confine the electric field to a local area to achieve higher stimulation selectivity.

[0043] Preferably, the stimulation parameters include one or more of the following: stimulation waveform, stimulation frequency, stimulation amplitude, stimulation pulse width, on / off cycle and / or stimulation duration;

[0044] The stimulation waveform includes asymmetric biphasic waves and symmetrical biphasic waves; wherein: the asymmetric biphasic wave consists of waves with different pulse widths or amplitudes and opposite phases; the symmetrical biphasic wave consists of two pulse waves with the same pulse width and amplitude but opposite phases, one of which generates the stimulation effect, and the other is equivalent to a balance wave; the stimulation frequency is the number of pulses generated per second during the stimulation process, in Hz; the stimulation amplitude is the current intensity, in mA; the stimulation pulse width is the time span of a single pulse, that is, the pulse duration;

[0045] Preferably, the default range of the stimulation frequency is 15–40 Hz; theoretically, the higher the frequency, the greater the muscle contraction effect, but the neurotransmitters at the neuromuscular junction will be depleted more quickly, thus easily causing fatigue; while too low a frequency cannot stimulate the muscle to produce a sufficiently strong contraction, thus causing low-frequency fatigue; the default range of the stimulation amplitude is 1–30 mA; the default range of the stimulation pulse width is 200–500 μs; when the stimulation pulse width is small, a very high amplitude is required to achieve sufficient depolarization to contract the muscle; while when the stimulation pulse width is wide, it can usually penetrate the subcutaneous tissue more deeply, increasing the number of activated nerve fibers, thus easily causing a more obvious muscle contraction response; the on / off cycle is the sum of the on-time and off-time; continuous electrical stimulation will cause muscle fatigue, thus leading to a decrease in muscle contraction force;

[0046] Step S2: Acquire feedback signals in real time and extract feedback features; obtain customized stimulation parameters based on feedback features; repeat this step until the iteration cutoff condition or optimization cutoff condition is met; send the customized stimulation parameters to the stimulation device.

[0047] The process of obtaining customized stimulation parameters based on feedback features specifically includes the following steps:

[0048] Step S21: Obtain stimulus parameter samples; specifically: obtain N stimulus parameter samples that match the stimulus object attributes; wherein: the stimulus parameter samples contain K types of stimulus parameters; sp n，k For the nth stimulus parameter sample, the parameter is of type k; k = 1 to K; n = 1 to N;

[0049] Preferred: N≥9;

[0050] Preferred method: Based on prior experience, obtain a sample of stimulus parameters that match the attributes of the stimulus object; Further method: Obtain a stimulus strategy that matches the stimulus object, extract stimulus parameters from the stimulus strategy, and organize the stimulus parameters in order to obtain a sample of stimulus parameters; After completing the customization of stimulus parameters, fill the customized stimulus parameters into the stimulus strategy to obtain a customized stimulus strategy.

[0051] Preferred method: Based on the stimulus object attributes, query a preset stimulus object attribute and stimulus parameter comparison table to obtain a sample of stimulus parameters that match the stimulus object attributes;

[0052] Step S22: Select 3 stimulus parameter samples from N stimulus parameter samples; determine whether the iteration cutoff condition is met. If yes, end step S2; otherwise, proceed to the next step.

[0053] The iteration cutoff condition is: the three collected parameter samples have been repeatedly taken, or the number of times any (any two) of the three collected parameter samples have been repeatedly taken exceeds the selection count threshold (or the proportion of repeated selections in the selection count exceeds the ratio of the selection count to N); if the iteration cutoff condition is met but the optimization cutoff condition is not met, the optimal solution is not obtained.

[0054] Preferred method: Select a threshold value for the number of times; for example, equal to N / 5 to N / 2.

[0055] The step of selecting 3 stimulus parameter samples from N stimulus parameter samples specifically involves selecting the 3 most similar stimulus parameter samples from the N stimulus parameter samples; Euclidean distance can be used to determine the degree of similarity.

[0056] Alternative: The step of selecting 3 stimulus parameter samples from N stimulus parameter samples specifically means: selecting 3 stimulus parameter samples whose similarity is within the similarity limit from N stimulus parameter samples;

[0057] Alternative: The step of selecting 3 stimulus parameter samples from N stimulus parameter samples specifically involves: clustering the N stimulus parameter samples to obtain M clusters; randomly selecting one cluster and then randomly selecting 3 stimulus parameter samples from it; In existing technologies, when performing customized optimization, aimless optimization is often used, and the optimization results can easily cross the inherent logical boundaries of the stimulus parameters, causing random values ​​to be in a non-logical region. Limiting the optimization method to three similar stimulus parameter samples or cluster ranges, and limiting the optimization process to a logical range, allows for random optimization within the logical space constructed by the stimulus parameter samples. This can improve customization efficiency while exploring the optimization space to achieve the goal of optimal customization.

[0058] Preferred: M < N;

[0059] Step S23: Obtain any two combinations of unprocessed stimulus parameter types, and the two stimulus parameter types in the combination are called the first parameter type k1 and the second parameter type k2, respectively; construct a two-dimensional coordinate plane using the two stimulus parameter types;

[0060] The construction of a two-dimensional coordinate plane using the two stimulus parameter types specifically involves using the first parameter type as the horizontal axis and the second parameter type as the vertical axis.

[0061] Alternatively: The construction of a two-dimensional coordinate plane using the two stimulus parameter types specifically involves using the second parameter type as the horizontal axis and the first parameter type as the vertical axis;

[0062] Furthermore, before constructing the two-dimensional coordinate plane, the stimulus parameter types are subjected to consistent changes such as numerical scaling to ensure that the values ​​of the two stimulus parameter types are compatible.

[0063] Preferred: When the number of stimulus parameter types is odd, for the last stimulus parameter type, its stimulus parameter value is set to the average value (or minimum, maximum, or other feature value) of the corresponding stimulus parameter values ​​among the three stimulus sample parameters;

[0064] Step S24: Obtain the stimulus parameters of the first parameter type k1 and the second parameter type k2 from the three stimulus parameter samples respectively to construct the corresponding stimulus parameter combinations. The stimulus parameter combinations corresponding to the three stimulus parameter samples are respectively called the first combination U1 = (u1 k1 u1 k2 ), the second combination U2=(u2 k1 u2 k2 ), the third combination U3=(u3 k1 u3 k2Place the three stimulus parameters into a two-dimensional coordinate plane; randomly select a point in the triangular plane constructed by the three stimulus parameters using a fully random method, and use the horizontal and vertical coordinate values ​​corresponding to the random point as the current stimulus parameter values ​​for the first parameter type and the second parameter type, respectively.

[0065] The step of randomly selecting a point within a triangular plane constructed from the three stimulus parameters involves: generating two random numbers r1 and r2 between [0, 1]; and calculating the random point Ux = (ux) within the triangular plane using the following formulas (1) and (2). k1 ux k2 );

[0066] ux k1 =u1 k1 +r1(u2 k1 -u1 k1 )+r2(u3 k1 -u1 k1 (1);

[0067] ux k2 =u1 k2 +r1(u2 k2 -u1 k2 )+r2(u3 k2 -u1 k2 (2);

[0068] Preferred method: When entering this step for the first time, a fully random method is used; when entering this step in subsequent times, a semi-random method or a combination of semi-random and fully random methods is used.

[0069] The semi-random method specifically involves: first, calculating the first random point constraint (α1, β1, γ1) of the previously acquired random point within the triangular plane; for the currently acquired three stimulus parameter samples, calculating the centroid coordinates within the triangular plane constructed from the combination of the three stimulus parameters based on the first random point constraint; and then calculating the centroid coordinates (uv... k1 uv k2 The corresponding horizontal and vertical coordinate values ​​are used as the current stimulus parameter values ​​for the first parameter type and the second parameter type, respectively.

[0070] The first step is to calculate the first random point constraint (α1, β1, γ1) of the previously obtained random point in the triangular plane; specifically, the first random point constraint (α1, β1, γ1) is calculated using the following formulas (3) to (10);

[0071]

[0072]

[0073] α1+β1+γ1=1, α1, β1, γ1>0 (10);

[0074] For the three stimulus parameter samples currently acquired, the centroid coordinates are calculated within the triangular plane constructed by the combination of the three stimulus parameters based on the first random point constraint; specifically, the centroid coordinates are calculated based on the following formulas (11) and (12);

[0075] UV k1 =α1u1 k1 +β1u2 k1 +γ1u3 k1 (11);

[0076] UV k2 =α1u1 k2 +β1u2 k2 +γ1u3 k2 (12);

[0077] The semi-random plus fully random method is as follows: First, the centroid coordinates are calculated using a semi-random method; then, using a fully random method, for the three stimulus parameter samples currently obtained, a random point is randomly selected within the triangular plane constructed by the combination of the three stimulus parameters; a second random point is selected within the straight line range formed by the centroid coordinates and the random point, and the horizontal and vertical coordinate values ​​corresponding to the second random point are used as the current stimulus parameter values ​​for the corresponding first parameter type and the corresponding second parameter type, respectively.

[0078] The centroid coordinates (uv) k1 uv k2 ) and random points (ux) k1 ux k2 Select a second random point (ux2) within the range of the straight line formed by the two points. k1 ux2 k2 Specifically, this involves generating a random number r between [0, 1]; and calculating the second random point (ux2) based on the following equations (13) and (14). k1 ux2 k2 );

[0079] ux2 k1 =uv k1 +r(ux k1 -uv k1 (13);

[0080] ux2 k2 =uv k2 +r(ux k2 -uv k2 (14);

[0081] Step S25: Determine if there are still unprocessed stimulus parameter types. If yes, return to step S23; otherwise, proceed to the next step.

[0082] Step S26: Apply a stimulus signal to the target location based on the current stimulus parameter values ​​for all stimulus parameter types; collect feedback signals in real time and extract feedback features; determine the current matching degree between the feedback features and the target feedback features; if the current matching degree is greater than the quasi-stimulation parameter, replace the quasi-stimulation parameter with the current stimulus parameter;

[0083] Preferred setting: Set the default value of the quasi-stimulus parameter to the stimulus parameter value corresponding to any stimulus parameter sample;

[0084] Alternative: Set the default value of the quasi-stimulus parameter to the stimulus parameter value corresponding to the stimulus parameter sample that best matches the stimulus object attribute;

[0085] Preferred approach: Develop a stimulus strategy that matches the target stimulus based on the current stimulus parameter values ​​for all stimulus parameter types, and apply the stimulus signal to the target location based on the stimulus strategy;

[0086] Preferred method: Apply a stimulation signal to the target location based on quasi-stimulation parameters through simulation, and obtain the simulated value of the feedback signal as the feedback signal;

[0087] Step S27: Determine whether the optimization cutoff condition is met. If so, use the quasi-stimulation parameter as the customized stimulation parameter and send the customized stimulation parameter to the stimulation device; otherwise, return to step S22.

[0088] Preferably, the optimization cutoff condition is that the stimulus target is achieved and the feedback characteristics meet the target feedback characteristic requirements;

[0089] Alternative: The optimization cutoff condition is that the current matching degree corresponding to the current stimulus parameter value is greater than or equal to the target matching degree, then the optimization cutoff condition is satisfied.

[0090] Preferably, after receiving customized stimulation parameters, the stimulation device saves the customized stimulation parameters in the stimulation controller;

[0091] Preferably, the feedback signal includes one or a combination of brain signals acquired by non-invasive methods (such as electroencephalography, EEG), visual feedback, proprioceptive feedback, electrotactile stimulation, motor-related potentials (ERPs), and motor evoked potentials (MEPs); furthermore, an adaptive feature extraction method is adopted for each type of signal, such as: time-domain analysis, frequency-domain analysis, time-frequency analysis, entropy feature extraction, monitoring changes in brain activity caused by visual stimulation, monitoring changes in muscle activity or joint position, monitoring changes in skin potential or brain activity caused by electrical stimulation, monitoring changes in brain activity caused by specific motor tasks or motor imagery, monitoring changes in muscle activity caused by transcranial magnetic stimulation or electrical stimulation, Fourier transform (FFT), etc.

[0092] Wherein: the visual feedback is provided through a display or other visual display device; the proprioceptive feedback is provided through passive hand movements stimulated by functional electrical stimulation (FES); the electrotactile stimulation is provided through electrotactile stimulation of the tongue; the motor evoked potential (MEP) feedback is through MEP evoked by transcranial magnetic stimulation; when multiple feedback types are selected in combination, it is a multimodal feedback method, which can provide more comprehensive feedback information; preferably: the stimulation parameters are converted into a format and included in the control command, which is sent to the functional electrical stimulation (FES) device to drive the electrodes to stimulate the target location;

[0093] Based on the same inventive concept, the present invention also provides an intelligent assisted customized parameter setting system, the system being used to implement the above-mentioned intelligent assisted customized parameter setting method;

[0094] Based on the same inventive concept, the present invention also provides an intelligent assisted customized parameter setting server, the server being used to implement the above-mentioned intelligent assisted customized parameter setting method;

[0095] A computer program (also referred to as a program, software, software application, script, or code) can be written in any form of programming language, including assembly or interpreted languages, declarative or procedural languages, and can be deployed in any form, including as a standalone program or as a module, component, subroutine, object, or other unit suitable for use in a computing environment. A computer program may, but does not necessarily, correspond to a file in a file system. A program can be stored as part of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to said program, or in multiple co-located files (e.g., a file storing one or more modules, subroutines, or code portions). A computer program can be deployed to execute on a single computer or on multiple computers located at a single site or distributed across multiple sites and interconnected by a communications network.

[0096] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0097] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0098] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0099] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for intelligent assisted customized parameter setting, characterized in that, The method includes: Step S1: Set the stimulation device at the target location; the stimulation device is used to apply a stimulation signal to the target location based on stimulation parameters; Step S2: Real-time acquisition of feedback signals and extraction of feedback features; Obtain customized stimulation parameters based on feedback characteristics; repeat this step until the iteration cutoff condition or optimization cutoff condition is met; send the customized stimulation parameters to the stimulation device. The process of obtaining customized stimulation parameters based on feedback features specifically includes the following steps: Step S21: Obtain N stimulus parameter samples; specifically: obtain N stimulus parameter samples that match the stimulus object attributes based on the stimulus object attributes; wherein: the stimulus parameter samples contain K types of stimulus parameters; n is the stimulus parameter sample number, k is the stimulus parameter type number; k = 1 to K; n = 1 to N; Step S22: Select 3 stimulus parameter samples from the stimulus parameter samples; determine whether the iteration cutoff condition is met. If yes, end step S2; otherwise, proceed to the next step. The iteration cutoff condition is: the number of times the three collected parameter samples have been repeatedly taken exceeds the selection count threshold; The step of selecting 3 stimulus parameter samples from N stimulus parameter samples specifically means: selecting 3 most similar stimulus parameter samples from N stimulus parameter samples; Step S23: Obtain any two combinations of unprocessed stimulus parameter types, and the two stimulus parameter types in the combination are called the first parameter type k1 and the second parameter type k2, respectively; construct a two-dimensional coordinate plane using the two stimulus parameter types; Step S24: Obtain the stimulus parameters of the first parameter type k1 and the second parameter type k2 from the three stimulus parameter samples respectively to construct the corresponding stimulus parameter combinations. The stimulus parameter combinations corresponding to the three stimulus parameter samples are respectively called the first combination U1 = (u1 k1 u1 k2 ), the second combination U2=(u2 k1 u2 k2 ), the third combination U3=(u3 k1 u3 k2 Place the three stimulus parameters into a two-dimensional coordinate plane; randomly select a point within the triangular plane constructed by the three stimulus parameters, and use the horizontal and vertical coordinate values ​​corresponding to the random point as the current stimulus parameter values ​​for the first parameter type and the second parameter type, respectively. The step of randomly selecting a point within a triangular plane constructed from the three stimulus parameters involves: generating two random numbers r1 and r2 between [0, 1]; and calculating the random point Ux = (ux) within the triangular plane using the following formulas (1) and (2). k1 ux k2 ); ux k1 =u1 k1 +r1(u2 k1 -u1 k1 )+r2(u3 k1 -u1 k1 ) (1); ux k2 =u1 k2 +r1(u2 k2 -u1 k2 )+r2(u3 k2 -u1 k2 ) (2); Step S25: Determine if there are still unprocessed stimulus parameter types. If yes, return to step S23; otherwise, proceed to the next step. Step S26: Apply a stimulus signal to the target location based on the current stimulus parameter values ​​for all stimulus parameter types; collect feedback signals in real time and extract feedback features; determine the degree of matching between the feedback features and the target feedback features; if the degree of matching is greater than that of the quasi-stimulation parameter, replace the quasi-stimulation parameter with the current stimulus parameter; Step S27: Determine whether the optimization cutoff condition is met. If so, use the quasi-stimulation parameter as the customized stimulation parameter and send the customized stimulation parameter to the stimulation device; otherwise, return to step S22. The optimization cutoff condition is that the stimulus target is reached or the feedback characteristics meet the target feedback characteristic requirements.

2. The intelligent assisted customized parameter setting method according to claim 1, characterized in that, The stimulation device includes stimulation electrodes, a stimulator, and a stimulation controller.

3. The intelligent assisted customized parameter setting method according to claim 2, characterized in that, The stimulating electrode is placed on the skin and connected to the stimulator via a wire.

4. The intelligent assisted customized parameter setting method according to claim 3, characterized in that, N≥9。 5. The intelligent assisted customized parameter setting method according to claim 4, characterized in that, The stimulation controller is connected to the stimulator via wired or wireless means, thereby enabling control of the stimulator and its stimulation electrodes.

6. The intelligent assisted customized parameter setting method according to claim 5, characterized in that, In the wireless method, communication between the stimulator and the stimulation controller is achieved through radio frequency transmission.

7. An intelligent assisted customized parameter setting system, characterized in that, The intelligent assisted customized parameter setting system is used to implement the intelligent assisted customized parameter setting method according to any one of claims 1-6.

8. An intelligent assisted customized parameter setting device, characterized in that, The intelligent assisted customized parameter setting device is used to implement the intelligent assisted customized parameter setting method according to any one of claims 1-6.

9. An intelligent assisted customized parameter design platform, characterized in that, The intelligent assisted customized parameter setting platform is used to implement the intelligent assisted customized parameter setting method according to any one of claims 1-6.

10. An intelligent assisted customized parameter setting server, characterized in that, The intelligent auxiliary customized parameter setting server is used to implement the intelligent auxiliary customized parameter setting method according to any one of claims 1-6.

Images