Programmable parameter configuration method and related apparatus

Abstract

A programmable parameter configuration method and a related apparatus. The method comprises: acquiring image data of a user (S1); preprocessing the image data of the user to obtain a preprocessing result, and obtaining a recommended programmable parameter on the basis of the preprocessing result (S2); and on the basis of a parameter compensation model, using the recommended programmable parameter to obtain an optimal programmable parameter (S3).

Description

Program-controlled parameter configuration method and related device

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on March 20, 2024, with application number 202410316865.2. The entire contents of the above application are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of medical device technology, for example, to a program-controlled parameter configuration method and related devices. Background Art

[0003] Deep brain stimulation (DBS) is an invasive neuromodulation technology that implants stimulating electrodes in specific neural structures of the human brain and uses a neurostimulator to deliver weak electrical pulses to change the electrical activity and function of brain neural circuits and networks in order to control and improve patient symptoms.

[0004] In current DBS treatments, therapeutic efficacy and adverse reactions are closely related to the distribution of the volume of tissue activated (VTA) near the electrode contacts. Ideally, the VTA should overlap closely with the target area to achieve optimal therapeutic efficacy and minimize adverse reactions. However, current programming methods cannot accurately predict and determine the VTA, making it difficult for physicians to select initial programming parameters. Physicians typically must try different combinations of electrode contacts and stimulation parameters to find the optimal programming parameters. However, this programming approach presents several challenges. First, initial programming parameters are difficult to determine, often requiring trial adjustments based on physician experience. Second, the programming process is time-consuming, typically taking several hours. This not only increases the physician's burden but can also lead to patient fatigue and inattention, impairing their observation and judgment. If a patient does not achieve optimal therapeutic effect after a single programming session, subsequent programming sessions may become more frequent, requiring multiple visits to the hospital and causing inconvenience. In addition, due to the huge number of possible contact and stimulation parameter combinations, it is impossible to try and evaluate them one by one. It is impossible to search all possible programming parameter combinations based on the doctor's experience and attempts; often the programming parameters can only be determined based on limited attempts, which may result in the determined programming parameters not being the optimal parameters for the patient. Summary of the Invention

[0005] The present application provides a programming parameter configuration method and related devices to solve the problems existing in three-dimensional visualization programming, such as the difficulty in determining the initial programming parameters, the long programming time, and the inability to search all possible programming parameter combinations based on the doctor's experience and attempts. Therefore, the programming parameters determined after a limited number of attempts may not be the optimal programming parameters for the patient.

[0006] In a first aspect, the present application provides a method for configuring programmable parameters for an in vivo implantable neurostimulator, the method comprising:

[0007] Obtain user's image data;

[0008] Preprocessing the user's image data to obtain a preprocessing result, and obtaining recommended program control parameters based on the preprocessing result;

[0009] Based on the parameter compensation model, the recommended program control parameters are used to obtain optimal program control parameters.

[0010] Optionally, the preprocessing of the user's image data to obtain a preprocessing result, and obtaining recommended program control parameters based on the preprocessing result, includes:

[0011] Segmenting the user's image data to obtain a target area;

[0012] Obtaining an electric field intensity scalar field of initial program-controlled parameters in three-dimensional space, and calculating an overlap rate between the electric field intensity scalar field and the target area;

[0013] By iteratively calculating the initial program control parameters, the overlap rate corresponding to the initial program control parameters in each iteration is obtained, and the highest overlap rate is obtained based on multiple overlap rates obtained in multiple iterations. Based on the highest overlap rate, the recommended program control parameters are obtained.

[0014] Optionally, obtaining the electric field intensity scalar field of the initial program-controlled parameters in three-dimensional space and calculating the overlap rate between the electric field intensity scalar field and the target area includes:

[0015] Segmenting the user's image data using a segmentation algorithm to obtain segmentation results of different tissues;

[0016] Orderly construct the contact array of the neurostimulator;

[0017] Taking initial program-controlled parameters as input, the contact array and the segmentation result as first objective function input, and obtaining the electric field intensity scalar field of the initial program-controlled parameters in three-dimensional space;

[0018] The coincidence rate is obtained by calculating the integral of the electric field intensity scalar field in the target area.

[0019] Optionally, the method for constructing the parameter compensation model includes:

[0020] Obtain multiple sets of actual program control parameters from actual treatment records;

[0021] Using each set of actual program control parameters and the corresponding recommended program control parameters, an error result corresponding to the set of actual program control parameters is obtained;

[0022] A parameter compensation model is constructed and iterative optimization is performed based on the multiple error results corresponding to the multiple groups of actual program-controlled parameters.

[0023] Optionally, constructing a parameter compensation model and performing iterative optimization based on multiple error results corresponding to the multiple sets of actual program-controlled parameters includes:

[0024] Obtaining the usage time of each set of actual program-controlled parameters;

[0025] Obtaining the weight of the error result corresponding to the set of actual program-controlled parameters according to the usage time;

[0026] The weights of the multiple error results are used as inputs of the parameter compensation model to obtain a final parameter compensation output result.

[0027] Optionally, the usage time of the actual program-controlled parameters includes the continuous usage time of each group of actual program-controlled parameters or the accumulated usage time of each group of actual program-controlled parameters.

[0028] Optionally, the program-controlled parameter configuration method further includes:

[0029] A plurality of first errors are obtained using the multiple sets of actual program parameters, the recommended program parameters corresponding to each set of actual program parameters, and a first objective function; wherein the input of the first objective function includes the actual program parameters, the recommended program parameters, the contact array, and the segmentation result; the first error is the difference between the first relative area and the second relative area; the first relative area is obtained based on the actual program parameters, the user's brain segmentation result, and the contact array and the first objective function; the second relative area is obtained based on the recommended program parameters, the user's brain segmentation result, and the contact array and the first objective function;

[0030] A first parameter compensation model is constructed using the multiple first errors and iterative optimization is performed.

[0031] Optionally, the program-controlled parameter configuration method further includes:

[0032] Using the multiple sets of actual program-controlled parameters and the recommended program-controlled parameters corresponding to each set of actual program-controlled parameters, a plurality of second errors are obtained; the second errors are direct differences between the actual program-controlled parameters and the corresponding recommended program-controlled parameters;

[0033] A second parameter compensation model is constructed using the multiple second errors and iteratively optimized.

[0034] Optionally, obtaining optimal program control parameters by using the recommended program control parameters based on the parameter compensation model includes:

[0035] Based on the parameter compensation model, obtaining a parameter compensation value;

[0036] The optimal program control parameters are obtained through the recommended program control parameters and the parameter compensation values.

[0037] On the other hand, the present application provides a device for configuring programmable parameters, wherein the programmable parameters are used for an in vivo implantable neurostimulator, the device comprising:

[0038] An image acquisition module configured to acquire image data of a user;

[0039] a recommended parameter acquisition module configured to pre-process the user's image data to obtain a pre-processing result, and obtain recommended program-controlled parameters based on the pre-processing result;

[0040] The optimal parameter configuration module is configured to obtain the optimal program control parameters using the recommended program control parameters based on the parameter compensation model.

[0041] In a third aspect, the present application proposes a medical system, comprising:

[0042] Implantable medical devices, which are implanted in the body of a user;

[0043] A program-controlled parameter configuration device, wherein the program-controlled parameter configuration device realizes program-controlled parameter configuration through any of the methods described above.

[0044] In a fourth aspect, the present application provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements any one of the above methods when executing the computer program.

[0045] In some optional embodiments, the electronic device is further provided with a display screen.

[0046] In a fifth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program implements any of the above methods when executed by a processor.

[0047] In a sixth aspect, the present application provides a computer program product, comprising at least one of a computer program and instructions, which, when executed by a processor, implements any of the above-mentioned methods or the functions of the device described in the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The present application is described below with reference to the accompanying drawings and embodiments.

[0049] FIG1 is a schematic diagram of a program-controlled parameter configuration method provided in an embodiment of the present application;

[0050] FIG2 is a schematic diagram of a program-controlled parameter setting process according to an embodiment of the present application;

[0051] FIG3 is a schematic diagram of a method for constructing an ordered array of contact state combinations provided in an embodiment of the present application;

[0052] FIG4 is a schematic diagram of a program-controlled parameter configuration device provided in an embodiment of the present application;

[0053] FIG5 is a schematic diagram of a medical system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0054] Below, the present application is further described in conjunction with the accompanying drawings and specific implementation methods. It should be noted that, under the premise of no conflict, the multiple embodiments or multiple technical features described below can be arbitrarily combined to form new embodiments.

[0055] Below, we first briefly describe one of the application fields (i.e., implantable devices) of the embodiments of the present application. An implantable neural stimulation system (or implantable medical system) includes a stimulator implanted in a user, such as a patient's body, and a programmable device disposed outside the patient's body. The neural regulation technology in the related art mainly involves implanting electrodes in specific structures (i.e., targets) in the body through stereotactic surgery, and the stimulator implanted in the patient's body emits discharge pulses to the target through the electrodes to regulate the electrical activity and function of the corresponding neural structures and networks, thereby improving symptoms and alleviating pain. Among them, the stimulator can be any one of an implantable neural electrical stimulation device, an implantable cardiac electrical stimulation system (also known as a pacemaker), an implantable drug delivery system (IDDS), and a lead adapter. Examples of implantable neurostimulation devices include Deep Brain Stimulation (DBS), Cortical Nerve Stimulation (CNS), Spinal Cord Stimulation (SCS), Sacral Nerve Stimulation (SNS), and Vagus Nerve Stimulation (VNS). A stimulator may include an IPG, extension leads, and electrode leads. An implantable pulse generator (IPG) is placed in the patient's body, receives programmable instructions from a programmable device, and provides controllable electrical stimulation energy to tissues in the body through sealed batteries and circuits. Through the implanted extension leads and electrode leads, one or two controllable, specific electrical stimulations are delivered to specific areas of tissue in the body.

[0056] The efficacy and adverse reactions of DBS therapy are related to the distribution of the volume of tissue activated (VTA) near the electrode contacts. Generally speaking, the greater the overlap between the VTA and the target area, the better the treatment effect. However, if the VTA extends beyond the target area, stimulation-related adverse reactions may occur.

[0057] This application does not limit implant devices. The embodiments solve the problems existing in three-dimensional visualization programming by configuring the programming parameters of the implant device: 1. The initial programming parameters are difficult to determine; 2. The programming time is too long; 3. It is impossible to search all possible programming parameter combinations based on the doctor's experience and attempts (the number of possible programming parameter combinations is very large and it is impossible to try and evaluate them one by one), so the programming parameters determined after a limited number of attempts may not be the optimal programming parameters for the patient.

[0058] The technical solution described in this application is not limited to a specific type or quantity of electrodes, nor is it limited to a specific type of pulse generator.

[0059] 1 , an embodiment of the present application provides a method for configuring programmable parameters for an in vivo implantable neurostimulator. The method includes:

[0060] S1. Obtaining imaging data of the patient; the imaging data includes computed tomography (CT), magnetic resonance imaging (MRI), and other imaging data;

[0061] S2. Preprocessing the patient's imaging data to obtain preprocessing results, and obtaining recommended programming parameters based on the preprocessing results;

[0062] S3. Based on the parameter compensation model, the recommended program control parameters are used to obtain the optimal program control parameters.

[0063] The principle and effect of the above technical solution are as follows: It obtains information such as the patient's physiology and lesion condition; this information is crucial for subsequent programming parameter configuration, as it helps determine the most appropriate neurostimulator parameters for the patient. Useful information, such as the patient's neural structure and lesion severity, is extracted from the imaging data. This step typically involves techniques such as image enhancement, segmentation, and registration to ensure more accurate parameter recommendations. Based on the extracted patient information and combined with a processing model, the system recommends a preliminary set of programming parameters. A parameter compensation model is an algorithm or mathematical model that optimizes and adjusts the preliminary recommended programming parameters to achieve the optimal stimulation effect. This step typically involves technologies such as machine learning and artificial intelligence, using a trained model to find the most appropriate programming parameter configuration for the patient. This method can quickly and accurately provide the optimal programming parameter configuration for the patient, significantly shortening the treatment cycle and improving treatment efficiency. Because this method provides more precise and personalized treatment and management plans, patients may experience better treatment outcomes, thereby increasing patient satisfaction.

[0064] In one embodiment of the present application, a method for configuring program control parameters, wherein the patient's imaging data is preprocessed to obtain a preprocessing result, and recommended program control parameters are obtained based on the preprocessing result, including:

[0065] Segment the patient's imaging data to obtain the target area;

[0066] Obtain the electric field intensity scalar field of the initial program-controlled parameters in VTA, i.e., three-dimensional space, and calculate the coincidence rate between the target area and the electric field intensity scalar field;

[0067] By iteratively calculating the initial program control parameters, the overlap rate corresponding to the initial program control parameters in each iteration is obtained, and the highest overlap rate is obtained based on multiple overlap rates obtained in multiple iterations. Based on the highest overlap rate, the recommended program control parameters are obtained.

[0068] In some optional embodiments, obtaining an electric field intensity scalar field of the initial program-controlled parameters in three-dimensional space and calculating an overlap ratio between the target area and the electric field intensity scalar field include:

[0069] Segment the patient's imaging data using a segmentation algorithm to obtain segmentation results for different tissues;

[0070] Orderly construct the contact array of the neurostimulator;

[0071] The initial programming parameters are used as input, the contact array and the segmentation result are used as input of the first objective function, and a scalar field of electric field intensity of the initial programming parameters in three-dimensional space is obtained through finite element analysis simulation. The first objective function can be abstracted as an S function. The initial programming parameters are any set of programming parameters within a safe range that can be used for different diseases (such as Parkinson's disease).

[0072] The coincidence rate is obtained by calculating the integral of the electric field intensity scalar field in the target area.

[0073] In some optional embodiments, obtaining recommended program control parameters based on the highest overlap rate includes:

[0074] Through an iterative algorithm, the program control parameter combination with the highest overlap rate is obtained as the recommended program control parameter; the iterative algorithm can be an algorithm such as linear programming, gradient descent, or back propagation neural network, which can search for the program control parameter combination with the highest overlap rate as the recommended program control parameter.

[0075] The working principle of the above technical solution is:

[0076] The technical solution described in this application is not limited to a specific type or number of electrodes, nor is it limited to a specific type of pulse generator. If the electrodes on the left and right sides of the brain each have n contacts, the contacts on the left electrode are numbered 0, 1, 2, ..., n-1 from the most ventral to the most dorsal, and the contacts on the right electrode are numbered n, n+1, n+2, ..., 2n-1 from the most ventral to the most dorsal, and the pulse generator is C. During treatment, each contact has three possible states:

[0077] 1. Used as negative electrode (-);

[0078] 2. Used as positive electrode (+);

[0079] 3. Close (_).

[0080] There are two possible states for the pulse generator:

[0081] 1. Used as positive electrode (+);

[0082] 2. Close (_).

[0083] If closed (_) is recorded as 0, used as negative pole (-) is recorded as 1, and used as positive pole (+) is recorded as 2, then the selection combination of all contacts can be represented by an ordered array contact_list, as shown in Figure 3.

[0084] The T1 brain image obtained by scanning the patient before DBS surgery can be calculated using a common segmentation algorithm to obtain the segmentation results (brain_masks) of gray matter, white matter, cerebrospinal fluid, blood vessels, etc. In this embodiment, the segmentation result is the brain region segmentation result. Different tissue types have different electrical conductivities, which will affect the subsequent finite element simulation calculation of VTA (i.e., the electric field intensity scalar field in three-dimensional space); program-controlled parameters (amplitude V, frequency F, pulse width P), contact_list, brain_masks. The calculation process of VTA can be abstracted as the following S function: The S function is the first objective function; VTA = S(V, F, P, contact_list, brain_masks)

[0085] Where VTA is the electric field intensity scalar field in three-dimensional space.

[0086] The patient's T1 brain image can be segmented manually, using a rule-based algorithm, or using a trained neural network to obtain a target volume, denoted as target_mask. target_mask is a scalar field in three-dimensional space, where voxels inside the target volume have a value of 1 and voxels outside the target volume have a value of 0. The overlap ratio is defined as the integral of the VTA within the target_mask as follows:

[0087] The off-target ratio is defined as the integral of VTA outside the target_mask area, as follows:

[0088] Among them, C Whole Space target_mask represents the complement of target_mask relative to Whole Space (C represents the complement symbol), that is, the area remaining after subtracting the target_mask area from Whole Space, which is the area you don't want to stimulate. Written under the integral symbol, it represents the volume integral of the VTA within this spatial region, that is, the integral of the stimulation intensity within the area you don't want to stimulate. This integral value is defined as the off-target rate. The higher the off-target rate, the more off-target areas are stimulated.

[0089] To simplify calculations, the VTA can also be converted into a scalar field in three-dimensional space using binary truncation. Specifically, given a given electric field strength threshold, all voxels below the threshold are set to 0, and all voxels above the threshold are set to 1. The overlap ratio can then be simplified to the ratio of the volume of the overlap between the VTA and target_mask to the volume of the VTA. In this case, the miss rate = 1 minus the overlap ratio.

[0090] When the program control parameters are given, the VTA is determined immediately, and the coincidence rate and miss rate are also determined accordingly.

[0091] The algorithm's initial iteration assumes that therapeutic efficacy is proportional to the overlap rate and adverse reactions are proportional to the off-target rate. Therefore, the recommended programming parameters are those that maximize the overlap rate. Using algorithms such as linear programming, gradient descent, or back-propagation neural networks, the combination of programming parameters with the highest overlap rate is searched for and used as the recommended programming parameters. Physicians can then fine-tune the recommended parameters based on this set of parameters, significantly shortening programming time and improving both efficiency and efficacy.

[0092] The above process can be abstracted into a function that takes "maximizing the overlap rate" as the goal and searches for "recommended program control parameters" as follows:

[0093] In some optional embodiments, the method for constructing the parameter compensation model includes:

[0094] Acquire multiple sets of actual program control parameters in the actual treatment record; illustratively, the actual program control parameters and corresponding recommended program control parameters in the user's actual treatment record can be acquired through a database;

[0095] Using each set of actual program control parameters and the corresponding recommended program control parameters, the error results corresponding to the set of actual program control parameters are obtained;

[0096] A parameter compensation model is constructed and iteratively optimized based on multiple error results corresponding to multiple sets of actual programming parameters. For example, multiple sets of actual programming parameters in the actual treatment records of multiple users and the recommended programming parameters corresponding to each set of actual programming parameters can be obtained through the database to obtain multiple error results. The parameter compensation model can adopt matrix least squares method, support vector machine, or more complex neural network.

[0097] In some embodiments, multiple first errors are obtained using multiple sets of actual programming parameters, recommended programming parameters corresponding to each set of actual programming parameters, and all other inputs required by the first objective function; all other inputs include contact arrays and brain segmentation results; the first error is the difference between the first relative area and the second relative area; the first relative area is obtained based on the actual programming parameters, the number of contact groups and the patient's brain segmentation results (brain_masks) and based on the first objective function, that is, the relative position relationship between the actual VTA and the brain segmentation results (brain_masks); the second relative area is obtained based on the recommended programming parameters, the number of contact groups and the patient's brain segmentation results (brain_masks) and based on the first objective function, that is, the relative position relationship between the preset VTA and the brain segmentation results (brain_masks); the first objective function can be abstracted as an S function; the input of the first objective function includes programming parameters, contact arrays and brain segmentation results.

[0098] A first parameter compensation model is constructed through multiple first errors and iterative optimization is performed.

[0099] In some other embodiments, a plurality of second errors are obtained using a plurality of sets of actual program control parameters and a recommended program control parameter corresponding to each set of actual program control parameters; the second errors are direct differences between the actual program control parameters and the corresponding recommended program control parameters;

[0100] A second parameter compensation model is constructed through multiple second errors and iterative optimization is performed.

[0101] The working principle of the above technical solution is as follows: with reference to FIG2 , for a specific patient, using the preoperative MRI image and the postoperative CT image as input, the programming parameters that maximize the overlap rate can be calculated. There is an unknown error distribution between the programming parameters and the actual programming parameters that maximize the clinical efficacy, namely, the "parameter error Ep". Similarly, there is also an unknown error distribution between the VTA (i.e., the recommended VTA) calculated by the programming parameters that maximize the overlap rate and the actual VTA calculated by the programming parameters that maximize the clinical efficacy, namely, the "VTA error Ev". Because the programming parameters that maximize the overlap rate are known, if Ep can be fitted, then only the error Ep needs to be compensated to obtain the programming parameters that maximize the clinical efficacy. In addition, if Ev can be fitted, since the recommended VTA can be calculated by the programming parameters that maximize the overlap rate, only the error Ev needs to be compensated to obtain the actual VTA that maximizes the clinical efficacy, and then the programming parameters that maximize the clinical efficacy can be inversely solved by the S function.

[0102] Each time a patient uses a program-controlled device for treatment, the device records the treatment log and uploads it to the company's server. This server then accumulates a large amount of data generated by the treatment, including but not limited to the specific parameters of each treatment and the date of the treatment. Using these specific parameters and the MRI and CT images obtained during the DBS procedure, we can calculate the actual VTA corresponding to these treatment parameters. The difference between the actual programming parameters and the VTA corresponding to the recommended programming parameters can be obtained through the VTA corresponding to the recommended programming parameters. A training model is established through multiple first errors (i.e., the differences between multiple VTAs), recommended programming parameters, and actual programming parameters for training, and iterative optimization is performed according to the amount of data collected to obtain a first parameter compensation model. The first parameter compensation model outputs the final first error. The relative position relationship between the VTA corresponding to the recommended programming parameters and the brain area segmentation result (brain_masks) and the final first error output by the model is used to obtain the optimal relative position relationship between the VTA and the brain area segmentation result (brain_masks). Based on the first objective function, the final programming parameters, i.e., the optimal programming parameters, are obtained. The training model can adopt matrix least squares method, support vector machine, or more complex neural network.

[0103] A training model can also be established based on the second error, recommended program control parameters, and actual program control parameters to obtain a second parameter compensation model. The final second error can be obtained through the second parameter compensation model. The final program control parameters, i.e., the optimal program control parameters, can be obtained through the recommended program control parameters and the final second error. The training model can adopt matrix least squares method, support vector machine, or more complex neural network.

[0104] When a large amount of programming data is input, Efit can effectively fit the distribution of Ep or Ev, thereby helping us compensate for the error and obtain the programming parameters that maximize clinical efficacy, namely the "optimal programming parameters."

[0105] The above technical solution provides the following advantages: During the construction of the parameter compensation model, actual programming parameters from actual treatment records and actual treatment records in a database are obtained. Then, based on the first objective function, the first error is calculated using the actual programming parameters, the corresponding recommended programming parameters, the number of contact groups, and the patient's brain segmentation results. Alternatively, the second error can be directly calculated using the actual programming parameters and the corresponding recommended programming parameters. The parameter compensation model allows for a more accurate prediction of the final error, which is then fitted to the recommended programming parameters to configure optimal programming parameters, thereby improving treatment accuracy and effectiveness. Using multiple first or second errors, a parameter compensation model can be constructed and iteratively optimized. This process can be accomplished using a training model such as matrix least squares, support vector machines, or complex neural networks. Through continuous iterative optimization, the model gradually fits the error distribution, thereby helping to compensate for errors and obtain optimal programming parameters. Through the iterative optimization process of the parameter compensation model, optimal programming parameters are obtained that maximize treatment effectiveness and minimize adverse reactions. Physicians can then perform programming based on these parameters, providing patients with more precise and personalized treatment and management plans.

[0106] The target nucleus based on the recommended programming parameters obtained theoretically may not be the optimal target nucleus. However, obtaining the actual target nucleus based on the parameter compensation model through the patient's actual usage records can help improve treatment effects and patient satisfaction, while promoting the advancement of medical technology and the sustainable development of the medical system.

[0107] An embodiment of the present application provides a program-controlled parameter configuration method, which constructs a parameter compensation model and performs iterative optimization based on multiple error results corresponding to multiple sets of actual program-controlled parameters, including:

[0108] Get the usage time of each set of actual program control parameters;

[0109] The weight of the error result corresponding to the set of actual program-controlled parameters is obtained by using the duration;

[0110] The weights of the multiple error results are used as inputs to a parameter compensation model to obtain a final parameter compensation output result.

[0111] The working principle of the above technical solution is to record the actual usage time of the programmed parameters through the programming device used by the patient. This usage time can reflect the stability and reliability of the parameters. The longer the usage time, the higher the parameter stability and reliability. The parameter with the longest usage time is selected as the programming parameter that maximizes clinical efficacy. Because the lower the programming frequency, the longer the usage time of the same parameter indicates the patient's lower willingness to actively seek medical treatment for parameter adjustment, it is inferred that the patient's therapeutic effect under this programming parameter is more significant and stable. Parameters with longer usage time are assigned a greater efficacy weight. Based on the usage time, the weight of the error result corresponding to the actual programming parameter is calculated. Next, the error result weight is used as the input of the parameter compensation model for iterative optimization. The iterative optimization process can be accomplished using a training model such as matrix least squares, support vector machine, or complex neural network. Through iterative optimization, the model gradually fits the error distribution and adjusts the parameter accuracy based on the size of the weight. Through the iterative optimization process, the final parameter compensation output is obtained. This result serves as a reference for optimizing the programming parameters, providing doctors with more precise and personalized treatment and management plans.

[0112] In some embodiments, the duration of use can be the continuous use time of a set of programmed parameters. Continuous use time can refer to the length of time a patient continuously uses the parameters during a treatment. If a patient uses the same parameter configuration for a long period of time during a treatment, it can indicate that the parameters are stable and effective during that particular treatment.

[0113] In other embodiments, the usage time may be the cumulative usage time of a certain set of programming parameters. For example, a patient uses a certain set of actual programming parameters for one hour today and one hour tomorrow, for a total of 50 hours of use over a period of time. Another set of actual programming parameters is used for a total of 30 hours over the same period of time. The weight of the corresponding error of the actual programming parameters used for a total of 50 hours is higher than the weight of the corresponding error of the actual programming parameters used for a total of 30 hours.

[0114] The effect of the above technical solution is: by considering the actual usage time of the programmed parameters and the weight of the error results, the accuracy and reliability of the parameters can be more accurately evaluated, thereby providing a more precise parameter configuration solution. Through more precise and reliable parameter configuration, the treatment effect can be greatly improved, and the patient's cure rate and recovery level can be improved. The usage time can reflect the stability and reliability of the parameters. By weighted calculation of the weights, the reliability and stability of the parameters can be more accurately evaluated, thereby improving the reliability of the entire medical system. Through an iterative optimization process, the accuracy and reliability of the parameter compensation model can be continuously improved, thereby providing doctors with more accurate and personalized treatment and management plans, shortening the treatment cycle, and improving treatment efficiency.

[0115] An embodiment of the present application provides a method for configuring program-controlled parameters, which obtains the weight of an error result corresponding to an actual program-controlled parameter by using a duration, including:

[0116] Statistical analysis is performed on multiple usage times to obtain a duration statistical result. The usage time here can be continuous usage time or cumulative usage time;

[0117] The duration statistics results are grouped to obtain grouping results; wherein the grouping rule may be the proportion of the statistical duration in the duration distribution, for example, the duration distribution between 0% and 5% may be grouped as one group, the duration distribution between 5% and 10% may be grouped as another group, and so on;

[0118] The weight of the error result corresponding to the actual program-controlled parameter is obtained according to the grouping result.

[0119] Among some possible methods, the weights are obtained as follows:

[0120] Where i is the group, M is the total number of groups; W i is the weight corresponding to the i-th group.

[0121] The working principle and effect of the above technical solution is to collect and record the usage time data of multiple actual program-controlled parameters. This data reflects the time and stability of different parameters in actual use.

[0122] Next, statistical analysis is performed on the collected usage time data to obtain duration statistics. This step may include calculating statistics such as average usage time, maximum usage time, and minimum usage time to fully understand the distribution characteristics of the data.

[0123] Then, based on the results of the statistical analysis, the usage duration is grouped. The grouping rule can be based on the proportion of the duration in the overall distribution. For example, the usage duration between 0% and 5% is grouped together, the usage duration between 5% and 10% is grouped together, and so on. This grouping helps to distinguish the differences in usage frequency and stability of different parameters.

[0124] After the grouping is completed, the weight of each group is calculated based on the grouping results. The weight calculation formula is Where i is the group, N is the total number of groups; W i is the weight corresponding to group i. The formula ensures that the sum of the weights is 1, and the weight of each group is proportional to its position in the grouping sequence. This means that parameter groups with higher frequency of use and better stability receive larger weights. The calculated weights serve as input to the parameter compensation model. The parameter compensation model uses this weight information, combined with the error results (i.e., the difference between the recommended program parameters and the actual program parameters), to perform iterative optimization to find the optimal parameter compensation configuration.

[0125] In summary, this method provides important input information for the parameter compensation model by statistically analyzing usage duration, grouping, and calculating weights based on the distribution characteristics of duration. This helps improve the accuracy and reliability of programmed parameters, thereby optimizing treatment outcomes and enhancing patient satisfaction.

[0126] In other embodiments, if the usage time of a group of actual program-controlled parameters is much longer than that of other groups, the weight of the error result corresponding to this group of actual program-controlled parameters is set to be infinitely close to 1; or the weight of the error result of the program-controlled parameter with the longest usage time is set to be infinitely close to 1.

[0127] In some other embodiments, multiple groups of actual program-controlled parameters with higher usage times may be selected, and weights of corresponding error results may be allocated according to the usage times corresponding to the selected actual program-controlled parameters.

[0128] In one embodiment of the present application, a method for configuring program control parameters, based on a parameter compensation model, using the recommended program control parameters to obtain optimal program control parameters, includes:

[0129] Obtaining a parameter compensation value based on a parameter compensation model; the parameter compensation model includes a first parameter compensation model or a second parameter compensation model;

[0130] Obtain the optimal program control parameters by recommending program control parameters and parameter compensation values.

[0131] In some embodiments, the parameter compensation model is a first parameter compensation model; the specific process is as follows:

[0132] 1) Obtain the VTA corresponding to the recommended programming parameters;

[0133] Input: recommended programming parameters, first objective function, contact array, and brain segmentation results;

[0134] Output: VTA corresponding to the recommended program control parameters;

[0135] 2) Calculate the first compensation, i.e., the final first error, based on the first parameter compensation model:

[0136] Input: recommended program control parameters;

[0137] Processing: The first parameter compensation model receives the recommended program control parameters and calculates the corresponding first compensation, i.e., the final first error, according to the internal logic or algorithm of the model;

[0138] Output: final first error;

[0139] 3) VTA after compensation:

[0140] Input: final first error, VTA corresponding to recommended program control parameters;

[0141] Processing: Apply the calculated first error to the VTA corresponding to the recommended program control parameters to obtain the final VTA, which represents the VTA after compensation;

[0142] Output: compensated VTA;

[0143] 4) Determine the optimal program control parameters:

[0144] Input: compensated VTA, first objective function, contact array, and brain segmentation results;

[0145] Processing: According to the final electric field strength scalar field and the first objective function, the optimal program control parameters are determined in reverse;

[0146] Output: Optimal programming parameters.

[0147] In summary, this method calculates the final first error using a first parameter compensation model, adjusts the VTA, and ultimately determines the optimal programming parameters. This process, based on mathematical models and a data-driven approach, ensures the accuracy and reliability of parameter configuration, thereby optimizing treatment outcomes.

[0148] In some other embodiments, the parameter compensation model is a second parameter compensation model. Based on the second parameter compensation model, the optimal compensation value of the corresponding recommended programming parameter is calculated. The optimal compensation value is the optimal direct difference between the actual programming parameter and the corresponding recommended programming parameter. The optimal programming parameter is obtained through the optimal compensation value and the corresponding recommended programming parameter.

[0149] This method uses mathematical models and data-driven methods to automatically calculate the optimal compensation values ​​and optimal programming parameters for recommended program control parameters, reducing the need for manual intervention. This saves time and labor costs and reduces the risk of human error.

[0150] This method can be adapted to different implementations and treatment scenarios. By adjusting the parameter compensation model and objective function, it can be customized according to actual needs. This flexibility makes the method widely applicable and can be applied to a variety of programmable parameter configuration requirements.

[0151] In summary, by obtaining the optimal programming parameters based on the parameter compensation model and recommended programming parameters, this method can improve treatment effects, automate parameter configuration, improve accuracy and stability, and be scalable and adaptable.

[0152] The benefits of this application include at least the following aspects: By using a parameter compensation model and recommended programming parameters, optimal programming parameters can be obtained. This ensures that the optimal electric field strength scalar field is obtained during treatment, thereby improving efficacy. It provides baseline parameters for DBS post-operative programming and a reference benchmark for subsequent programming, significantly reducing programming time. This avoids prolonged programming time, poor therapeutic effect feedback caused by patient fatigue and inattention, and thus improves the efficiency and effectiveness of programming. The target area is obtained through a segmentation algorithm, and the overlap rate is calculated using the initial programming parameters. An iterative calculation is performed to obtain the highest overlap rate. Based on the highest overlap rate, recommended programming parameters are obtained, thereby automatically performing parameter optimization. This reduces manual intervention and improves the efficiency and accuracy of parameter optimization. By constructing a parameter compensation model and performing iterative optimization, error results can be obtained based on the actual programming parameters in the actual treatment record and the corresponding recommended programming parameters, thereby further optimizing the parameter configuration. This can continuously improve the accuracy and stability of the parameter configuration. It provides a reference for setting up programming after DBS surgery, reduces the difference in efficacy caused by differences in the diagnosis and treatment experience of programming doctors, and reduces the frequency of patient programming, bringing convenience to patients and saving health expenses.

[0153] The embodiment of the present application provides a device for configuring programmable parameters, wherein the programmable parameters are used for an in vivo implantable neurostimulator, and the device comprises:

[0154] An image acquisition module configured to acquire imaging data of a patient;

[0155] a recommended parameter acquisition module configured to preprocess the patient's imaging data to obtain a preprocessing result, and obtain recommended program-controlled parameters based on the preprocessing result;

[0156] The optimal parameter configuration module is configured to obtain the optimal program control parameters using the recommended program control parameters based on the parameter compensation model.

[0157] The recommended parameter acquisition module includes:

[0158] A target region acquisition unit is configured to segment the patient's imaging data to obtain a target region;

[0159] The coincidence rate acquisition unit is configured to obtain the electric field intensity scalar field of the initial program-controlled parameters in the VTA, i.e., the three-dimensional space, and calculate the coincidence rate between the electric field intensity scalar field and the target area;

[0160] a maximum overlap rate obtaining unit configured to obtain the overlap rate corresponding to the initial program-controlled parameters in each iteration by iteratively calculating the initial program-controlled parameters, and obtain the maximum overlap rate based on multiple overlap rates obtained in multiple iterations;

[0161] The recommended parameter acquisition unit is configured to obtain recommended program control parameters based on the highest overlap rate.

[0162] The recommended parameter acquisition unit includes:

[0163] A segmentation unit is configured to segment the patient's imaging data using a segmentation algorithm to obtain segmentation results of different tissues;

[0164] a contact construction unit, configured to construct a contact array of a neurostimulator in an orderly manner;

[0165] The first calculation unit is configured to take the initial programming parameters as input, the contact array and the segmentation result as input of a first objective function, and obtain a scalar field of electric field intensity of the initial programming parameters in a three-dimensional space; wherein the first objective function can be abstracted as an S-function; wherein the initial programming parameters are any set of programming parameters within a safe range that can be commonly used for different diseases (e.g., Parkinson's disease);

[0166] The second calculation unit is configured to calculate the integral of the electric field intensity scalar field in the target area to obtain the overlap rate.

[0167] The highest overlap rate acquisition unit is configured to obtain the program control parameter combination with the highest overlap rate as the recommended program control parameter through an iterative algorithm; and the program control parameter combination with the highest overlap rate can be searched out as the recommended program control parameter through algorithms such as linear programming, gradient descent, or back propagation neural network.

[0168] The technical solution described in this application is not limited to a specific type or number of electrodes, nor is it limited to a specific type of pulse generator. If the electrodes on the left and right sides of the brain each have n contacts, the contacts on the left electrode are numbered 0, 1, 2, ..., n-1 from the most ventral to the most dorsal, and the contacts on the right electrode are numbered n, n+1, n+2, ..., 2n-1 from the most ventral to the most dorsal, and the pulse generator is C. During treatment, each contact has three possible states:

[0169] 1. Used as negative electrode (-);

[0170] 2. Used as positive electrode (+);

[0171] 3. Close (_).

[0172] There are two possible states for the pulse generator:

[0173] 1. Used as positive electrode (+);

[0174] 2. Close (_).

[0175] If closed (_) is recorded as 0, used as negative pole (-) is recorded as 1, and used as positive pole (+) is recorded as 2, then the selection combination of all contacts can be represented by an ordered array contact_list, as shown in Figure 3.

[0176] The T1 brain image obtained by scanning the patient before DBS surgery can be calculated using a common segmentation algorithm to obtain the segmentation results (brain_masks) of gray matter, white matter, cerebrospinal fluid, blood vessels, etc. In this embodiment, the segmentation result is the brain region segmentation result. Different tissue types have different electrical conductivities, which will affect the subsequent finite element simulation calculation of VTA (i.e., the electric field intensity scalar field in three-dimensional space); program-controlled parameters (amplitude V, frequency F, pulse width P), contact_list, brain_masks. The calculation process of VTA can be abstracted as the following S function: The S function is the first objective function; VTA = S(V, F, P, contact_list, brain_masks)

[0177] Where VTA is the electric field intensity scalar field in three-dimensional space.

[0178] The patient's T1 brain image can be segmented manually, using a rule-based algorithm, or using a trained neural network to obtain a target volume, denoted as target_mask. target_mask is a scalar field in three-dimensional space, where voxels inside the target volume have a value of 1 and voxels outside the target volume have a value of 0. The overlap ratio is defined as the integral of the VTA within the target_mask as follows:

[0179] The off-target ratio is defined as the integral of VTA outside the target_mask area, as follows:

[0180] Among them, C Whole Space target_mask represents the complement of target_mask relative to Whole Space (C represents the complement symbol), that is, the area remaining after subtracting the target_mask area from Whole Space, which is the area you don't want to stimulate. Written under the integral symbol, it represents the volume integral of the VTA within this spatial region, that is, the integral of the stimulation intensity within the area you don't want to stimulate. This integral value is defined as the off-target rate. The higher the off-target rate, the more off-target areas are stimulated.

[0181] To simplify calculations, the VTA can also be converted into a scalar field in three-dimensional space using binary truncation. Specifically, given a given electric field strength threshold, all voxels below the threshold are set to 0, and all voxels above the threshold are set to 1. The overlap ratio can then be simplified to the ratio of the volume of the overlap between the VTA and target_mask to the volume of the VTA. In this case, the miss rate = 1 minus the overlap ratio.

[0182] When the program control parameters are given, the VTA is determined immediately, and the coincidence rate and miss rate are also determined accordingly.

[0183] The algorithm's initial iteration assumes that therapeutic efficacy is proportional to the overlap rate and adverse reactions are proportional to the off-target rate. Therefore, the recommended programming parameters are those that maximize the overlap rate. Using algorithms such as linear programming, gradient descent, or back-propagation neural networks, the combination of programming parameters with the highest overlap rate is searched for and used as the recommended programming parameters. Physicians can then fine-tune the recommended parameters based on this set of parameters, significantly shortening programming time and improving both efficiency and efficacy.

[0184] The above process can be abstracted into a function that takes "maximizing the overlap rate" as the goal and searches for "recommended program control parameters" as follows:

[0185] In some optional embodiments, the optimal parameter configuration module includes a parameter compensation model construction submodule, and the parameter compensation model construction submodule includes:

[0186] an actual record acquisition unit configured to acquire multiple sets of actual program-controlled parameters in the actual treatment record; illustratively, the actual program-controlled parameters and corresponding recommended program-controlled parameters in the actual treatment record may be acquired through a database;

[0187] An error obtaining unit is configured to obtain an error result corresponding to each set of actual program control parameters using the corresponding recommended program control parameters;

[0188] The model building unit is configured to build a parameter compensation model and perform iterative optimization based on multiple error results corresponding to multiple sets of actual program-controlled parameters.

[0189] The parameter compensation model includes a first parameter compensation model or a second parameter compensation model.

[0190] Among them, the method for constructing the first parameter compensation model is: using multiple groups of actual programming parameters, the recommended programming parameters corresponding to each group of actual programming parameters, and all other inputs required by the first objective function to obtain multiple first errors; the first error is the difference between the first relative area and the second relative area; the first relative area is obtained based on the actual programming parameters, the number of contact groups and the patient's brain area segmentation result (brain_masks) and the first objective function, that is, the relative position relationship between the actual VTA and the brain area segmentation result (brain_masks); the second relative area is obtained based on the recommended programming parameters, the number of contact groups and the patient's brain area segmentation result (brain_masks) and the first objective function, that is, the relative position relationship between the preset VTA and the brain area segmentation result (brain_masks); the first objective function can be abstracted as an S function; the input of the first objective function includes programming parameters, contact arrays and brain area segmentation results.

[0191] A first parameter compensation model is constructed through multiple first errors and iterative optimization is performed.

[0192] The construction method of the second parameter compensation model is:

[0193] Using multiple sets of actual program control parameters and recommended program control parameters corresponding to each set of actual program control parameters, a plurality of second errors are obtained; the second errors are direct differences between the actual program control parameters and the recommended program control parameters;

[0194] A second parameter compensation model is constructed through multiple second errors and iterative optimization is performed.

[0195] The model building unit also includes:

[0196] A duration acquisition unit is configured to acquire the usage duration of each set of actual program-controlled parameters;

[0197] A weight obtaining unit, configured to obtain the weight of the error result corresponding to the set of actual parameters by using the duration;

[0198] The input module is configured to use the weights of the multiple error results as inputs to a parameter compensation model to obtain a final parameter compensation output result.

[0199] In some embodiments, the usage duration of the actual programmed parameters includes the continuous usage duration of each set of actual programmed parameters.

[0200] In other embodiments, the usage time of the actual program-controlled parameters includes the accumulated usage time of each group of actual program-controlled parameters.

[0201] In some embodiments, the weight acquisition unit includes:

[0202] A statistical unit is configured to perform statistical analysis on multiple continuous usage durations to obtain duration statistics results;

[0203] A grouping unit, configured to group the duration statistics results to obtain grouping results;

[0204] The calculation unit is configured to obtain the weight of the error result corresponding to the actual program-controlled parameter according to the grouping result.

[0205] In other embodiments, if the usage time of a group of actual program-controlled parameters is much longer than that of other groups, the weight of the error result corresponding to this group of actual program-controlled parameters is set to be infinitely close to 1; or the weight of the error result of the program-controlled parameter with the longest usage time is set to be infinitely close to 1.

[0206] In some other embodiments, multiple groups of actual program-controlled parameters with earlier usage times may be selected, and weights of corresponding error results may be allocated according to the usage times corresponding to the selected actual program-controlled parameters.

[0207] In some embodiments, the optimal parameter configuration module includes:

[0208] a compensation obtaining unit configured to obtain a parameter compensation value based on a parameter compensation model; the parameter compensation model includes a first parameter compensation model or a second parameter compensation model;

[0209] The final configuration unit is set to obtain the optimal program control parameters through the recommended program control parameters and parameter compensation values.

[0210] In some embodiments, the parameter compensation model is a first parameter compensation model; the specific process is as follows:

[0211] 1) Obtain the VTA corresponding to the recommended programming parameters;

[0212] Input: recommended programming parameters, first objective function, contact array, and brain segmentation results;

[0213] Output: VTA corresponding to the recommended program control parameters;

[0214] 2) Calculate the first compensation, i.e., the final first error, based on the first parameter compensation model:

[0215] Input: recommended program control parameters;

[0216] Processing: The first parameter compensation model receives the recommended program control parameters and calculates the corresponding first compensation, i.e., the final first error, according to the internal logic or algorithm of the model;

[0217] Output: final first error;

[0218] 3) VTA after compensation:

[0219] Input: final first error, VTA corresponding to recommended program control parameters;

[0220] Processing: Apply the calculated first error to the VTA corresponding to the recommended program control parameters to obtain the final VTA, which represents the VTA after compensation;

[0221] Output: compensated VTA;

[0222] 4) Determine the optimal program control parameters:

[0223] Input: compensated VTA, first objective function, contact array, and brain segmentation results;

[0224] Processing: According to the final electric field strength scalar field and the first objective function, the optimal program control parameters are determined in reverse;

[0225] Output: Optimal programming parameters.

[0226] In some other embodiments, the parameter compensation model is a second parameter compensation model. Based on the second parameter compensation model, the optimal compensation value of the corresponding recommended programming parameter is calculated. The optimal compensation value is the optimal direct difference between the actual programming parameter and the corresponding recommended programming parameter. The optimal programming parameter is obtained through the optimal compensation value and the corresponding recommended programming parameter.

[0227] The working principle and effect of the above modules refer to the program-controlled parameter configuration method in this application and will not be described in detail here.

[0228] The present application also provides a medical system, as shown in FIG5 , comprising:

[0229] Implantable medical devices, which are implanted in a patient's body;

[0230] A program-controlled parameter configuration device, wherein the program-controlled parameter configuration device realizes program-controlled parameter configuration through any program-controlled parameter configuration method recorded in the embodiments.

[0231] The present application also provides an electronic device, which includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the above-mentioned program-controlled parameter configuration method when executing the computer program.

[0232] The present application also provides a computer program product, comprising a computer program / instruction, which implements any of the above-mentioned program-controlled parameter configuration methods when executed by a processor.

[0233] In some optional embodiments, the electronic device is further provided with a display screen.

[0234] An embodiment of the present application also provides a computer-readable storage medium, which is used to store a computer program. When the computer program is executed, the steps of the program-controlled parameter configuration method in the embodiment of the present application are implemented. Its specific implementation method is consistent with the implementation method and the technical effect achieved in the above-mentioned program-controlled parameter configuration embodiment, and some contents are not repeated here.

[0235] In the present application, a readable storage medium can be any tangible medium that contains or stores a program that can be used by or in combination with an instruction execution system, device, or device. A program product can use any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination of the above. The readable storage medium (a non-exhaustive list) can include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM) or 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.

[0236] Computer-readable storage media may include a data signal propagated in baseband or as part of a carrier wave, wherein readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium that can send, propagate, or transmit a program for use by an instruction execution system, apparatus, or device or used in conjunction with it. The program code contained on the readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, radio frequency (RF), etc., or any suitable combination thereof. The program code for performing the operations of this application may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., as well as conventional procedural programming languages ​​such as C or similar programming languages. The program code may be executed entirely on a user computing device, partially on an associated device, as a separate software package, partially on a user computing device, partially on a remote computing device, or entirely on a remote computing device or server. In situations involving a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or it can be connected to an external computing device (for example, through the Internet using an Internet service provider).

Claims

1. A method for configuring programmable parameters for an in vivo implantable neurostimulator, the method comprising: Obtain user's image data; Preprocessing the user's image data to obtain a preprocessing result, and obtaining recommended program control parameters based on the preprocessing result; Based on the parameter compensation model, the recommended program control parameters are used to obtain optimal program control parameters.

2. The program-controlled parameter configuration method according to claim 1, wherein: The preprocessing of the user's image data to obtain a preprocessing result, and obtaining recommended program control parameters based on the preprocessing result, includes: Segmenting the user's image data to obtain a target area; Obtaining an electric field intensity scalar field of initial program-controlled parameters in three-dimensional space, and calculating an overlap rate between the electric field intensity scalar field and the target area; By iteratively calculating the initial program control parameters, the overlap rate corresponding to the initial program control parameters in each iteration is obtained, and the highest overlap rate is obtained based on multiple overlap rates obtained in multiple iterations. Based on the highest overlap rate, the recommended program control parameters are obtained.

3. The program-controlled parameter configuration method according to claim 2, wherein: The step of obtaining the electric field intensity scalar field of the initial program-controlled parameters in the three-dimensional space and calculating the coincidence rate between the electric field intensity scalar field and the target area includes: Segmenting the user's image data using a segmentation algorithm to obtain segmentation results of different tissues; Orderly construct the contact array of the neurostimulator; Taking initial program-controlled parameters as input, the contact array and the segmentation result as first objective function input, and obtaining the electric field intensity scalar field of the initial program-controlled parameters in three-dimensional space; The coincidence rate is obtained by calculating the integral of the electric field intensity scalar field in the target area.

4. The program-controlled parameter configuration method according to claim 1, wherein: The method for constructing the parameter compensation model includes: Obtain multiple sets of actual program control parameters from actual treatment records; Using each set of actual program control parameters and the corresponding recommended program control parameters, an error result corresponding to the set of actual program control parameters is obtained; A parameter compensation model is constructed and iterative optimization is performed based on the multiple error results corresponding to the multiple groups of actual program-controlled parameters.

5. The program-controlled parameter configuration method according to claim 4, wherein: The method of constructing a parameter compensation model and performing iterative optimization based on the multiple error results corresponding to the multiple sets of actual program-controlled parameters includes: Obtaining the usage time of each set of actual program-controlled parameters; Obtaining the weight of the error result corresponding to the set of actual program-controlled parameters according to the usage time; The weights of the multiple error results are used as inputs of the parameter compensation model to obtain a final parameter compensation output result.

6. The program-controlled parameter configuration method according to claim 5, wherein: The usage time of the actual program-controlled parameters includes the continuous usage time of each group of actual program-controlled parameters or the accumulated usage time of each group of actual program-controlled parameters.

7. The program-controlled parameter configuration method according to claim 4, further comprising: A plurality of first errors are obtained using the multiple sets of actual program parameters, the recommended program parameters corresponding to each set of actual program parameters, and a first objective function; wherein the input of the first objective function includes the actual program parameters, the recommended program parameters, the contact array, and the segmentation result; the first error is the difference between the first relative area and the second relative area; the first relative area is obtained based on the actual program parameters, the user's brain segmentation result, and the contact array and the first objective function; the second relative area is obtained based on the recommended program parameters, the user's brain segmentation result, and the contact array and the first objective function; A first parameter compensation model is constructed using the multiple first errors and iterative optimization is performed.

8. The program-controlled parameter configuration method according to claim 4, further comprising: Obtaining a plurality of second errors using the plurality of groups of actual program-controlled parameters and the recommended program-controlled parameters corresponding to each group of actual program-controlled parameters; The second error is the direct difference between the actual program control parameter and the corresponding recommended program control parameter; A second parameter compensation model is constructed using the multiple second errors and iteratively optimized.

9. The program-controlled parameter configuration method according to claim 1, wherein: The method of obtaining the optimal program control parameters by using the recommended program control parameters based on the parameter compensation model includes: Based on the parameter compensation model, obtaining a parameter compensation value; The optimal program control parameters are obtained through the recommended program control parameters and the parameter compensation values.

10. A device for configuring programmable parameters for an in vivo implantable neurostimulator, the device comprising: An image acquisition module configured to acquire image data of a user; a recommended parameter acquisition module configured to pre-process the user's image data to obtain a pre-processing result, and obtain recommended program-controlled parameters based on the pre-processing result; The optimal parameter configuration module is configured to obtain the optimal program control parameters using the recommended program control parameters based on the parameter compensation model.

11. A medical system comprising: Implantable medical devices, which are implanted in the body of a user; A program-controlled parameter configuration device, wherein the program-controlled parameter configuration device implements program-controlled parameter configuration through the method described in any one of claims 1-9.

12. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 9 when executing the computer program.

13. A computer-readable storage medium storing a computer program, wherein the computer program implements the method according to any one of claims 1 to 9 when executed by a processor.

14. A computer program product, comprising at least one of a computer program and instructions, wherein when the computer program or instructions are executed by a processor, the computer program or instructions implement the method according to any one of claims 1 to 9, or implement the functions of the apparatus according to claim 10.

Images