Bioelectric data processing system for end-plate zone localization

Abstract

The present application relates to the technical field of health care informatics, and discloses a bioelectric data processing system for end-plate area positioning, which comprises a data acquisition module, a logical space mapping module, a load feature extraction module and a path correction module. The data acquisition module acquires multi-channel source bioelectric data of a sensing domain medium surface. The logical space mapping module establishes a deep axial cascade logical relationship and converts the multi-channel source bioelectric data into a logical processing machine base. The load feature extraction module extracts an energy envelope and determines a virtual mapping load. The path correction module calculates a coordinate offset correction amount according to a mapping rule of the virtual mapping load and a virtual conduction resistance parameter, updates a space coordinate definition of the logical processing machine base, and generates a target area space coordinate. The present application converts medium dynamic variables into a space conduction path logical compensation amount, eliminates coordinate artifacts induced by data intensity fluctuations, and maintains the spatial resolution of the positioning result.

Description

Technical Field

[0001] This invention relates to a bioelectrical data processing system for locating the endplate region, belonging to the field of healthcare informatics technology. Background Technology

[0002] Accurately locating the motor endplate of skeletal muscle is a prerequisite for implementing precision treatment. Currently, high-density surface electromyography (EMG) signal acquisition technology is commonly used to analyze the highest amplitude or phase reversal point of the signal to determine the spatial location of the endplate. However, human muscle tissue is a complex non-homogeneous electrical conduction medium with a volumetric conduction effect. When the subject is in a dynamic contraction state, the impedance distribution inside the tissue fluctuates in real time with muscle deformation, causing nonlinear distortion of the potential signal during spatial conduction. This makes the acquired signal characteristics unable to truly reflect the physical location of its source, resulting in a phenomenon of positional drift of the positioning center during dynamic processes.

[0003] To mitigate the aforementioned drift phenomenon, industry practices typically focus on increasing the density of the electrode array and attempting to fit the conduction path using signal processing algorithms. However, due to the nonlinear coupling between the biomechanical characteristics of human tissue and the electrical information flow, simply increasing the data sampling volume or using probabilistic model deduction cannot fundamentally eliminate signal artifacts caused by media compression and tissue heterogeneity. Against this backdrop, existing technologies suffer from the following shortcomings: 1. The positioning process relies too heavily on the assumption of tissue homogeneity, resulting in a lack of objective physical consistency in the positioning results under dynamic conditions; 2. There is a lack of physical compensation mechanisms for conduction path distortion caused by dynamic contraction, leading to spatial deviations in the positioning coordinates; 3. The identification of discrete interference signals is limited. With weaker capabilities, it is easy to misjudge crosstalk signals from adjacent muscle groups as germination signals from the target endplate region. For example, Chinese invention patent CN113769275A discloses a method and system for automatic target localization in transcranial magnetic stimulation therapy. It uses a binary search algorithm to shorten the target search time. Essentially, it is a spatial mapping based on anatomical statistical laws, simplifying bioelectrical feedback into logical judgment. The limitation of this scheme is that it assumes that the volumetric conduction medium is an isotropic static rigid body and ignores the transient impedance nonlinear drift caused by changes in muscle fiber thickness and interstitial fluid compression during active muscle contraction. Existing compensation or filtering algorithms can only eliminate noise at the signal layer and cannot correct the geometric deflection of the conduction path from the physical source level.

[0004] Therefore, the technical problem to be solved by this invention is how to construct an analytical framework that simulates the dynamic equilibrium of a physical continuum to track the essence of bioelectric signal transmission and correct the conduction path deviation caused by dynamic loads. Summary of the Invention

[0005] To address the problems mentioned in the background art, the technical solution of the present invention is as follows: A bioelectrical data processing system for locating the endplate region, the system comprising:

[0006] The data acquisition module is used to acquire multi-channel source-end bioelectrical data distributed on the surface of the target sensing domain medium.

[0007] The logical space mapping module is used to establish a cascaded logical relationship about the depth axis for multi-channel source bioelectric data according to the spatial acquisition dimension, and to convert multi-channel source bioelectric data into multiple logic processing bases;

[0008] The load feature extraction module is used to extract the real-time energy envelope of multi-channel source-end bioelectrical data and determine the real-time energy envelope as a virtual mapping load characterizing the pressure state of the target sensing domain medium.

[0009] The path correction module retrieves preset virtual conduction resistance parameters characterizing the conduction properties of the target sensing domain medium. Based on the positive correlation exponential function mapping rule between the virtual mapped load and the virtual conduction resistance parameters, it calculates the spatial coordinate offset correction amount generated by the medium in the target sensing domain. The path correction module is also used to perform differential updates on the spatial coordinate definitions of multiple logic processing bases along the depth axis using the spatial coordinate offset correction amount to offset the signal phase deviation caused by the nonlinear offset of the medium and generate the spatial coordinates of the positioning target area. The system maintains the positioning resolution of the positioning target area at 1mm by converting the dynamic variables of the source medium into logical compensation amounts for the spatial conduction path, thereby eliminating coordinate artifacts induced by data intensity fluctuations.

[0010] Preferably, the system further includes: a data arbitration module for performing consistency checks among multiple logic processing bases; the data arbitration module defines the virtual signal flow rate Φ by calculating the product of the characteristic envelope value of the multi-channel source bioelectric data and the spatial conduction velocity; after the logic space mapping module establishes the cascaded logic relationship, the data arbitration module performs real-time verification of the flow spatial coherence between adjacent logic processing bases; when a sudden pulse interference deviates from the conservation window in a certain acquisition channel, the data arbitration module reduces the calculation weight of the acquisition channel according to the flow consistency deviation to remove crosstalk noise from different sources in the same frequency band.

[0011] Preferably, the system further includes: a dynamic buffer module for establishing stability monitoring logic for multi-channel source bioelectrical data; the dynamic buffer module defines the signal consistency residual as a virtual buffer amount L; when the main signal strength drops below a preset threshold, the dynamic buffer module activates the buffer response mode, retrieves the previous cycle physical inertial parameters cached in the memory, and uses the buffer adjustment operator to perform smooth maintenance on the current spatial coordinate offset correction amount, so as to suppress random jumps in positioning coordinates during signal discontinuity.

[0012] Preferably, the system further includes: a deviation sensing module and a weight adjustment module; the deviation sensing module is used to calculate the virtual loss factor μ corresponding to each acquisition channel based on the phase lead characteristics between adjacent logic processing bases; the weight adjustment module is used to perform nonlinear gain compensation on the processing matrix of multiple logic processing bases according to the spatial distribution gradient of the virtual loss factor μ, so as to correct the positioning offset caused by the uneven contact impedance of the input nodes.

[0013] Preferred virtual loss factor The calculation logic follows the following formula: ,in, P represents the energy loss gradient between adjacent logic processing bases, and P is the virtual mapped load determined by the load feature extraction module.

[0014] Preferably, when performing spatial coordinate definition adjustment, the path correction module uses an iterative convergence algorithm to feed the virtual mapped load back to the nonlinear correction term of the virtual transmission resistance parameter in real time, so as to update the adaptive medium deformation model in real time.

[0015] Preferably, the logical space mapping module converts the arrangement matrix of the acquisition nodes into a continuum flow model at the logical level by establishing a topology mapping table, wherein each logical processing base corresponds to a depth profile of the target sensing domain medium.

[0016] Preferably, when reducing the computational weight, the data arbitration module retrieves the conservation parameters of adjacent logic processing bases to perform logical masking on abnormal nodes, and calls the cubic spline interpolation algorithm to reconstruct the feature data of the disturbed nodes.

[0017] Preferably, the data acquisition module integrates high-pass filtering logic and 50Hz adaptive notch filtering logic to filter out baseline drift and environmental electromagnetic interference before extracting multi-channel source-end bioelectrical data.

[0018] Preferably, the system further includes: a positioning output module, used to convert the spatial coordinates of the positioning target area into three-dimensional visualization vector data, and output the three-dimensional visualization vector data to an external guiding device in real time, so that the spatial coordinates are maintained within a preset accuracy range of 1mm.

[0019] Compared with the prior art, the beneficial effects of the present invention are:

[0020] 1. In bioelectric data processing, by constructing a virtual machine rack topology and the virtual tension parameter Γ between each logical rack, the spatially distributed sampled signal is converted into a continuous flow model with physical constraints. The system does not rely on searching for extreme points or phase reversal points of the signal amplitude, but instead traces the germination point of the bioelectric signal by solving the coordinates of singular points where the virtual tension gradient in the entire field approaches zero. This positioning mechanism based on physical symmetry allows the positioning process to avoid the nonlinear distortion of electrical conduction caused by the nonhomogeneity of human soft tissue, thereby locking onto the physical essence of signal conduction and ensuring that the positioning results do not drift with the fluctuation of tissue impedance.

[0021] 2. The coordinate locking module integrates a load extraction unit and a path correction unit. By defining the envelope energy of the multi-channel signal as a virtual mapped load P and combining it with the preset virtual conduction resistance parameter K to perform nonlinear compensation, the mechanism converts the physiological pressure generated by high-intensity muscle contraction into the compensation amount ΔH of the spatial conduction path. It adjusts the spatial coordinate definition of each logic rack in the depth axis in real time and corrects the signal phase deviation caused by medium deformation. This path correction logic based on physical load mapping eliminates the coordinate artifacts induced by drastic fluctuations in contraction intensity, enabling the positioning system to maintain millimeter-level spatial resolution under different muscle force levels.

[0022] 3. Utilizing the principle of second-by-second flow conservation between virtual machine racks, the virtual signal second-by-second flow rate Φ is defined by calculating the product of the signal envelope and the transmission velocity. The arbitration module performs real-time verification of the spatial consistency of flow between adjacent logical racks. When a sudden pulse interference deviates from the conservation window in a certain acquisition channel, the system reduces the calculation weight of that channel based on the flow consistency deviation, or performs logical masking on abnormal nodes using the conservation parameters of adjacent racks. This signal arbitration mechanism based on physical causal chains preserves the original phase characteristics of the main signal while removing crosstalk noise from different sources in the same frequency band, solving the problem of positioning accuracy being affected by far-field crosstalk in complex electromagnetic environments. Attached Figure Description

[0023] Fig. 1 This is a schematic diagram of the cascaded logic mapping and path correction principle architecture of the bioelectric data processing system of the present invention;

[0024] Fig. 2 This invention provides a dynamic operation flow and environmental adaptive logic diagram for the integrated impedance balancing operation system. Detailed Implementation

[0025] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto. The following embodiments are intended to illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the present invention.

[0026] A bioelectric data processing system for endplate region localization includes a data acquisition module, a logical space mapping module, a load feature extraction module, a path correction module, a data arbitration module, and a localization output module. The data acquisition module acquires multi-channel source-end bioelectric data distributed on the surface of the target sensing domain medium. The logical space mapping module establishes a cascaded logical relationship about the depth axis according to the spatial acquisition dimension and converts the signal to multiple logical processing bases. The load feature extraction module extracts the real-time energy envelope and determines the virtual mapped load. The path correction module calculates the spatial coordinate offset correction amount according to physical mapping rules and performs coordinate definition updates. The data arbitration module suppresses heterogeneous pulses through flow spatial coherence verification. Interference, combined with the three-dimensional spatial transformation of the positioning output module, jointly constructs an analytical framework simulating the dynamic equilibrium of a physical continuum, enabling the tracking of the essence of bioelectric signal transmission and the locking of the spatial coordinates of the endplate area. Addressing the volumetric conduction effect caused by human muscle tissue as a non-homogeneous electrical conduction medium, the data acquisition module integrates high-pass filtering logic and 50Hz adaptive notch filtering logic. Before acquiring multi-channel source-end bioelectric data distributed on the surface of the target sensing domain medium, baseline drift and environmental electromagnetic interference are filtered out. The logic space mapping module establishes a topological mapping table, mapping the sampling channels of the electrode array as a set of serially arranged logic processing units, which are defined as a three-dimensional data tensor unit. Its physical essence is based on the target perceptual domain medium at a specific anatomical depth. The potential intensity distribution matrix at the location is reconstructed using the following steps: First, obtain the coordinates of the two-dimensional discrete sampling nodes distributed on the skin surface. and its corresponding potential time series Secondly, using a cubic spline interpolation algorithm, based on the volumetric conduction attenuation coefficient at anatomical depth, the two-dimensional surface potential is projected onto a preset depth axial profile, forming continuous logic layers corresponding to different depths. Finally, each logic processing unit represents the cross-sectional electrical information flow model of that specific depth profile, thereby transforming discrete channel signals into a spatially continuous three-dimensional data structure. This provides a topological base for subsequent depth axial cascading and establishes cascaded logical relationships along the depth axis according to the spatial acquisition dimension. The potential intensity of each channel is reconstructed into a continuum flow model at the logical level, where each logic processing unit corresponds to the depth profile of the target sensing domain medium, allowing the positioning process to avoid interference with tissue. The system relies on the assumption of homogeneity. The logic processing base generated by the logic space mapping module is defined as a three-dimensional tensor data unit corresponding to a specific anatomical depth. By establishing a topological mapping table, the sampling channels of the electrode array are mapped as data nodes arranged in series. A cascaded logical relationship about the depth axis is established according to the spatial acquisition dimension. During the reconstruction process, the system sets the depth axis sampling step size δ to 0.5 mm. The Laplace differential operator is called to perform a second-order difference operation on the potential gradient between adjacent bases, transforming the planar distributed sampling points into a physical continuous flow model in three-dimensional space. This allows the system to track the physical trajectory of bioelectric signals conducted from deep muscle fibers to the body surface and remove non-real phase reversal points induced by discrete interference pulses.

[0027] Based on the physiological pressure generated when muscles are in a state of high-intensity contraction, which induces a physical deflection of the conduction path, the load feature extraction module calculates the real-time energy envelope of multi-channel source-end bioelectrical data. The specific calculation method for the real-time energy envelope adopts the root mean square algorithm, that is, the root mean square energy value of the signal is defined as the virtual mapped load P characterizing the pressure state of the medium. The calculation formula is as follows: ,in, The virtual mapped payload at the current moment. The amplitude of the multi-channel source-end bioelectrical data signal after filtering. The sliding window duration for calculating the preset energy envelope was ensured by explicitly employing the RMS standard algorithm to calculate load characteristics. To ensure the physical uniqueness and computational stability of values ​​under different dynamic working conditions, before the system starts the positioning process, an external pressure sensor is used to acquire the stepping mechanical load on the surface of the target sensing domain medium. This external displacement monitoring tool is connected to the input port of the data acquisition module through a physical channel with a 12-bit resolution analog-to-digital converter. During the system initialization phase, a linear mapping table between displacement physical quantities and voltage code values ​​is established by performing reference sampling with a step size of 0.50mm on the displacement stroke from 0mm to 10mm. This provides real-time and synchronous tissue deformation depth data for subsequent calculations. The data acquisition module synchronously collects the reference signal characteristics of the electrode array under different pressure gradients. The virtual conduction resistance parameter K is defined as the rate of change of virtual mapped load within a unit depth coordinate displacement. This parameter is derived from the formula... Wherein K is the virtual conduction resistance parameter in mV / mm, ΔP is the virtual mapped load change in mV, and Δd is the tissue deformation displacement obtained by the external physical displacement monitoring tool in mm. The path correction module also has a pre-stored mapping database based on human biomechanical feature classification. In non-sensor application scenarios, the system obtains the subject's preset physiological index parameters and performs feature matching in the mapping database to retrieve the corresponding reference resistance constant. and will Defined as the current virtual transmission resistance parameter Among the preset physiological parameters are the subcutaneous fat thickness at the target site of the subject. Body Mass Index and muscle cross-sectional area The mapping database is established by pre-collecting deformation characteristics of different physiological index samples under standard load. By combining a real-time calibration mode based on external displacement monitoring tools with a database retrieval mode based on physiological characteristics, the system can flexibly acquire parameters according to actual application scenarios. This allows for precise compensation of conduction path offset in different individuals and muscle regions. The path correction module combines parameter K with the real-time virtual mapping load P determined by the load feature extraction module, and calculates the spatial coordinate offset correction ΔS through ratio calculation. The formula is as follows: To achieve high-precision convergence of the path correction module under dynamic load, the iterative convergence algorithm adopts a feedback adjustment mechanism based on the least squares criterion. The adaptive medium deformation model involved follows the following mathematical evolution formula: ,in, The updated adaptive virtual conduction resistance parameters; This refers to the virtual transmission resistance parameter for the current iteration step; Real-time virtual mapping of load; This is the spatial coordinate offset correction calculated for the current iteration step; The preset convergence gain factor is used to adjust the model's response sensitivity to medium deformation; the specific execution steps of this iterative convergence algorithm are as follows: Obtain the virtual mapped load for the current period. And retrieve the preset reference transmission resistance as the initial parameter. According to the formula Calculate the spatial coordinate offset correction for the current iteration step; then calculate the result... Substituting the above adaptive medium deformation model formula, the drag parameters for the next iteration step are calculated. This allows for real-time fitting of the nonlinear deformation characteristics of the medium; the absolute residual between two adjacent iterations is calculated. and compare it with a preset convergence threshold. Perform a comparison; if the residual is less than If the algorithm converges, the final output is determined. And update the spatial coordinate definition of the logic processing base; if not satisfied, then... The system accumulates and calculates the spatial coordinate offset correction for the current iteration step, continuing the loop. This correction is used to update the spatial coordinate difference definition across multiple logic processing bases along the depth axis, offsetting signal phase deviations caused by tissue deformation. The path correction module retrieves a preset virtual conduction resistance parameter K, characterizing the medium conduction properties of the target sensing domain. Based on the positive correlation exponential function mapping rule between the virtual mapped load P and the virtual conduction resistance parameter K, the system performs a correction. This positive correlation exponential function mapping rule refers to the nonlinear exponential evolution logic between displacement response and physical load in the adaptive medium deformation model constructed within the system. Specifically, to ensure the final coordinate offset correction... It can realistically characterize the nonlinear deformation features of human soft tissue under pressure, and virtual conduction resistance parameters. It is not a fixed constant, but varies with the virtual mapped load. The increase of exhibits an exponentially decaying dynamic variable, and its mathematical expression is as follows: ,in, For virtual transmission resistance parameters that are updated in real time; The initial conduction resistance reference value of the target sensing domain medium under resting zero-load conditions; The preset viscoelastic compressibility coefficient of the medium is used to characterize the sensitivity of the effect of tissue deformation on the electrical conduction path. The virtual mapped load determined by the load feature extraction module; It is a natural constant. Supported by this exponential mapping model, the path correction module follows the formula... Calculate the spatial coordinate offset correction ΔS caused by the target sensing field medium. At this time, Substituting the expression into the calculation formula, we can see that... This results in the spatial coordinate offset correction amount. With virtual mapped load A substantial positive correlation exponential function mapping relationship is established between them, accurately offsetting the signal phase deviation caused by the nonlinear displacement of the medium. Furthermore, the spatial coordinate offset correction ΔS is used to perform differential updates on the spatial coordinates of multiple logic processing bases along the depth axis to compensate for the signal phase deviation caused by tissue deformation. Here, ΔS is the spatial coordinate offset correction, P is the virtual mapped load, and K is the virtual conduction resistance parameter. Considering that far-field crosstalk noise in complex electromagnetic environments can easily destroy the phase characteristics of the main signal, the data arbitration module determines the virtual signal flow rate per second by calculating the product of the feature envelope value A and the spatial conduction velocity v. Where Φ is the virtual signal flow rate per second, A is the feature envelope value, and v is the spatial propagation velocity. The spatial propagation velocity v is determined based on the peak time delay of the cross-correlation between adjacent sampling nodes. After the cascaded logic relationship is established, the data arbitration module performs real-time verification of the spatial coherence of the flow between adjacent logic processing bases. It monitors the spatial gradient change of Φ through a sliding window. When a flow consistency deviation of a certain acquisition channel is detected to exceed the preset 15% threshold, the system automatically reduces the calculation weight of that acquisition channel or retrieves the conservation parameters of adjacent nodes to perform logical masking on the abnormal nodes, thus removing crosstalk noise from different sources in the same frequency band.

[0028] To address signal-to-noise ratio drops caused by muscle fatigue or deep contraction intervals, the system utilizes a dynamic buffer module to establish stability monitoring logic. The signal consistency residual is defined as a virtual buffer quantity L. When the main signal strength drops below a preset threshold, the dynamic buffer module activates a buffer response mode, retrieves the physical inertia parameters from the previous sampling period cached in memory, and uses a buffer adjustment operator to smooth and maintain the current coordinate offset correction ΔS. During signal discontinuity periods, logical inertia suppresses random jumps in positioning coordinates, ensuring the physical continuity of the positioning trajectory. For positioning offsets caused by uneven electrode-skin contact impedance, the deviation sensing module calculates the virtual loss factor for each acquisition channel based on the phase lead characteristics between adjacent logic processing bases. Where μ is the virtual loss factor, Let P be the energy loss gradient between adjacent logic processing bases, and P be the virtual mapped load. The potential intensity corresponding to the logic processing base is quantized by the energy characterization value E, which is defined as the root mean square energy value of the multi-channel signals within the corresponding logic processing base. For the i-th layer logic processing base, its energy characterization value... The calculation formula is: ,in The potential sample value of the j-th logic node within the rack, where N is the total number of nodes covered by the rack; energy loss gradient. This refers to the rate of change of spatial attenuation of an electrical signal as it propagates along the depth axis, due to the depth sampling step size of the logic processing chassis. Discrete distribution, gradient Numerically, the first-order finite difference method is used to solve the problem, and its computational logic follows the following formula: ,in This represents the energy level of deeply adjacent logic processing sockets. This represents the energy level of the shallow logic processing socket. The gradient is the preset depth axis sampling step size. The vector direction points in the negative direction of the depth axis, that is, the opposite direction of conduction from the deep layers of human tissue to the body surface. Through this discretized numerical solution method, the system can transform the abstract energy field change into a calculable spatial loss parameter. The weight adjustment module performs nonlinear gain compensation on the processing matrices of multiple logic processing bases according to the spatial distribution gradient of the virtual loss factor μ, automatically offsetting the changes in the physical properties of the contact layer. In the process of acquiring the spatial coordinates of the target area after path correction and weight adjustment, the positioning output module uses the gradient search operator to perform equilibrium point optimization in the three-dimensional logic space. Virtual tension parameters are defined in the three-dimensional space formed by the logic processing bases. The logical relationship formula between this parameter and the core system parameters is as follows: ,in, Virtual tension characterizes the potential energy intensity of bioelectrical signals during medium conduction. For virtual mapping payload; For virtual transmission resistance parameters that are updated in real time; The virtual loss factor calculated by the deviation sensing module is used to reconstruct the three-dimensional tensor data of multiple logic processing bases into a scalar potential field with physical constraints during the positioning process. And calculate the virtual tension in the potential field. spatial gradient vector Since the germination point of bioelectric signals corresponds to the extreme center of potential energy distribution in physical mechanism, the system locks the coordinates of spatial singularities where the gradient vector magnitude approaches zero in the entire field by performing the second-order Laplace difference operation. This allows for reverse tracking and precise location of the physical position of the signal germination center in the three-dimensional topology, ensuring that the positioning accuracy is not affected by tissue impedance fluctuations. The system reconstructs the potential distribution between each layer of logic processing bases into a scalar potential field with physical constraints and calculates the field strength gradient vector of the scalar potential field in the three-dimensional topology. By identifying and locking the spatial singularity position where the gradient magnitude approaches zero in the entire field, the system reverse tracks the coordinates of the germination center of bioelectric signals in the deep layers of the medium. This procedure does not rely on the extreme value of the signal amplitude of a single sampling node, but utilizes the convergence characteristics of the dynamic equilibrium of the physical continuum to remove the non-real phase reversal points induced by discrete interference pulses.

[0029] Example 1: During isometric contraction of the human biceps brachii at 50% of its maximum spontaneous contraction intensity, the target sensing domain medium undergoes physical deformation under the combined effects of mechanical compression from the electrode array and muscle bulging. This causes the trajectory of the potential extreme point to shift away from the anatomical central axis as the muscle contraction intensity increases. At this time, the data acquisition module collects 128 channels of source-end bioelectrical data distributed on the skin surface and uses a frequency window of 20Hz to 500Hz to remove low-frequency mechanical artifacts and high-frequency electromagnetic interference, providing a basic data stream for subsequent logic processing base reconstruction. In this process, the logic space mapping module executes a specific channel mapping algorithm: First, the system presets the detection depth to 8mm and divides this detection depth into 16 logical levels, i.e., the interlayer spacing. Subsequently, the system calls the Laplace weighted interpolation operator, using the reciprocal of the spatial distance as weights, to map the discrete voltage values ​​of 128 physical sampling nodes on the body surface into virtual node potentials of 16 logic processing bases; each logic processing base represents the potential scalar field of that depth profile. Through this spatial reconstruction process, the system converts the two-dimensional surface electromyography distribution into a three-dimensional cascaded logic image, thereby realizing the data conversion from discrete physical channels to continuous logic flow fields.

[0030] In this contraction scenario, the load feature extraction module determines the virtual mapped load P to be 0.62mV in real time. Simultaneously, the path correction module retrieves the preset virtual conduction resistance parameter K of the biceps brachii tissue as 0.31mV / mm and calculates the spatial coordinate offset correction ΔS as 2.0mm based on the positive correlation exponential function mapping rule. This spatial coordinate offset correction ΔS directly acts on the 16 logic processing bases generated by the logic spatial mapping module. By dynamically accumulating the depth axial coordinates of each base, it offsets the volumetric conduction path distortion induced by deep fiber contraction, achieving causal synergy between the virtual mapped load and the coordinate correction logic. This ensures that the positioning coordinates remain anchored to the spatial projection position of the physical endplate area under dynamic load conditions. The virtual mapped load P = 0.62mV is calculated by the load feature extraction module using the root mean square algorithm on the current 128 channel data, representing the electrical energy envelope strength during muscle contraction. The spatial coordinate offset correction... The solution process strictly follows the physical constitutive equations: ,Right now Because the path correction module has pre-established a linear mapping table between displacement physical quantities and voltage code values, this 2.0mm correction represents the equivalent physical displacement compensation produced by the tissue under 50% shrinkage strength, and is not fabricated out of thin air; by applying this correction... In the spatial coordinate definition of the compensation logic processing base, the system successfully corrected the original positioning deviation of 3.23mm caused by medium deformation to 0.92mm, thus supporting millimeter-level spatial resolution at the physical mechanism level. In response to the crosstalk problem of electrical signals between adjacent muscle groups caused by high-intensity contraction, the data arbitration module calculates the virtual signal flow rate Φ at each sampling point and monitors the spatial gradient change of Φ between adjacent logic processing bases through a sliding window. When the crosstalk signal from the brachialis muscle causes the flow consistency deviation of a certain acquisition channel to reach 18%, which exceeds the preset threshold, the data arbitration module automatically executes the procedure of reducing the calculation weight of that channel, stripping the crosstalk noise from the positioning logic, resolving the contradiction between high signal strength and low spatial selectivity, and finally outputting three-dimensional visualization vector data that maintains a consistent physical correspondence between spatial coordinates and muscle anatomy.

[0031] Example 2: In the verification experiment of the motor endplate region of the tibialis anterior muscle, electrophysiological simulation signals of isolated skeletal muscle and sample data collected by the human tibialis anterior muscle electrode array were obtained using a physical experimental platform. This physical experimental platform has multi-channel synchronous acquisition capabilities, with a voltage measurement resolution set to 0.1 μV. Before the data enters the processing module, the original biometrics are converted into technical indexes using hash mapping through anonymization preprocessing logic, ensuring that the subsequent analysis process runs in a desensitized state. The core parameter sampling frequency... The settings take into account the balance between signal bandwidth and data processing load. When the effective bandwidth of the target bioelectric signal is distributed in the range of 20Hz to 500Hz, in order to satisfy the Nyquist sampling theorem and preserve phase characteristics, the sampling frequency is... The frequency was set to 2000Hz, and the test environment was actively superimposed with Gaussian white noise with a signal-to-noise ratio of 20dB and simulated power frequency interference harmonics with a frequency of 50Hz. The control group used an amplitude detection algorithm, while the experimental group used the bioelectric data processing system of this invention. The muscle contraction intensity was set to four gradients, namely 10%, 30%, 50%, and 70% of the maximum spontaneous contraction intensity, and pressure tests were conducted under extreme conditions where the contraction intensity exceeded 85%.

[0032] Under low load conditions with 10% contraction intensity, the experimental group determined a virtual mapped load P of 0.12mV, and the spatial coordinate offset correction ΔS determined by the path correction module was 0.41mm. At this time, the positioning center error was 0.84mm. As the contraction intensity increased to a moderate load of 50%, muscle bulging caused signal phase drift, and the energy envelope extracted in real time by the experimental group showed an upward trend. The virtual mapped load P increased to 0.65mV, and the virtual conduction resistance parameter K of the tibialis anterior muscle retrieved by the system was 0.28mV / mm, according to the formula... The calculated spatial coordinate offset correction ΔS was 2.32 mm. After updating the coordinates of the logic processing base, the positioning error of the control group (without the path correction module) increased to 3.23 mm, while the positioning resolution of the experimental group remained at 0.92 mm. When crosstalk signals of the same frequency band were introduced into the environment, the system detected abnormal fluctuations in the feature envelope value A. The data arbitration module identified a spatial coherence deviation of 21% in the virtual signal per-second flow rate Φ and automatically executed a procedure to reduce the calculation weight of this channel. The calculation formula is as follows: Where Φ is the virtual signal flow rate per second, A is the characteristic envelope value, and v is the spatial conduction velocity. If the data arbitration module is removed, the crosstalk signal is included in the energy envelope, causing the virtual mapped load P to be artificially high, and the spatial coordinate offset correction ΔS produces an additional deviation of 1.52 mm. When the contraction intensity exceeds 85%, the growth rate of the virtual mapped load P slows down and tends to flatten with the increase of intensity. The data shows that the tissue strain enters the saturation zone, and the positioning resolution fluctuates from 0.98 mm to 1.16 mm. This data shows the performance inflection point of the system under extreme overload conditions and provides experimental basis for determining the value range of the virtual conduction resistance parameter K. Finally, the experimental group stabilized the positioning resolution of the tibialis anterior muscle motor endplate area within 1 mm.

[0033] Example 3: This example combines Figs. 1-2 The bioelectrical data processing system used for endplate region localization is described, such as... Fig. 1 As shown, the data processing flow begins with the sensing domain medium surface data acquisition module, which acquires multi-channel source-end bioelectrical data. This data flows to the logic space mapping module, which establishes a depth-axis cascaded logic relationship and converts the data into a logic processing base. Simultaneously, the load feature extraction module is responsible for extracting the energy envelope and determining the virtual mapping load. This virtual mapping load is transmitted as an input to the path correction module. The path correction module combines the input virtual conduction resistance parameters, calculates the coordinate offset correction amount according to predetermined rules, and updates the spatial coordinate definition of the logic processing base, thereby ultimately generating and outputting the spatial coordinates of the target area.

[0034] like Fig. 2 As shown, the operator performs a skin contact layer impedance balancing operation as the initiator. This operation is accompanied by the injection of a constant current detection signal. For the target sensing domain medium system, after acquiring multi-channel source-end bioelectrical data, the process of establishing a deep axial cascade logic relationship and extracting the energy envelope and determining the virtual image load is carried out in parallel. The data results of the above process are aggregated and used to calculate the coordinate offset correction and update the logic base, thereby generating the spatial coordinates of the target area. During this process, the environmental acquisition unit operates independently. When the background noise power fluctuation is detected, the logic step of adaptively adjusting the consistency deviation threshold is automatically triggered.

[0035] Example 4: In the scenario of initializing the parameters of the tibialis anterior muscle of the subject, due to the physical differences in the physiological cross-sectional area of ​​skeletal muscle and the thickness of subcutaneous fat among different subjects, the system uses a pressure sensor to acquire the stepping mechanical load acting on the surface of the target sensing domain medium before starting the positioning process. The data acquisition module simultaneously acquires the reference signal characteristics of the 64-channel electrode array under different pressure gradients. During this process, the system uses a hash mapping algorithm to isolate the original electrophysiological signal characteristics from the subject's identity, ensuring that the data processing runs in a desensitized state.

[0036] The system defines the virtual transmission resistance parameter K as the rate of change of the virtual mapped load within a unit depth coordinate displacement, and its calculation logic follows the formula. Where K is the virtual conduction resistance parameter, in mV / mm; ΔP is the change in virtual mapped load, in mV; and Δd is the tissue deformation displacement obtained by an external physical displacement monitoring tool, in mm. This provides the path correction module with physically traceable parameter input. The logic space mapping module establishes a horizontal mapping reference based on the geometric arrangement coordinates of the electrode array and converts multi-channel source-end bioelectrical data into 10-level logic processing bases. During this reconstruction process, the system sets the sampling step size δ along the depth axis to 0.5mm and calls the Laplace differential operator to perform a second-order difference operation on the potential gradient between adjacent logic processing bases, thereby transforming the planar sampling points into a three-dimensional physical continuous flow model. This allows the system to track the physical trajectory of bioelectrical signals from deep muscle fibers to the body surface at the logical level. To correct the technical parameters used in the data arbitration module to determine the spatial coherence of the flow, the system introduces adaptive adjustment logic based on signal-to-noise ratio changes. The system performs a root mean square operation on the 2s reference background electrical signal obtained during the initialization phase, defining it as the background noise power. Simultaneously, the peak value of the first typical action potential envelope captured by the system is defined as the reference signal power. Reference signal-to-noise ratio Defined as the ratio of the two, that is During system operation, real-time signal-to-noise ratio The average energy of the current sampled sequence is estimated in real time using a sliding window. The ratio is obtained by defining the standard deviation of the background noise of the current environmental acquisition unit as the judgment benchmark σ. When the system detects that the background noise power in the environment fluctuates, causing the real-time signal-to-noise ratio to drop, the operator is automatically invoked. A correction is performed on the initial consistency deviation threshold of 15%. Set as It is based on the per-second flow rate of multi-channel bioelectrical signals of human skeletal muscle under quasi-static contraction. The inherent physiological coefficient of variation, and the physical significance of using a logarithmic function operator to perform correction, lies in: following the principle of dynamic range compression in signal processing, making the threshold... As the signal-to-noise ratio decreases, the logarithmic function exhibits a non-linear monotonically increasing behavior. This allows for a moderate relaxation of the conservation decision window in noisy environments, ensuring the system does not misjudge legitimate signals due to environmental interference. Simultaneously, the slow growth characteristic of the logarithmic function effectively suppresses the divergence of the threshold under extreme interference, where T is the corrected flow consistency deviation threshold. This is the initial consistency deviation threshold. As the reference signal-to-noise ratio, To improve the real-time signal-to-noise ratio, this adaptive adjustment procedure enhances the system's accuracy in identifying crosstalk pulses from different sources.

[0037] Example 5: In a field deployment scenario of performing skin contact layer impedance balancing on a 64-channel array electrode, the system initiates a pre-calibration procedure to determine the initial virtual loss factor μ corresponding to each channel. The data acquisition module injects a constant current detection signal with a frequency of 1000Hz and an amplitude of 10μA into each electrode. The deviation sensing module calculates the loop potential difference of each acquisition node relative to the full array reference potential. The weight adjustment module establishes a normalized mapping table based on the initial potential gradient, extracts the deviation vector of each acquisition channel relative to the array mean potential, and uses a nonlinear gain compensation unit to adjust the background noise baseline of each channel to a uniform 5μV root mean square level, eliminating the asymmetry of the original data caused by the uneven distribution of the skin stratum corneum thickness.

[0038] When the system encounters electromagnetic pulse interference in the interventional treatment room, the logic space mapping module performs discrete grid initialization along the depth axis, dividing the detection depth from 0mm to 5mm into 10 levels of logic processing bases, with layer spacing... Set as The data arbitration module continuously acquires the raw background field data stream for 2 seconds without muscle contraction. It determines the standard deviation σ of the background noise of the current environment by calculating the time variation coefficient of the feature envelope value A within the sampling window. The system updates the weight allocation of the flow consistency deviation threshold in real time based on the currently locked judgment benchmark, and performs Laplace second-order difference iteration operation between each layer of logic processing base to stabilize the spatial continuity of the virtual signal flow rate Φ in the deep axial transmission process.

[0039] Example 6: In the case of performing multi-channel array electrode electrical balance adjustment on the gastrocnemius muscle motor endplate region, the system injects a constant current detection signal with a frequency of 1000Hz and an amplitude of 10μA into the sensing domain medium through the array electrodes. The deviation sensing module calculates the loop potential difference of each acquisition node relative to the full array reference potential and determines the initial virtual loss factor μ corresponding to each channel. The weight adjustment module establishes a normalized mapping table based on the initial potential gradient, extracts the deviation vector of each acquisition channel relative to the array mean potential, and uses a nonlinear gain compensation unit to adjust the background noise baseline of each channel to a uniform 5μV root mean square level to compensate for the signal phase truncation caused by the uneven distribution of individual skin electrolytes.

[0040] The logical space mapping module performs hierarchical adaptive calibration of the depth axis for sensing domain media of different thicknesses, and obtains the subcutaneous tissue thickness through displacement monitoring tools. To accurately reflect the conduction characteristics of bioelectrical signals in non-homogeneous tissues, the system performs non-uniform hierarchical partitioning logic based on the acoustic impedance gradient of the target sensing domain medium. Specifically, the system does not perform simple spatial division, but rather divides the total thickness... As a global constraint, it identifies the anatomical interface between subcutaneous fat and deep muscle; within the neighborhood of this anatomical interface, the system automatically reduces the interlayer spacing. To increase the sampling density of the logic processing unit, the interlayer spacing is correspondingly expanded in regions with uniform organizational properties. This dynamic hierarchical allocation strategy based on tissue heterogeneity ensures that the logic processing base has higher spatial resolution in interface regions with severe signal distortion. This guides the reconstructed continuum flow model to more accurately fit the nonlinear flipping characteristics of the potential, ensuring that the spatial vector of the target area remains within 1mm accuracy under dynamic physiological loads. When the dynamic buffer module monitors the real-time signal-to-noise ratio... When the signal drops below the 3dB threshold, the system automatically activates the physical inertia maintenance mode. It retrieves the spatial coordinate offset correction ΔS from the previous sampling period cached in the memory to fill the current positioning solution sequence. By using logical inertia to suppress the random coordinate jumps induced by the idle calculation step during the weak signal period, the spatial vector of the positioning target area is always maintained within the 1mm accuracy range under dynamic physiological load.

[0041] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0042] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A bioelectrical data processing system for locating the endplate region, characterized in that the system... include: The data acquisition module is used to acquire multi-channel source-end bioelectrical data distributed on the surface of the target sensing domain medium. The logical space mapping module is used to establish a cascaded logical relationship about the depth axis for multi-channel source bioelectric data according to the spatial acquisition dimension, and to convert multi-channel source bioelectric data into multiple logic processing bases; The load feature extraction module is used to extract the real-time energy envelope of multi-channel source-end bioelectrical data and determine the real-time energy envelope as a virtual mapping load characterizing the pressure state of the target sensing domain medium. The path correction module is used to retrieve preset virtual conduction resistance parameters that characterize the conduction properties of the target sensing domain medium, and calculate the spatial coordinate offset correction amount generated by the target sensing domain medium based on the positive correlation exponential function mapping rule between the virtual mapping load and the virtual conduction resistance parameters. The path correction module is also used to perform differential updates on the spatial coordinate definitions of multiple logic processing bases in the depth axis using spatial coordinate offset correction amounts, in order to offset the signal phase deviation caused by nonlinear offset of the medium and generate the spatial coordinates of the positioning target area; wherein, the system maintains the positioning resolution of the positioning target area at 1mm by converting the dynamic variables of the source medium into the logical compensation amount of the spatial transmission path.

2. The bioelectrical data processing system for locating the endplate region according to claim 1, characterized in that, The system also includes: a data arbitration module for performing consistency checks between multiple logic processing bases; the data arbitration module defines the virtual signal flow rate Φ by calculating the product of the characteristic envelope value of the multi-channel source bioelectric data and the spatial conduction velocity; after the logic space mapping module establishes the cascaded logic relationship, the data arbitration module performs real-time verification of the flow spatial coherence between adjacent logic processing bases; when a sudden pulse interference deviates from the conservation window in a certain acquisition channel, the data arbitration module reduces the calculation weight of that acquisition channel according to the flow consistency deviation in order to remove crosstalk noise from different sources in the same frequency band.

3. A bioelectrical data processing system for locating the endplate region according to claim 2, characterized in that, The system also includes: a dynamic buffer module, used to establish stability monitoring logic for multi-channel source bioelectric data; the dynamic buffer module defines the signal consistency residual as a virtual buffer amount L; when the main signal strength drops below a preset threshold, the dynamic buffer module activates the buffer response mode, retrieves the previous cycle physical inertial parameters cached in the memory, and uses the buffer adjustment operator to perform smooth maintenance on the current spatial coordinate offset correction amount, so as to suppress random jumps in positioning coordinates during signal discontinuity.

4. A bioelectrical data processing system for locating the endplate region according to claim 1, characterized in that, The system also includes: a deviation sensing module and a weight adjustment module; the deviation sensing module is used to calculate the virtual loss factor μ corresponding to each acquisition channel based on the phase lead characteristics between adjacent logic processing bases; the weight adjustment module is used to perform nonlinear gain compensation on the processing matrix of multiple logic processing bases according to the spatial distribution gradient of the virtual loss factor μ, so as to correct the positioning offset caused by the uneven contact impedance of the input nodes.

5. A bioelectrical data processing system for locating the endplate region according to claim 4, characterized in that, Virtual loss factor The calculation logic follows the following formula: ,in, P represents the energy loss gradient between adjacent logic processing bases, and P is the virtual mapped load determined by the load feature extraction module.

6. A bioelectrical data processing system for locating the endplate region according to claim 1, characterized in that, When performing spatial coordinate definition calibration, the path correction module uses an iterative convergence algorithm to feed back the virtual mapped load to the nonlinear correction term of the virtual transmission resistance parameter in real time, so as to update the adaptive medium deformation model in real time.

7. A bioelectrical data processing system for locating the endplate region according to claim 1, characterized in that, The logical space mapping module converts the arrangement matrix of the acquisition nodes into a continuum flow model at the logical level by establishing a topology mapping table, where each logical processing base corresponds to a depth profile of the target sensing domain medium.

8. A bioelectrical data processing system for locating the endplate region according to claim 2, characterized in that, When reducing the computational weight, the data arbitration module retrieves the conservation parameters of adjacent logic processing bases to perform logical masking on abnormal nodes and calls the cubic spline interpolation algorithm to reconstruct the feature data of the disturbed nodes.

9. A bioelectrical data processing system for locating the endplate region according to claim 1, characterized in that, The data acquisition module integrates high-pass filtering logic and 50Hz adaptive notch filtering logic to filter out baseline drift and environmental electromagnetic interference before extracting multi-channel source bioelectrical data.

10. A bioelectrical data processing system for locating the endplate region according to claim 1, characterized in that, The system also includes a positioning output module, which converts the spatial coordinates of the positioning target area into three-dimensional visualization vector data and outputs the three-dimensional visualization vector data to an external guidance device in real time.

Images