Implantable spinal cord stimulation system and method of adjusting stimulation parameters thereof

Abstract

The application provides an implantable spinal cord stimulation system and an adjustment method of stimulation parameters thereof, the adjustment method comprising: acquiring a nerve conduction gain value and a complex impedance measurement value of a target electrode contact at multiple frequencies; sequentially performing impedance causality verification processing and hierarchical relaxation time distribution (DRT) decoupling processing on the complex impedance measurement value to obtain a first parameter for characterizing electrochemical characteristics of a tissue interface; autonomously identifying an electrode-tissue interface physiological state through an unsupervised clustering model according to a time sequence variation characteristic of the first parameter and the nerve conduction gain value; and determining a control instruction based on a physiological state identifier output by the clustering model to adjust a current stimulation parameter and a sampling period.

Description

Technical Field

[0001] This application relates to the field of neuromodulation technology, and more specifically, to an implantable spinal cord electrical stimulation system and a method for adjusting its stimulation parameters. Background Technology

[0002] Spinal cord stimulation (SCS) technology delivers weak electrical pulses to the spinal cord nerves via implanted electrodes, and has become an important clinical neuromodulation technique for treating chronic intractable pain, lower limb ischemic diseases, and other conditions. However, in the long-term use of implanted spinal cord stimulation systems, the physiological state of the electrode-tissue interface undergoes complex dynamic changes, which significantly affect the stimulation effect.

[0003] In current clinical practice, doctors typically rely on patient feedback and regular imaging examinations to manually adjust stimulation parameters. However, this method suffers from significant lag and subjectivity. Furthermore, minute electrode shifts that may occur during daily activities are difficult to detect in a timely manner. If these changes are not adjusted promptly, they can lead to reduced analgesic effects (treatment window drift) or, in severe cases, abnormal stimulation diffusion causing sensory abnormalities (such as tingling), or even damage to nerve tissue.

[0004] Therefore, the parameter adjustments in the existing scheme lack specificity and cannot achieve adaptive parameter adjustment. Summary of the Invention

[0005] This application addresses the shortcomings of existing methods by proposing an implantable spinal cord stimulation system and a method for adjusting its stimulation parameters. The aim is to solve the problem that existing solutions lack targeted parameter adjustments and cannot achieve adaptive parameter adjustment.

[0006] In a first aspect, embodiments of this application provide an implantable spinal cord stimulation system, comprising: Multiple electrodes implanted in the spinal cord; An electrical stimulation driving module is used to apply an electrical stimulation signal to the target electrode contact among the plurality of electrodes; Impedance measurement module is used to acquire the complex impedance measurement values ​​of the target electrode contact at multiple frequencies; The signal acquisition module is used to periodically acquire spinal cord evoked action potential (ECAP) signals. The data processing module is used to obtain the neural conduction gain value based on the ECAP signal, and to perform impedance causality verification processing and layered relaxation time distribution (DRT) decoupling processing on the complex impedance measurement value in sequence to obtain the first parameter used to characterize the electrochemical properties of the tissue interface. The signal generation module is used to autonomously identify the physiological state of the electrode-tissue interface through an unsupervised clustering model based on the temporal variation characteristics of the first parameter and the neural conduction gain value. The physiological state of the electrode-tissue interface includes at least one of the following: stable period, acute edema period, fibrous encapsulation formation period, and electrode displacement warning period. The instruction determination module is used to determine control instructions based on the physiological state identifiers output by the clustering model; The parameter adjustment module is used to control the electrical stimulation drive module based on the control command to adjust the current stimulation parameters and the sampling period of the ECAP signal; The stimulation parameters include: stimulation voltage, and / or stimulation pulse width.

[0007] Secondly, embodiments of this application provide a method for adjusting stimulation parameters of an implantable spinal cord stimulation system, comprising: Acquire neural conduction gain values ​​and complex impedance measurements of the target electrode contact at multiple frequencies; The complex impedance measurements are sequentially subjected to impedance causality verification and layered relaxation time distribution (DRT) decoupling to obtain the first parameter used to characterize the electrochemical properties of the tissue interface. Based on the temporal variation characteristics of the first parameter and the neural conduction gain value, the physiological state of the electrode-tissue interface is autonomously identified through an unsupervised clustering model. The physiological state of the electrode-tissue interface includes at least one of the following: stable period, acute edema period, fibrous encapsulation formation period, and electrode displacement warning period. Control commands are determined based on the physiological state identifiers output by the clustering model to adjust the current stimulation parameters and sampling period; The stimulation parameters include: stimulation voltage, and / or stimulation pulse width; The sampling period is the period used to acquire spinal cord evoked action potential (ECAP) signals, and the neural conduction gain value is obtained based on the ECAP signals.

[0008] Thirdly, embodiments of this application also disclose a signal processing system, including the implantable spinal cord stimulation system as described in the first aspect.

[0009] Fourthly, embodiments of this application also disclose a computer-readable storage medium storing a computer program that, when executed by a processor, implements one or more of the adjustment methods described in the embodiments of the second aspect of this application.

[0010] Fifthly, embodiments of this application also disclose a computer program product, including a computer program that, when executed by a processor, implements one or more of the adjustment methods described in the embodiments of the second aspect of this application.

[0011] The beneficial technical effects of the technical solutions provided in this application include: By performing feature decoupling on the acquired complex impedance measurements at multiple frequencies, key parameters characterizing the interface properties can be obtained. Based on the temporal variation characteristics of these parameters, accurate judgment of the interface's physiological state can be achieved. Furthermore, based on the signals corresponding to different interface physiological states, adaptive closed-loop adjustment of the stimulation parameters can be performed.

[0012] Additional aspects and advantages of this application will be set forth in part in the description which follows, and will become apparent from the description or may be learned by practice of this application. Attached Figure Description

[0013] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a schematic diagram of the structure of an implantable spinal cord stimulation system provided in an embodiment of this application; Figure 2 A flowchart illustrating a method for adjusting stimulation parameters of an implantable spinal cord electrical stimulation system provided in this application embodiment; Figure 3 This is a schematic diagram of the signal processing system provided in an embodiment of this application. Detailed Implementation The embodiments of this application are described below with reference to the accompanying drawings. It should be understood that the embodiments described below with reference to the accompanying drawings are exemplary descriptions for explaining the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions of the embodiments of this application.

[0014] Those skilled in the art will understand that, unless specifically stated otherwise, the terms "described" and "the" as used herein may also include plural forms. It should be further understood that the term "comprising" as used in this application's specification means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude implementations of other features, information, data, steps, operations, elements, components, and / or combinations thereof supported by this art. It should be understood that when we say an element is "connected" or "coupled" to another element, the element may be directly connected or coupled to the other element, or it may mean that the element and the other element are connected through an intermediate element. Furthermore, "connected" or "coupled" as used herein may include wireless connections or wireless coupling. The term "and / or" as used herein refers to at least one of the items defined by the term; for example, "A and / or B" may be implemented as "A," or as "B," or as "A and B."

[0015] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0016] Spinal cord stimulation (SCS) delivers weak electrical pulses to the spinal nerves via implanted electrodes, treating chronic intractable pain, lower limb ischemic diseases, and other conditions, and has become an important clinical neuromodulation technique. However, after electrode implantation, the surrounding tissues undergo physiological processes such as acute edema and fibrous encapsulation, leading to changes in the electrical properties of the electrode-tissue interface. For example, acute edema reduces tissue resistivity, while fibrous encapsulation significantly increases charge transfer resistance. Furthermore, minute electrode displacements caused by daily activities are difficult to detect in a timely manner. If these changes are not adjusted promptly according to the stimulation parameters, they can lead to weakened analgesic effects (treatment window drift) or, in severe cases, abnormal stimulation diffusion causing sensory abnormalities (such as tingling), and even damage to nerve tissue.

[0017] Current technologies cannot effectively distinguish the contributions of different physical mechanisms such as solution resistance, charge transfer resistance, and interfacial dispersion effects. Therefore, it is difficult to accurately determine whether the interface is in a state of edema, fibrosis, or displacement, resulting in a lack of targeted parameter adjustments and hindering the achievement of truly intelligent closed-loop control.

[0018] The technical solution of this application and how it solves the above-mentioned technical problems are described in detail below with specific embodiments. It should be noted that the following embodiments can be referenced, borrowed, or combined with each other, and the same terms, similar features, and similar implementation steps in different embodiments will not be described again.

[0019] The implantable spinal cord stimulation system and its stimulation parameter adjustment method proposed in this application will be described in detail below with reference to the accompanying drawings.

[0020] In some embodiments, an implantable spinal cord stimulation system is provided, such as Figure 1 As shown, the implantable spinal cord stimulation system 10 includes: multiple electrodes 11 implanted in the spinal cord, an impedance measurement module 12, a data processing module 13, a signal generation module 14, a command determination module 15, and a parameter adjustment module 16, as well as an electrical stimulation drive module 17 and a signal acquisition module 18. The electrical stimulation drive module 17, multiple electrodes 11, impedance measurement module 12, data processing module 13, signal generation module 14, command determination module 15, and parameter adjustment module 16 are connected sequentially. The data processing module 13 is also connected to the signal acquisition module 18. The parameter adjustment module 16 is also connected to the electrical stimulation drive module 17. The command determination module 15 is also connected to the impedance measurement module 12 and the signal acquisition module 18.

[0021] In some embodiments, the electrostimulation driving module 17 is used to apply an electrostimulation signal to a target electrode contact among the plurality of electrodes; the impedance measurement module 12 is used to acquire the complex impedance measurement values ​​of the target electrode contact at multiple frequencies; the signal acquisition module 18 is used to periodically acquire spinal cord evoked action potential (ECAP) signals. The data processing module 13 is used to obtain a neural conduction gain value based on the ECAP signal, and sequentially perform impedance causality verification processing and hierarchical relaxation time distribution (DRT) decoupling processing on the complex impedance measurement value to obtain a first parameter for characterizing the electrochemical properties of the tissue interface; the signal generation module 14 is used to autonomously identify the physiological state of the electrode-tissue interface through an unsupervised clustering model based on the temporal variation characteristics of the first parameter and the neural conduction gain value; the instruction determination module 15 is used to determine a control instruction based on the physiological state identifier output by the clustering model; the parameter adjustment module 16 is used to control the electrostimulation driving module 17 based on the control instruction to adjust the current stimulation parameters and the sampling period of the ECAP signal.

[0022] Optionally, the aforementioned physiological states of the electrode-tissue interface include at least one of the following: stable phase, acute edema phase, fibrous encapsulation formation phase, and electrode displacement warning phase.

[0023] Optionally, the above stimulation parameters may include: stimulation voltage, and / or stimulation pulse width.

[0024] In some embodiments, the first parameter includes: solution resistance, charge transfer resistance, and dispersion index of a constant phase element.

[0025] In some embodiments, when the signal generation module 14 identifies the physiological state as a stable period through an unsupervised clustering model, it is specifically used to: calculate the change coefficients of the solution resistance, the charge transfer resistance, and the diffusion index over a continuous preset monitoring period; when the change coefficients of the solution resistance, the charge transfer resistance, and the diffusion index are all less than their respective preset stable thresholds, the clustering model outputs a stable period status identifier.

[0026] In some embodiments, if the clustering model outputs a stable state identifier, the control command is used to instruct the periodic monitoring of the complex impedance measurement value, and the acquisition frequency of the complex impedance measurement value is reduced from a first frequency to a second frequency, and the sampling period of the ECAP signal is extended from a first duration to a second duration.

[0027] In some embodiments, when the signal generation module 14 identifies the physiological state as acute edema phase through an unsupervised clustering model, it is specifically used to: obtain a rate of change of the solution resistance relative to an initial reference value within a first preset time period; when the first rate of change exceeds a first threshold and the fluctuation amplitude of the neural conduction gain value is less than a preset interference threshold, the clustering model outputs an acute edema phase status identifier.

[0028] In some embodiments, if the clustering model outputs an acute edema phase status identifier, the control command is used to instruct the current stimulation voltage to be adjusted to the target stimulation voltage.

[0029] Optionally, the target stimulation voltage is determined based on the voltage adjustment amount and the boundary value of the safe voltage range, wherein the voltage adjustment amount is obtained based on the first rate of change and a preset compensation coefficient.

[0030] Optionally, the compensation coefficient can range from 0.3 to 0.7.

[0031] In some embodiments, the data processing module 13 is further configured to: obtain an initial stimulation voltage based on the current stimulation voltage and the voltage adjustment amount; and, If the initial stimulation voltage is within the safe voltage range, then the initial stimulation voltage is determined as the target stimulation voltage; If the initial stimulation voltage is less than the lower limit of the safe voltage range, then the lower limit is determined as the target stimulation voltage; If the initial stimulation voltage is greater than the upper limit of the safe voltage range, then the upper limit is determined as the target stimulation voltage.

[0032] In some embodiments, when the signal generation module 14 identifies the physiological state as the fiber encapsulation formation stage through an unsupervised clustering model, it is specifically used to: monitor the rising rate of the charge transfer resistance and the decreasing trend of the diffusion index, as well as the changing trend of the neural conduction gain value; when the changing trend of the neural conduction gain value is a continuous slow decreasing trend, the rising rate of the charge transfer resistance exceeds a second threshold, and the diffusion index is lower than a preset diffusion critical value, the clustering model outputs a fiber encapsulation formation stage state identifier.

[0033] In some embodiments, the signal generation module 14 is further configured to: divide the fiber wrapping formation period into a light wrapping period, a medium wrapping period, and a heavy wrapping period according to the relationship between the absolute value of the charge transfer resistance and the first resistance threshold and the second resistance threshold, and output the corresponding grading identifier.

[0034] Optionally, if the grading indicator corresponds to a mild envelopment period, the control command is used to instruct that the current stimulation voltage be kept constant and the current stimulation pulse width be increased within a preset compensation range.

[0035] Optionally, if the grading identifier corresponds to the moderate encapsulation period, the control command is used to instruct the current stimulation voltage and the current stimulation pulse width to be synchronously increased according to a preset compensation gradient.

[0036] Optionally, if the grading identifier corresponds to the period of severe envelopment, the control command is used to instruct the current stimulation parameters to be adjusted to a preset safe upper limit value and to trigger a second alarm signal.

[0037] In some embodiments, when the signal generation module 14 identifies the physiological state as an electrode displacement warning period through an unsupervised clustering model, it is specifically used to: obtain the rate of change of charge transfer resistance of adjacent first electrode contacts and second electrode contacts in the target electrode contacts within the same time period; when the second rate of change of charge transfer resistance of the first electrode contact is positive, and the third rate of change of charge transfer resistance of the second electrode contact is negative, and the absolute value of the difference between the second rate of change and the third rate of change exceeds a preset displacement judgment threshold, the clustering model outputs an electrode displacement warning period status identifier.

[0038] In some embodiments, the signal generation module 14 is specifically used to: obtain the rate of change of charge transfer resistance of adjacent first electrode contact and second electrode contact in the target electrode contact within the same time period; when the second rate of change of charge transfer resistance of the first electrode contact is negative, and the third rate of change of charge transfer resistance of the second electrode contact is positive, and the absolute value of the difference between the second rate of change and the third rate of change exceeds a preset displacement judgment threshold, the clustering model outputs an electrode displacement warning period status identifier.

[0039] In some embodiments, the first electrode contact and the second electrode contact are two adjacent electrode contacts among the target electrode contacts.

[0040] In some embodiments, if the clustering model outputs an electrode displacement warning period status identifier, the control command is also used to instruct switching to the backup electrode contact and maintaining the current stimulation parameters for stimulation. At the same time, a first alarm signal is triggered and the ECAP signal and the complex impedance measurement value of the backup electrode contact are monitored.

[0041] In some embodiments, when performing the layered relaxation time distribution (DRT) decoupling process, the data processing module 13 is specifically used to: perform deconvolution processing on the complex impedance measurement value after impedance causality verification using a relaxation time distribution algorithm to obtain a relaxation time distribution spectrum, and extract the peak parameters corresponding to different characteristic frequency bands in the relaxation time distribution spectrum as the first parameter.

[0042] In some embodiments, when performing impedance causality verification processing, the data processing module 13 is specifically used for: The complex impedance measurements are preprocessed using sliding window smoothing and frequency domain symmetric extrapolation. A lightweight linear Kramers-Kronig fitting model was used to fit the preprocessed complex impedance measurements, and the normalized residual between the measured impedance and the theoretical fitted impedance was calculated. The validity of impedance measurements is determined based on the residual threshold of the implanted electrode: if the residual of a single frequency point exceeds the residual threshold, the complex impedance measurement at that frequency point is determined to be an invalid complex impedance measurement and is discarded. The determined valid complex impedance measurement is used for layered DRT decoupling processing. If the proportion of invalid complex impedance measurements exceeds a predetermined proportion, the complex impedance measurements are deemed invalid as a whole, and the complex impedance measurements are re-acquired.

[0043] It should be understood that the controller in the above embodiments may include a processor. The processor may include a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0044] The implantable spinal cord stimulation system provided in this application embodiment can achieve the following: Figure 2 The various processes implemented in the method embodiments shown will not be described again here to avoid repetition.

[0045] Based on the same inventive concept, embodiments of this application provide a method for adjusting stimulation parameters of an implantable spinal cord stimulation system. This implantable spinal cord stimulation system may include the implantable spinal cord stimulation system described in the above embodiments. Figure 2 As shown, the method includes: S1. Obtain the neural conduction gain value and the complex impedance measurement value of the target electrode contact at multiple frequencies.

[0046] In some embodiments, the neural conduction gain value can be obtained based on the acquisition of spinal cord evoked action potential (ECAP) signals. Optionally, the ECAP signal can be obtained by applying a low-intensity probe pulse (the amplitude of which is lower than the clinical therapeutic stimulus amplitude and does not produce abnormal somatosensory interference) to the excitation contacts within the electrode array to induce a compound action potential in the dorsal column nerve of the spinal cord, and then acquiring the signal through the recording contacts within the electrode array. Both the excitation contacts and the recording contacts are considered target electrode contacts.

[0047] In this embodiment, the neural conduction gain value is calculated by extracting the peak amplitude of the ECAP signal waveform. Optionally, the neural conduction gain value is a standardized quantitative indicator, which can be determined as the ratio of the peak amplitude of the ECAP signal under the same probe stimulus intensity to the baseline ECAP amplitude at the initial implantation stage.

[0048] It should be noted that if the nerve conduction gain value is fixed at 1, it indicates that the spinal cord nerve conduction pathway is intact and the electrode and nerve are well coupled. If the nerve conduction gain value decreases slowly and continuously, it indicates that a fibrous capsule has formed on the electrode surface, reducing the efficiency of electric field penetration and increasing the difficulty of nerve activation. If the nerve conduction gain value decreases abruptly, it indicates that the distance between the electrode contact and the spinal cord tissue changes abruptly, and there is a risk of displacement. If the nerve conduction gain value fluctuates slightly and randomly without a long-term decay trend, it indicates that there is only interference from tissue edema mediators and the nerve axonal conduction function is not impaired.

[0049] In some embodiments, the multiple frequencies may include multiple frequencies in the low-frequency band, mid-frequency band, and high-frequency band within the 10Hz to 10kHz frequency range. Optionally, in embodiments of this application, the 10Hz to 10kHz frequency range is divided into: a high-frequency band (e.g., 1kHz to 10kHz) for characterizing solution resistance characteristics, a mid-frequency band (e.g., 100Hz to 1kHz) for characterizing charge transfer characteristics, and a low-frequency band (e.g., 10Hz to 100Hz) for characterizing interfacial diffusion and dispersion characteristics.

[0050] In some embodiments, by applying an electrical stimulation signal within a frequency range of 10 Hz to 10 kHz to the target electrode contact, the impedance amplitude and phase angle of the target electrode contact at multiple frequencies are acquired to obtain complex impedance measurements. For example, within a preset monitoring period, AC signals with an amplitude of 100 μV at multiple frequencies (10 logarithmically distributed frequencies within a frequency range of 1 Hz to 100 kHz) are applied to the target electrode contact, and the voltage and current response signals at each frequency are acquired synchronously. Complex impedance measurements at each frequency are obtained through lock-in amplification or a fast Fourier transform algorithm. It should be understood that this measurement process does not affect normal neurostimulation therapy.

[0051] In some embodiments, the complex impedance measurement includes impedance magnitude and phase angle. Optionally, the real and imaginary parts of the complex impedance can be calculated based on the impedance magnitude and phase angle for subsequent characteristic decoupling processing.

[0052] In some embodiments, the frequency at which the complex impedance measurement value is obtained can be a first frequency.

[0053] It should be noted that the acquisition frequency of complex impedance measurements refers to the time interval between two monitoring measurements. For example, a decrease in the acquisition frequency indicates a larger time interval between two monitoring measurements.

[0054] In some embodiments, a target electrode contact refers to an electrode contact implanted in the patient's body (such as in the epidural space of the spinal cord or a specific nucleus in the brain) for applying electrical stimulation pulses.

[0055] In some embodiments, the implantation location of the electrode can be determined by image fusion navigation technology, and a specific contact point can be selected as the target electrode contact point.

[0056] In some embodiments, intraoperative CT images are rigidly registered with preoperative MRI images to generate fused images; a three-dimensional model of the spinal cord or brain is reconstructed based on the fused images, and the correspondence between electrode contacts and nerve segments is marked on the three-dimensional model; based on the correspondence, the implantation location and configuration parameters of the electrodes are determined; and the target electrode contact is determined from the electrodes based on the configuration parameters.

[0057] In this embodiment, rigid registration of intraoperative CT images with preoperative MRI images can be understood as: using translation and rotation algorithms, the high-definition preoperative MRI images are overlaid on the intraoperative CT images to obtain a fused image that has both accurate coordinates and a clear neural display of the navigation map. This fused image includes complete structures such as bones, nerves, and blood vessels.

[0058] In some embodiments, a 3D model reconstruction is further performed based on the fused image. The reconstruction process may employ voxel reconstruction or surface rendering.

[0059] Optionally, voxel reconstruction directly utilizes the three-dimensional volume data of fused images. By setting different grayscale thresholds, isosurfaces of different tissues such as the spinal cord, brain parenchyma, and cerebrospinal fluid are extracted to construct a three-dimensional anatomical model.

[0060] Optionally, the surface rendering algorithm first extracts the contour lines of the region of interest, and then generates a smooth 3D surface model by fitting triangular facets.

[0061] In some embodiments, on the reconstructed 3D model, image segmentation algorithms are used to automatically identify the positions of electrode contacts, and an anatomical atlas database is used to calibrate the correspondence between electrode contacts and nerve segments. For example, in a spinal cord stimulation scenario, the spatial distance between the electrode contacts and specific neural structures such as the dorsal horn of the spinal cord and the dorsal root entry area can be calculated based on the spatial coordinates of the electrode contacts, and marked on the model with different colors or labels. After calibration, the electrode implantation location and configuration parameters can be further determined based on the correspondence between electrode contacts and nerve segments.

[0062] Optional configuration parameters include: electrode contact combination (such as single-pole, bipolar or tripolar mode), contact activation sequence, etc.

[0063] Then, based on the location of the target nerve segment, the electric field distribution generated by different contact point combinations can be calculated, and a configuration scheme that maximizes coverage of the target area and minimizes the impact on non-target areas can be selected. Then, based on this configuration scheme, the most suitable contact point as the stimulation source is selected from the electrode array as the target electrode contact point.

[0064] It should be understood that the selection of target electrode contacts is not fixed and can be dynamically adjusted based on feedback from intraoperative electromyographic response data. For example, if the initially selected target electrode contacts induce a response in a non-target muscle group during the test, an adjacent contact can be automatically reselected as the new target electrode contact.

[0065] S2. The complex impedance measurement values ​​are sequentially subjected to impedance causality verification and layered relaxation time distribution (DRT) decoupling processing to obtain the first parameter used to characterize the electrochemical properties of the tissue interface.

[0066] In some embodiments, the first parameter includes: solution resistance, charge transfer resistance, and dispersion index of a constant phase element. Optionally, solution resistance is used to characterize the ion concentration and conductivity of the tissue fluid surrounding the electrode. For example, during acute edema, the tissue fluid contains more water and ions, resulting in better conductivity and a decrease in solution resistance.

[0067] Optionally, charge transfer resistance is used to characterize the ease with which charge exchange occurs between the electrode and the tissue. For example, during the fibrous encapsulation stage, the electrode surface is encapsulated by a dense layer of insulating fibrous tissue, making it extremely difficult for charges to "cross" over, thus increasing the charge transfer resistance.

[0068] Optionally, the dispersion index of a constant-phase element is used to characterize the surface roughness and inhomogeneity of the electrode. That is, the dispersion index represents the smoothness of the electrode surface. The dispersion index ranges from 0 to 1. For example, during the fiber encapsulation period, as rough fiber structures accumulate on the electrode surface, the inhomogeneity of the electrode interface intensifies, and the dispersion index typically decreases gradually (deviating more from 1).

[0069] In some embodiments, the impedance causality verification process in step S2 may include: The complex impedance measurements are preprocessed using sliding window smoothing and frequency domain symmetric extrapolation. A lightweight linear Kramers-Kronig fitting model was used to fit the preprocessed complex impedance measurements, and the normalized residual between the measured impedance and the theoretical fitted impedance was calculated. The validity of impedance measurements is determined based on the residual threshold of the implanted electrode: if the residual of a single frequency point exceeds the residual threshold, the complex impedance measurement at that frequency point is determined to be an invalid complex impedance measurement and is discarded. The determined valid complex impedance measurement is used for layered DRT decoupling processing. If the proportion of invalid complex impedance measurements exceeds a predetermined proportion, the complex impedance measurements are deemed invalid as a whole, and the complex impedance measurements are re-acquired.

[0070] In this embodiment, the frequency domain symmetric extrapolation preprocessing may include: converting the measured 10Hz~10kHz impedance to the logarithmic frequency axis, using the midpoint of the logarithm of the measured frequency band as the center of symmetry, and using a piecewise polynomial trend function to synchronously and symmetrically supplement virtual frequency points and corresponding complex impedance values ​​on both sides of the low frequency below 10Hz and the high frequency above 10kHz, and splicing them together to form a continuous impedance spectrum of the entire frequency band.

[0071] It should be noted that in this embodiment, during the verification process, not only is the data of a single frequency point verified (e.g., if the residual of a single frequency point exceeds 2%, that frequency point is discarded), but the data of the whole system is also verified (e.g., if the residual of more than 30% of the frequency points exceeds the standard, all complex impedance data in this group are discarded, that is, the complex impedance measurement of this whole frequency band is contaminated by body position motion artifacts and transient tissue disturbances, and should be discarded). Only the effective complex impedance with acceptable residuals is used for layered DRT decoupling.

[0072] Since patients' turning over and limb movements cause instantaneous impedance jumps in the spinal cord electrodes, impedance causality verification can accurately identify non-physical abnormal impedances, avoid distorted interface parameters obtained by DRT decoupling, and effectively prevent misjudgment of physiological state.

[0073] Directly performing DRT deconvolution on noisy, non-causal distorted impedances will generate spurious peaks in the relaxation time spectrum, distorting solution resistance, charge transfer resistance, and dispersion index, leading to inaccurate subsequent clustering results. Therefore, in this embodiment, the approach of "first removing invalid impedances through impedance causality verification, and then performing layered relaxation time distribution DRT decoupling to extract precise electrochemical features" avoids generating spurious peaks in the relaxation time spectrum and distorting solution resistance, charge transfer resistance, and dispersion index, thus affecting subsequent clustering results. Clearly, without impedance causality verification, incorrect parameter adjustments due to motion artifacts contaminating the data will occur.

[0074] In some embodiments, the hierarchical relaxation time distribution (DRT) decoupling process in step S2 may include: using a relaxation time distribution algorithm to perform deconvolution processing on the complex impedance measurement value after impedance causality verification to obtain a relaxation time distribution spectrum, and extracting the peak parameters corresponding to different characteristic frequency bands in the relaxation time distribution spectrum as the first parameter.

[0075] Optionally, the effective complex impedance measurement value is deconvolved using a relaxation time distribution algorithm to obtain a relaxation time distribution spectrum, and the peak parameters corresponding to different characteristic frequency bands in the relaxation time distribution spectrum are extracted as the first parameter.

[0076] It should be noted that in this embodiment, the specific implementation process of using the hierarchical relaxation time distribution algorithm to deconvolve the complex impedance measurement value to obtain the relaxation time distribution spectrum can be referred to existing related technologies, and will not be elaborated here.

[0077] Optionally, the solution resistance can be determined by taking the value at the high-frequency limit (e.g., 1kHz~10kHz) in the relaxation time distribution spectrum, or by determining the remaining resistance value after subtracting the area of ​​all relaxation peaks.

[0078] Optionally, one or more characteristic peaks will appear in the mid-frequency region (e.g., 100Hz~1kHz) of the relaxation time distribution spectrum. These peaks correspond to the charge transfer process on the electrode surface. The area under the peak obtained by integrating the peak shape of the identified characteristic peaks can be used as the magnitude of the charge transfer resistance.

[0079] Optionally, the full width at half maximum (FWHM) of the mid-frequency characteristic peaks in the relaxation time distribution spectrum can be obtained through Gaussian fitting, or the dispersion index can be directly calculated by inversion using the shape factor of the relaxation time distribution spectrum. It should be understood that the dispersion index is negatively correlated with the peak width of the mid-frequency characteristic peaks. Therefore, the larger the FWHM of the characteristic peaks, the higher the inhomogeneity of the electrode interface, and the smaller the corresponding dispersion index value.

[0080] S3. Based on the temporal variation characteristics of the first parameter and the neural conduction gain value, the physiological state of the electrode-tissue interface is autonomously identified through an unsupervised clustering model.

[0081] In some embodiments, the physiological state of the electrode-tissue interface includes at least one of the following: stable phase, acute edema phase, fibrous encapsulation formation phase, and electrode displacement warning phase.

[0082] In some embodiments, the first parameter is an electrochemical parameter characterizing the microscopic electrical structure (e.g., liquid, membrane, interface roughness) of the electrode-tissue interface. It can reflect physicochemical changes in the tissue surrounding the electrode, but cannot distinguish whether the nerve itself is damaged. The nerve conduction gain is a neurophysiological parameter characterizing the conduction function of spinal cord nerves. It can reflect the actual functional response of the nerve, thus distinguishing whether the impedance change is caused by tissue inflammation or electrode position changes. Therefore, in the scheme of this application embodiment, combining the two can achieve accurate classification of the physiological state of the electrode-tissue interface; without the need for manually preset fixed thresholds, unsupervised clustering autonomously learns multi-feature coupling distribution, automatically classifying four physiological states, effectively avoiding the defects of existing technologies such as "manual stage division and easy misjudgment of single impedance."

[0083] In some embodiments, the unsupervised clustering model fuses two types of feature parameters and distinguishes four physiological states based on feature combinations to achieve accurate classification: 1) The characteristic combination of the steady period: the temporal fluctuations of solution resistance, charge transfer resistance, and diffusion index are minimal, and the neural conduction gain is stable without attenuation. Therefore, there is no inflammation at the interface and the neural coupling is constant. 2) The characteristic combination of acute edema: the solution resistance is significantly reduced (electrochemical), and the nerve conduction gain only fluctuates slightly randomly and does not continue to decline (no nerve damage). Therefore, there is only an increase in tissue fluid and no fibrous capsule, which excludes misdiagnosis of fibrosis. 3) The characteristic combination of the fiber encapsulation formation period: the charge transfer resistance continues to increase, the diffusion index gradually decreases (electrochemical), and the nerve conduction gain slowly decays over a long period of time (the electric field is blocked by the capsule, and the nerve activation efficiency continues to decline). Then: chronic foreign body fibrosis, distinguishing it from short-term edema. 4) Characteristic combination of electrode displacement warning period: the charge transfer resistance of adjacent contacts jumps in the opposite direction (spatial impedance gradient change), and the nerve conduction gain drops sharply in an instant. Then: the distance between the electrode and the nerve changes abruptly, distinguishing it from simple motion noise (noise only has small impedance fluctuations and no sudden drop in gain).

[0084] It should be noted that the acute edema period can generally be understood as 1 day to 3 weeks post-surgery, which is the acute reaction period after surgical trauma, mainly manifested as local tissue inflammation and fluid exudation. During this period, the concentration of the conductive medium between the electrode and the tissue increases, and the conductivity is significantly enhanced. From the perspective of impedance monitoring, as the degree of tissue edema increases, the resistance to ion migration decreases, and the solution resistance will show a significant downward trend.

[0085] It should also be noted that the fibrous encapsulation formation period can generally be understood as 1 to 3 months post-surgery. This is a chronic adaptation process of the body to the implanted foreign body, characterized by the formation of a fibrous capsule on the electrode surface as acute inflammation subsides. This capsule acts as an electrical insulator, increasing the difficulty for charges to cross the electrode-tissue interface, thus leading to a significant increase in charge transfer resistance.

[0086] Furthermore, uneven growth of fibrous tissue leads to increased roughness of the electrode surface microstructure, resulting in a gradual decrease in the dispersion index.

[0087] It should also be noted that the warning period for electrode displacement is generally understood to be from 1 week to 3 months post-surgery, during which it is crucial to guard against abnormal changes in electrode position. Electrode displacement mainly occurs in the early postoperative period, especially within the first week, when the electrode is primarily fixed by physical clamping and has not yet formed biological adhesions with the surrounding tissue. During the subacute period of 1 to 3 months post-surgery, although a fibrous capsule begins to form, drastic changes in body position or external traction can still cause electrode displacement. Electrode displacement leads to abrupt changes in the geometry of the electrode-tissue contact, such as a sudden reduction in contact area or a change in the contact medium from high-resistivity tissue to low-resistivity body fluid. This abrupt change will manifest in the impedance spectrum as drastic jumps or abnormal fluctuations in the complex impedance amplitude and phase angle.

[0088] Therefore, the solution in this application embodiment can accurately determine the current physiological state of the electrode-tissue interface based on the neural conduction gain value and the temporal change characteristics of solution resistance, charge transfer resistance, and diffusion index, and generate corresponding state signals to trigger the displacement warning mechanism, or adjust the stimulation parameters, thereby ensuring the safe operation of the implanted device.

[0089] S4. Based on the physiological state identifiers output by the clustering model, determine control instructions to adjust the current stimulation parameters and sampling period.

[0090] Optionally, the sampling period can be the period used to acquire spinal cord evoked action potential (ECAP) signals.

[0091] In some embodiments, if the unsupervised clustering model identifies the physiological state as being in the acute edema stage or the fibrous encapsulation stage, the control command is used to instruct an upward adjustment of the current stimulation parameters.

[0092] In some embodiments, if the unsupervised clustering model identifies the physiological state as an electrode shift warning period, the control command is used to instruct the current stimulation parameters to be maintained, while simultaneously triggering a first alarm signal.

[0093] Optionally, the stimulation parameters include: stimulation voltage, and / or stimulation pulse width.

[0094] In some embodiments, identifying a stable physiological state using an unsupervised clustering model in step S3 may include: Calculate the coefficients of change of the solution resistance, the charge transfer resistance, and the dispersion index over a continuous preset monitoring period; When the coefficients of change of the solution resistance, the charge transfer resistance, and the dispersion index are all less than their respective preset stability thresholds, the clustering model outputs a stable state identifier.

[0095] In some embodiments, the preset monitoring period may be, for example, 7 days, 10 days, 14 days, etc., but is not limited to this.

[0096] In some embodiments, the coefficient of variation may be the standard deviation or mean of the first parameter over a continuous preset monitoring period, but is not limited thereto.

[0097] Optionally, when the standard deviation of the solution resistance, charge transfer resistance, and dispersion index is less than 5% (preset stability threshold) for 7 consecutive days, the interface can be determined to have entered the stable period, and a stable period status indicator can be output.

[0098] It should be understood that in this embodiment, the preset stability thresholds for solution resistance, charge transfer resistance, and dispersion index can be the same or different, and this application embodiment does not limit this.

[0099] In some embodiments, if the clustering model outputs a stable state identifier, the control command is used to instruct the periodic monitoring of the complex impedance measurement value, and the acquisition frequency of the complex impedance measurement value is reduced from a first frequency to a second frequency, and the sampling period of the ECAP signal is extended from a first duration to a second duration.

[0100] In some embodiments, control commands can be used to instruct the acquisition frequency of complex impedance measurements to be reduced from once a day (first frequency) to once a week (second frequency), thereby effectively reducing system power consumption.

[0101] In some embodiments, identifying the physiological state as acute edema phase using an unsupervised clustering model in step S3 may include: The first rate of change of the solution resistance relative to the initial reference value is obtained within a first preset time period; When the first rate of change exceeds the first threshold and the fluctuation amplitude of the neural conduction gain value is less than the preset interference threshold, the clustering model outputs an acute edema phase status identifier.

[0102] In some embodiments, the first preset time period may be a continuous period of 3 days, 5 days, etc., but is not limited to this.

[0103] In some embodiments, the solution resistance can be calculated daily. Compared to the baseline value on day 1 after implantation rate of change If for 3 consecutive days If the first threshold is reached, it can be determined that the interface has entered the acute edema stage, and the acute edema stage status indicator will be output.

[0104] In some embodiments, if the clustering model outputs an acute edema phase status identifier, the control command is used to instruct the current stimulation voltage to be adjusted to the target stimulation voltage.

[0105] Optionally, the target stimulation voltage is determined based on the voltage adjustment amount and the boundary value of the safe voltage range, wherein the voltage adjustment amount is obtained based on the first rate of change and a preset compensation coefficient.

[0106] Optionally, the compensation coefficient can range from 0.3 to 0.7.

[0107] In some embodiments, voltage adjustment amount Among them, the compensation coefficient The value ranges from 0.3 to 0.7, for example, it can be 0.5, but it is not limited to this.

[0108] Optionally, the compensation coefficient can be set to 0.5. When edema is in a period of rapid change, a lower compensation coefficient (e.g., 0.3) can be selected to ensure absolute safety; when edema tends to stabilize but impedance still has a slight drift, a higher compensation coefficient (e.g., 0.7) can be selected to improve the comfort of the stimulation experience. The setting of this value range is intended to balance the timeliness of impedance adaptive compensation with the safety of clinical stimulation.

[0109] It should be noted that during the acute edema phase, the increase in tissue fluid leads to an increase in solution resistance ( The number of cases decreases. According to Ohm's law (…). = × If the stimulator is in constant current mode, a decrease in impedance will cause a decrease in the actual voltage applied to the tissue, and the patient may feel that the stimulation is "weaker." In this case, the system needs to increase the voltage.

[0110] However, edema is a dynamic and nonlinear process. If the system is compensated with a 1:1 proportional voltage (i.e., a compensation factor of 1.0) exactly according to the rate of change of impedance, the voltage adjustment will be too large if the edema fluctuates drastically or if there is transient noise in the measurement. Excessive voltage may not only exceed the safety threshold but may also cause unexpected muscle twitching or pain.

[0111] Therefore, in impedance dynamic monitoring during the acute edema phase, the compensation coefficient is used... k The voltage is adjusted by proportionally controlling the rate of change of impedance, which ensures that the voltage can be adaptively adjusted based on physiological changes, while also ensuring the safety of stimulation and the comfort of the patient.

[0112] In some embodiments, the safety voltage boundary values ​​can be 0.5V and 10V, but are not limited thereto.

[0113] In some embodiments, if it is determined that the interface is in the acute edema phase, the control command is used to instruct the current stimulation voltage to be adjusted to the target stimulation voltage while keeping the current stimulation pulse unchanged.

[0114] In some embodiments, the target stimulation voltage can be determined based on the voltage adjustment amount and the boundary values ​​of the safe voltage range.

[0115] In some embodiments, the above adjustment method further includes: Based on the current stimulation voltage and the voltage adjustment amount, the initial stimulation voltage is obtained; If the initial stimulation voltage is within the safe voltage range, then the initial stimulation voltage is determined as the target stimulation voltage; If the initial stimulation voltage is less than the lower limit of the safe voltage range, then the lower limit is determined as the target stimulation voltage; If the initial stimulation voltage is greater than the upper limit of the safe voltage range, then the upper limit is determined as the target stimulation voltage.

[0116] In some embodiments, the preliminary target voltage is calculated. .like If the voltage does not exceed the safe voltage range [0.5V, 10V], then the target stimulation voltage is: .

[0117] like If the voltage exceeds the safe voltage range [0.5V, 10V], it will be limited. For example: if If the voltage is less than 0.5V, the target stimulation voltage is the lower limit of 0.5V; if If the voltage is greater than 10V, the target stimulation voltage is the upper limit of 10V.

[0118] In some embodiments, identifying the physiological state as the fibrous encapsulation formation stage using an unsupervised clustering model in step S3 may include: Monitor the rate of increase of the charge transfer resistance, the decreasing trend of the diffusion index, and the changing trend of the neural conduction gain value; When the neural conduction gain value shows a continuous and slow downward trend, the rate of increase of the charge transfer resistance exceeds the second threshold, and the diffusion index is lower than the preset diffusion critical value, the clustering model outputs a fiber encapsulation formation period status identifier.

[0119] In some embodiments, the second threshold may be 5% / day, but is not limited thereto.

[0120] In some embodiments, the diffusion threshold can be 0.8, but is not limited thereto.

[0121] Optionally, the rising rate of charge transfer resistance and the decreasing trend of diffusion index, as well as the changing trend of neural conduction gain, can be monitored. If the rising rate of charge transfer resistance is greater than 5% / day and the diffusion index is less than 0.8, and the changing trend of neural conduction gain is a continuous slow decreasing trend, then the interface can be determined to be in the fiber encapsulation formation stage, and a fiber encapsulation formation stage status indicator can be output.

[0122] In some embodiments, the above adjustment method may further include: The clustering model divides the fiber encapsulation formation period into a light encapsulation period, a moderate encapsulation period, and a heavy encapsulation period based on the relationship between the absolute value of the charge transfer resistance and the first resistance threshold and the second resistance threshold, and outputs corresponding classification labels.

[0123] In some embodiments, the first resistance threshold may be 5 kΩ and the second resistance threshold may be 20 kΩ, but is not limited thereto.

[0124] In some embodiments, if the absolute value of the charge transfer resistance is less than 5kΩ and greater than 1kΩ, it can be determined that the interface is in a light encapsulation period, and then the corresponding grade label can be output.

[0125] Optionally, if the grading indicator corresponds to a mild envelopment period, the control command is used to instruct that the current stimulation voltage be kept constant and the current stimulation pulse width be increased within a preset compensation range. For example, if the preset compensation range is 20%, the increase in the stimulation pulse width shall not exceed 20%, for example, it can be increased by 10%, but is not limited to this.

[0126] In some embodiments, if the absolute value of the charge transfer resistance is greater than 5kΩ and less than 20kΩ, the interface can be determined to be in a moderate encapsulation period, and the corresponding grade label can be output.

[0127] Optionally, if the grading indicator corresponds to the moderate encapsulation period, the control command is used to instruct the current stimulation voltage and the current stimulation pulse width to be increased synchronously according to a preset compensation gradient. For example, if the preset compensation gradient for the stimulation voltage is 0.5V / step and the preset compensation gradient for the stimulation pulse width is 50μs / step, then the control command is used to instruct the stimulation voltage to be increased by 0.5V / step, and simultaneously, the stimulation pulse width to be increased by 50μs / step.

[0128] In some embodiments, if the absolute value of the charge transfer resistance is greater than 20kΩ, it can be determined that the interface is in a heavily encapsulated period, and then the corresponding grade indicator can be output.

[0129] Optionally, if the grading indicator corresponds to the severe encapsulation period, the control command is used to instruct the current stimulation parameters to be adjusted to a preset safe upper limit and to trigger a second alarm signal. For example, if the preset safe upper limit for the stimulation voltage is 10V and the preset safe upper limit for the stimulation pulse width is 500μs, then the control command is used to instruct the current stimulation voltage to be adjusted to 10V and the current stimulation pulse width to be adjusted to 500μs, while simultaneously triggering a second alarm signal, such as displaying a risk warning message about treatment failure. It should be understood that the intensity of the second alarm signal is greater than the intensity of the first alarm signal for electrode displacement warning. That is, the severity of the severe encapsulation period is stronger than that of the electrode displacement warning period.

[0130] In some embodiments, identifying the physiological state as an electrode displacement warning period using an unsupervised clustering model in step S3 may include: The rate of change of charge transfer resistance of adjacent first and second electrode contacts among the target electrode contacts is obtained within the same time period; When the second rate of change of the charge transfer resistance of the first electrode contact is positive, and the third rate of change of the charge transfer resistance of the second electrode contact is negative, and the absolute value of the difference between the second rate of change and the third rate of change exceeds a preset displacement judgment threshold, the clustering model outputs an electrode displacement early warning period status indicator.

[0131] In some embodiments, the rate of change of charge transfer resistance of adjacent electrode contacts A and B over the same time period (e.g., 1 day) can be calculated. and .like Greater than 0, and Less than 0, and If the displacement determination threshold is reached, the interface is determined to be in the electrode displacement warning period, and then the electrode displacement warning period status indicator is output.

[0132] In some embodiments, identifying the physiological state as an electrode displacement warning period using an unsupervised clustering model in step S3 may include: The rate of change of charge transfer resistance of adjacent first and second electrode contacts among the target electrode contacts is obtained within the same time period; When the second rate of change of the charge transfer resistance of the first electrode contact is negative, and the third rate of change of the charge transfer resistance of the second electrode contact is positive, and the absolute value of the difference between the second rate of change and the third rate of change exceeds a preset displacement judgment threshold, the clustering model outputs an electrode displacement warning period status indicator.

[0133] In some embodiments, the rate of change of charge transfer resistance of adjacent electrode contacts A and B over the same time period (e.g., 1 day) can be calculated. and .like Less than 0, and Greater than 0, and If the displacement determination threshold is reached, the interface is determined to be in the electrode displacement warning period, and then the electrode displacement warning period status indicator is output.

[0134] Optionally, the first electrode contact and the second electrode contact are two adjacent electrode contacts among the target electrode contacts.

[0135] In some embodiments, if the clustering model outputs an electrode displacement warning period status identifier, the control command is further used to instruct switching to the backup electrode contact for stimulation, and to monitor the ECAP signal and the complex impedance measurement value of the backup electrode contact.

[0136] In some embodiments, if it is determined that the interface is in the electrode displacement warning period, the control command can be used to instruct to keep the current stimulation parameters unchanged to avoid abnormal stimulation spread; at the same time, it can also instruct to automatically switch to a preset backup electrode contact (e.g., an electrode contact of a nerve segment adjacent to the nerve segment where the target electrode contact is located) for stimulation; in addition, it can also trigger a first alarm signal to indicate that the electrode has been displaced.

[0137] In summary, the embodiments of this application provide an adaptive stimulation parameter adjustment method, which can achieve the following technical effects: By performing feature decoupling on the acquired complex impedance measurements at multiple frequencies, key parameters characterizing the interface properties can be obtained. Based on the time-series variation characteristics of these parameters, misjudgments from single measurements are avoided, thereby enabling accurate judgment of the interface's physiological state.

[0138] Furthermore, it offers differentiated parameter adjustment strategies for different physiological states and severity levels (such as mild, moderate, and severe fiber encapsulation). Additionally, it automatically reduces monitoring frequency during stable periods to achieve energy savings, effectively balancing system performance and battery life.

[0139] Furthermore, the solution in this application embodiment can also achieve the following technical effects: 1. Significantly improved accuracy in physiological state recognition: Through impedance causality verification and hierarchical DRT decoupling, three types of interface electrochemical parameters can be accurately extracted. At the same time, by combining neural conduction gain values ​​for clustering, edema, fibrosis and displacement, three easily confused physiological states can be effectively distinguished, thereby reducing the misjudgment rate in scenarios with multiple overlapping lesions.

[0140] 2. Full-process closed-loop autonomous control without manual intervention: Abandoning manual thresholds and manual phased rules, unsupervised clustering autonomously identifies physiological states based on dual-modal fusion features, automatically and synchronously adjusts stimulation parameters and sampling cycles, and achieves long-term implantation with full-process adaptive closed-loop control, solving the pain points of existing solutions being lagging behind and relying on doctors to manually adjust parameters.

[0141] Based on the same principles as the methods shown in the embodiments of this application, the embodiments of this application also provide a signal processing system, which includes the implantable spinal cord stimulation system provided in the above embodiments.

[0142] In an alternative embodiment, a signal processing system, such as Figure 3 As shown, Figure 3 The signal processing system 20 shown includes a processor 21 and a memory 23. The processor 21 is communicatively connected to the memory 23, for example, via a bus 22. Optionally, the signal processing system 20 may further include a transceiver 24, which can be used for signal interaction between the signal processing system and other electronic devices, such as signal transmission and / or signal reception. It should be noted that in practical applications, the transceiver 24 is not limited to one type, and the structure of this signal processing system 20 does not constitute a limitation on the embodiments of this application.

[0143] Processor 21 may be a CPU (Central Processing Unit), general-purpose processor, DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array), or other programmable logic device, transistor logic device, hardware component, or any combination thereof, including the chip or implantable spinal cord stimulation system described in any of the above embodiments. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 21 may also be a combination that implements computing functions, such as including one or more microprocessor combinations, a combination of DSP and microprocessor, etc.

[0144] Bus 22 may include a pathway for transmitting information between the aforementioned components. Bus 22 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 22 may be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 3 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0145] The memory 23 may be a ROM (Read-Only Memory) or other type of static storage device capable of storing static information and instructions, RAM (Random Access Memory) or other type of dynamic storage device capable of storing information and instructions, or it may be an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read-Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0146] The memory 23 is used to store computer programs that execute the embodiments of this application, and the execution is controlled by the processor 21. The processor 21 is used to execute the computer programs stored in the memory 23 to implement the steps shown in the foregoing method embodiments.

[0147] This application provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it can implement the steps and corresponding content of the aforementioned method embodiments.

[0148] This application also provides a computer program product, including a computer program that, when executed by a processor, can implement the steps and corresponding content of the aforementioned method embodiments.

[0149] In the description of this application, the terms "center," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicate directions or positional relationships based on the exemplary directions or positional relationships shown in the accompanying drawings. They are used to facilitate the description or simplification of the embodiments of this application and are not intended to indicate or imply that the device or component referred to must have a specific orientation or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0150] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0151] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "joining" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0152] In the description of this specification, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0153] The above description is only a partial embodiment of this application. It should be noted that for those skilled in the art, other similar implementation methods based on the technical concept of this application, without departing from the technical concept of this application, also fall within the protection scope of the embodiments of this application.

Claims

1. A method for adjusting stimulation parameters of an implantable spinal cord electrical stimulation system, characterized in that, include: Acquire neural conduction gain values ​​and complex impedance measurements of the target electrode contact at multiple frequencies; The complex impedance measurements are sequentially subjected to impedance causality verification and layered relaxation time distribution (DRT) decoupling to obtain a first parameter for characterizing the electrochemical properties of the tissue interface. The first parameter includes: solution resistance, charge transfer resistance, and dispersion index of constant phase element. Based on the temporal variation characteristics of the first parameter and the neural conduction gain value, the physiological state of the electrode-tissue interface is autonomously identified through an unsupervised clustering model. The physiological state of the electrode-tissue interface includes at least one of the following: stable period, acute edema period, fibrous encapsulation formation period, and electrode displacement warning period. Control commands are determined based on the physiological state identifiers output by the unsupervised clustering model. These control commands are used to adjust the current stimulation parameters and sampling period. The stimulation parameters include: stimulation voltage, and / or stimulation pulse width; The sampling period is the period used to acquire spinal cord evoked action potential (ECAP) signals, and the neural conduction gain value is obtained based on the ECAP signals; If a stable state identifier is output through an unsupervised clustering model, the control command is used to instruct a reduction in the monitoring frequency; If the unsupervised clustering model outputs an electrode shift warning period status indicator, the control command is used to instruct switching to the backup electrode contact for stimulation; If the unsupervised clustering model outputs a state identifier for the acute edema phase or a state identifier for the fibrous encapsulation phase, the control command is used to instruct an upward adjustment of the current stimulation parameters.

2. The adjustment method according to claim 1, characterized in that, Identifying physiological states as stable phases using unsupervised clustering models includes: Calculate the variation coefficients of the solution resistance, the charge transfer resistance, and the dispersion index over a continuous preset monitoring period, wherein the variation coefficient is the standard deviation or mean of the first parameter over the continuous preset monitoring period; When the coefficients of change of the solution resistance, the charge transfer resistance, and the dispersion index are all less than their respective preset stability thresholds, the unsupervised clustering model outputs a stable state identifier.

3. The adjustment method according to claim 2, characterized in that, If the unsupervised clustering model outputs a stable state identifier, the control command is used to instruct the periodic monitoring of the complex impedance measurement value, and the acquisition frequency of the complex impedance measurement value is reduced from a first frequency to a second frequency, and the sampling period of the ECAP signal is extended from a first duration to a second duration.

4. The adjustment method according to claim 1, characterized in that, The physiological state of acute edema was identified using an unsupervised clustering model, including: The rate of change of the solution resistance relative to the initial reference value is obtained during a first preset time period; When the rate of change exceeds a first threshold and the fluctuation amplitude of the neural conduction gain value is less than a preset interference threshold, the unsupervised clustering model outputs an acute edema phase status identifier.

5. The adjustment method according to claim 4, characterized in that, If the unsupervised clustering model outputs an acute edema phase status indicator, the control command is used to instruct the current stimulation voltage to be adjusted to the target stimulation voltage; The target stimulation voltage is determined based on the voltage adjustment amount and the boundary value of the safe voltage range, and the voltage adjustment amount is obtained based on the rate of change and a preset compensation coefficient. The compensation coefficient ranges from 0.3 to 0.

7.

6. The adjustment method according to claim 5, characterized in that, The adjustment method further includes: Based on the current stimulation voltage and the voltage adjustment amount, the initial stimulation voltage is obtained; If the initial stimulation voltage is within the safe voltage range, then the initial stimulation voltage is determined as the target stimulation voltage; If the initial stimulation voltage is less than the lower limit of the safe voltage range, then the lower limit is determined as the target stimulation voltage; If the initial stimulation voltage is greater than the upper limit of the safe voltage range, then the upper limit is determined as the target stimulation voltage.

7. The adjustment method according to claim 1, characterized in that, The physiological state of fibrous encapsulation formation was identified using an unsupervised clustering model, including: Monitor the rate of increase of the charge transfer resistance, the decreasing trend of the diffusion index, and the changing trend of the neural conduction gain value; When the neural conduction gain value shows a continuous and slow decreasing trend, the rate of increase of the charge transfer resistance exceeds the second threshold, and the diffusion index is lower than the preset diffusion critical value, the unsupervised clustering model outputs a fiber encapsulation formation period status identifier.

8. The adjustment method according to claim 1 or 7, characterized in that, The adjustment method further includes: The unsupervised clustering model divides the fiber encapsulation formation period into a light encapsulation period, a moderate encapsulation period, and a heavy encapsulation period based on the relationship between the absolute value of the charge transfer resistance and the first resistance threshold and the second resistance threshold, and outputs the corresponding classification labels. Wherein, if the grading identifier corresponds to the mild envelopment period, the control instruction is used to indicate that the current stimulation voltage should remain unchanged and the current stimulation pulse width should be increased within a preset compensation range; If the grading indicator corresponds to the moderate encapsulation period, the control command is used to instruct the current stimulation voltage and the current stimulation pulse width to be synchronously increased according to the preset compensation gradient; If the grading indicator corresponds to the severe envelopment period, the control command is used to instruct the current stimulation parameters to be adjusted to a preset safe upper limit value and to trigger a second alarm signal.

9. The adjustment method according to claim 1, characterized in that, Physiological states identified as electrode displacement warning periods using unsupervised clustering models include: The rate of change of charge transfer resistance of adjacent first electrode contact and second electrode contact among the target electrode contacts is obtained within the same time period, wherein the first electrode contact and the second electrode contact are two adjacent electrode contacts among the target electrode contacts; When the rate of change of charge transfer resistance of the first electrode contact is positive and the rate of change of charge transfer resistance of the second electrode contact is negative, and the difference between the absolute values ​​of the two rates of change exceeds a preset displacement judgment threshold, the unsupervised clustering model outputs an electrode displacement early warning period status identifier. or, When the rate of change of charge transfer resistance of the first electrode contact is negative and the rate of change of charge transfer resistance of the second electrode contact is positive, and the difference between the absolute values ​​of the two rates of change exceeds a preset displacement judgment threshold, the unsupervised clustering model outputs an electrode displacement early warning period status indicator.

10. The adjustment method according to claim 1 or 9, characterized in that, If the unsupervised clustering model outputs an electrode shift warning period status indicator, the control command is used to instruct switching to the backup electrode contact and maintaining the current stimulation parameters for stimulation. At the same time, a first alarm signal is triggered and the ECAP signal and the complex impedance measurement value of the backup electrode contact are monitored.

11. The adjustment method according to any one of claims 1-4, 7, and 9, characterized in that, The hierarchical relaxation time distribution (DRT) decoupling process includes: The relaxation time distribution algorithm is used to deconvolve the complex impedance measurement value after impedance causality verification to obtain the relaxation time distribution spectrum, and the peak parameters corresponding to different characteristic frequency bands in the relaxation time distribution spectrum are extracted as the first parameter.

12. The adjustment method according to any one of claims 1-4, 7, and 9, characterized in that, The impedance causality verification process includes: The complex impedance measurements are preprocessed using sliding window smoothing and frequency domain symmetric extrapolation. A lightweight linear Kramers-Kronig fitting model was used to fit the preprocessed complex impedance measurements, and the normalized residual between the measured impedance and the theoretical fitted impedance was calculated. The validity of impedance measurements is determined based on the residual threshold of the implanted electrode: if the residual of a single frequency point exceeds the residual threshold, the complex impedance measurement at that frequency point is determined to be an invalid complex impedance measurement and is discarded. The determined valid complex impedance measurement is used for layered DRT decoupling processing. If the proportion of invalid complex impedance measurements exceeds a predetermined proportion, the complex impedance measurements are deemed invalid as a whole, and the complex impedance measurements are re-acquired.

13. An implantable spinal cord electrical stimulation system, characterized in that, include: Multiple electrodes implanted in the spinal cord; An electrical stimulation driving module is used to apply an electrical stimulation signal to the target electrode contact among the plurality of electrodes; Impedance measurement module is used to acquire the complex impedance measurement values ​​of the target electrode contact at multiple frequencies; The signal acquisition module is used to periodically acquire spinal cord evoked action potential (ECAP) signals. The data processing module is used to obtain the neural conduction gain value based on the ECAP signal, and to perform impedance causality verification processing and layered relaxation time distribution (DRT) decoupling processing on the complex impedance measurement value in sequence to obtain a first parameter for characterizing the electrochemical properties of the tissue interface. The first parameter includes: solution resistance, charge transfer resistance, and diffusion index of constant phase element. The signal generation module is used to autonomously identify the physiological state of the electrode-tissue interface through an unsupervised clustering model based on the temporal variation characteristics of the first parameter and the neural conduction gain value. The physiological state of the electrode-tissue interface includes at least one of the following: stable period, acute edema period, fibrous encapsulation formation period, and electrode displacement warning period. The instruction determination module is used to determine control instructions based on the physiological state identifiers output by the unsupervised clustering model; A parameter adjustment module is used to control the electrical stimulation drive module based on the control command, wherein the control command is used to adjust the current stimulation parameters and the sampling period of the ECAP signal; The stimulation parameters include: stimulation voltage, and / or stimulation pulse width; If a stable state identifier is output through an unsupervised clustering model, the control command is used to instruct a reduction in the monitoring frequency; If the unsupervised clustering model outputs an electrode shift warning period status indicator, the control command is used to instruct switching to the backup electrode contact for stimulation; If the unsupervised clustering model outputs a state identifier for the acute edema phase or a state identifier for the fibrous encapsulation phase, the control command is used to instruct an upward adjustment of the current stimulation parameters.

14. The implantable spinal cord stimulation system according to claim 13, characterized in that, The signal generation module identifies a stable physiological state using an unsupervised clustering model, specifically for: Calculate the variation coefficients of the solution resistance, the charge transfer resistance, and the dispersion index over a continuous preset monitoring period, wherein the variation coefficient is the standard deviation or mean of the first parameter over the continuous preset monitoring period; When the coefficients of change of the solution resistance, the charge transfer resistance, and the dispersion index are all less than their respective preset stability thresholds, the unsupervised clustering model outputs a stable state identifier.

15. The implantable spinal cord stimulation system according to claim 14, characterized in that, If the unsupervised clustering model outputs a stable state identifier, the control command is used to instruct the periodic monitoring of the complex impedance measurement value, and the acquisition frequency of the complex impedance measurement value is reduced from a first frequency to a second frequency, and the sampling period of the ECAP signal is extended from a first duration to a second duration.

16. The implantable spinal cord stimulation system according to claim 13, characterized in that, When the signal generation module identifies the physiological state as acute edema using an unsupervised clustering model, it is specifically used to: obtain the rate of change of the solution resistance relative to the initial baseline value within a first preset time period; when the rate of change exceeds a first threshold and the fluctuation amplitude of the neural conduction gain value is less than a preset interference threshold, the unsupervised clustering model outputs an acute edema state identifier.

17. The implantable spinal cord stimulation system according to claim 16, characterized in that, If the unsupervised clustering model outputs an acute edema phase status indicator, the control command is used to instruct the current stimulation voltage to be adjusted to the target stimulation voltage; The target stimulation voltage is determined based on the voltage adjustment amount and the boundary value of the safe voltage range, and the voltage adjustment amount is obtained based on the rate of change and a preset compensation coefficient. The compensation coefficient ranges from 0.3 to 0.

7.

18. The implantable spinal cord stimulation system according to claim 17, characterized in that, The data processing module is also used for: Based on the current stimulation voltage and the voltage adjustment amount, the initial stimulation voltage is obtained; If the initial stimulation voltage is within the safe voltage range, then the initial stimulation voltage is determined as the target stimulation voltage; If the initial stimulation voltage is less than the lower limit of the safe voltage range, then the lower limit is determined as the target stimulation voltage; If the initial stimulation voltage is greater than the upper limit of the safe voltage range, then the upper limit is determined as the target stimulation voltage.

19. The implantable spinal cord stimulation system according to claim 13, characterized in that, The signal generation module identifies the physiological state as the fiber encapsulation formation stage through an unsupervised clustering model. Specifically, it monitors the rising rate of the charge transfer resistance, the decreasing trend of the diffusion index, and the changing trend of the neural conduction gain value. When the changing trend of the neural conduction gain value is a continuous and slow decreasing trend, the rising rate of the charge transfer resistance exceeds a second threshold, and the diffusion index is lower than a preset diffusion critical value, the unsupervised clustering model outputs a fiber encapsulation formation stage status identifier.

20. The implantable spinal cord stimulation system according to claim 13 or 19, characterized in that, The signal generation module is also used to: divide the fiber wrapping formation period into a light wrapping period, a medium wrapping period, and a heavy wrapping period according to the relationship between the absolute value of the charge transfer resistance and the first resistance threshold and the second resistance threshold, and output the corresponding grading identifier; Wherein, if the grading identifier corresponds to the mild envelopment period, the control instruction is used to indicate that the current stimulation voltage should remain unchanged and the current stimulation pulse width should be increased within a preset compensation range; If the grading indicator corresponds to the moderate encapsulation period, the control command is used to instruct the current stimulation voltage and the current stimulation pulse width to be synchronously increased according to the preset compensation gradient; If the grading indicator corresponds to the severe envelopment period, the control command is used to instruct the current stimulation parameters to be adjusted to a preset safe upper limit value and to trigger a second alarm signal.

21. The implantable spinal cord stimulation system according to claim 13, characterized in that, The signal generation module identifies the physiological state as the electrode displacement warning period through an unsupervised clustering model. Specifically, it is used to: obtain the rate of change of charge transfer resistance of adjacent first electrode contact and second electrode contact in the target electrode contact within the same time period, wherein the first electrode contact and the second electrode contact are two adjacent electrode contacts in the target electrode contact. When the rate of change of charge transfer resistance of the first electrode contact is positive and the rate of change of charge transfer resistance of the second electrode contact is negative, and the difference between the absolute values ​​of the two rates of change exceeds a preset displacement judgment threshold, the unsupervised clustering model outputs an electrode displacement early warning period status identifier. or, When the rate of change of charge transfer resistance of the first electrode contact is negative and the rate of change of charge transfer resistance of the second electrode contact is positive, and the difference between the absolute values ​​of the two rates of change exceeds a preset displacement judgment threshold, the unsupervised clustering model outputs an electrode displacement early warning period status indicator.

22. The implantable spinal cord stimulation system according to claim 13 or 21, characterized in that, If the unsupervised clustering model outputs an electrode shift warning period status indicator, the control command is also used to instruct switching to the backup electrode contact and maintaining the current stimulation parameters for stimulation. At the same time, it triggers a first alarm signal and monitors the ECAP signal and the complex impedance measurement value of the backup electrode contact.

23. The implantable spinal cord stimulation system according to any one of claims 13-16, 19, and 21, characterized in that, When performing the hierarchical relaxation time distribution (DRT) decoupling process, the data processing module is specifically used for: The relaxation time distribution algorithm is used to deconvolve the complex impedance measurement value after impedance causality verification to obtain the relaxation time distribution spectrum, and the peak parameters corresponding to different characteristic frequency bands in the relaxation time distribution spectrum are extracted as the first parameter.

24. The implantable spinal cord stimulation system according to any one of claims 13-16, 19, and 21, characterized in that, When performing impedance causality verification, the data processing module is specifically used for: The complex impedance measurements are preprocessed using sliding window smoothing and frequency domain symmetric extrapolation. A lightweight linear Kramers-Kronig fitting model was used to fit the preprocessed complex impedance measurements, and the normalized residual between the measured impedance and the theoretical fitted impedance was calculated. The validity of impedance measurements is determined based on the residual threshold of the implanted electrode: if the residual of a single frequency point exceeds the residual threshold, the complex impedance measurement at that frequency point is determined to be an invalid complex impedance measurement and is discarded. The determined valid complex impedance measurement is used for layered DRT decoupling processing. If the proportion of invalid complex impedance measurements exceeds a predetermined proportion, the complex impedance measurements are deemed invalid as a whole, and the complex impedance measurements are re-acquired.

25. A signal processing system, characterized in that, Including the implantable spinal cord stimulation system as described in any one of claims 13 to 24.

26. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the adjustment method according to any one of claims 1 to 12.

27. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the adjustment method according to any one of claims 1 to 12.

Images