Intelligent self-adaptive acupuncture electrical stimulation system and method

Abstract

The invention discloses an intelligent self-adaptive acupuncture electrical stimulation system and method, and the system comprises a signal collection module which is used for obtaining a multi-modal physiological signal of a tested animal in real time; the signal processing module is connected with the signal acquisition module and is used for performing filtering and noise reduction on the acquired physiological signals and extracting time domain and frequency domain characteristics; the physiological state recognition module is connected with the signal processing module and used for inputting the features into a pre-trained state recognition model; the parameter matching and regulating module is connected with the physiological state recognition module and is used for automatically determining an electrical stimulation parameter combination according to a recognition result; the electrical stimulation execution module comprises a multi-channel programmable electrical stimulator and an electrode needle; and the feedback control module forms a closed-loop control path with the signal acquisition module and the parameter regulation and control module and is used for dynamically updating the stimulation parameters. According to the technical scheme, the physiological changes of individuals can be responded in real time, and the stimulation strategy can be adjusted according to needs.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of acupuncture systems, and in particular to an intelligent adaptive acupuncture electric stimulation system and method. BACKGROUND

[0002] As an important treatment method of traditional Chinese medicine, acupuncture has good clinical effects in regulating the nervous system, improving metabolic function, and relieving pain. With the development of modern medical technology, traditional acupuncture gradually introduces electric stimulation methods, forming so-called "electroacupuncture therapy", that is, applying a pulse current with a certain frequency and voltage on the basis of acupuncture to enhance the efficacy and expand the indications. However, the existing electroacupuncture equipment still has great limitations in parameter control, targeting, electrode structure, and individualized treatment, etc.

[0003] Firstly, in the stimulation control mode, the traditional electroacupuncture instrument often adopts a single-channel output mode with fixed parameters, and the stimulation frequency, voltage and waveform type are mostly manually set, lacking real-time sensing and feedback mechanism for the physiological state of patients or animals, and it is difficult to cope with the dynamic changes of physiological reactions during treatment, resulting in unpredictable stimulation effect or mismatched stimulation intensity.

[0004] Secondly, in terms of targeting accuracy, the commonly used body surface patch electrode or conventional needle electrode cannot achieve precise action on deep nerves or specific tissue structures, the current diffusion range is large, and the stimulation depth and range are difficult to control, which may cause "shallow over-stimulation and deep under-stimulation" problems. In addition, the electroacupuncture device lacks dynamic monitoring and control of the impedance change of the electrode and tissue interface, further affecting the consistency and safety of stimulation.

[0005] Thirdly, in terms of acupoint positioning and treatment standardization, most clinical acupuncture operations still rely on the experience of doctors to select and position acupoints, lacking objective and visualized auxiliary means, resulting in poor repeatability of the treatment process and difficulty in meeting the requirements of operation standardization and accuracy for research or large-scale treatment. SUMMARY

[0006] The purpose of the embodiments of the present application is to provide an intelligent adaptive acupuncture electric stimulation system and method to solve the technical problem of low treatment accuracy and safety of the existing acupuncture electric stimulation system.

[0007] In order to solve the above technical problems, the embodiments of the present application provide an intelligent adaptive acupuncture electric stimulation system, which adopts the following technical solutions:

[0008] An intelligent adaptive acupuncture electric stimulation system, comprising:

[0009] A signal acquisition module comprising an electromyography sensor, an impedance sensor, a temperature sensor and a photoplethysmography sensor, for acquiring real-time multi-modal physiological signals of a test animal;

[0010] The signal processing module is connected with the signal acquisition module and includes a signal filtering subunit and a feature extraction subunit, and is used for filtering and denoising the collected physiological signals and extracting time domain and frequency domain features;

[0011] The physiological state recognition module is connected with the signal processing module and is used for inputting the features into a pre-trained state recognition model including a support vector machine classifier and an LSTM network to identify whether the current state is a stress state, a hyperglycemia state or a sympathetic nerve activation state;

[0012] The parameter matching and regulation module is connected with the physiological state recognition module and is used for automatically determining an electric stimulation parameter combination including a frequency, a voltage, a waveform type and a pulse width according to the recognition result;

[0013] The electric stimulation execution module includes a multi-channel programmable electric stimulator and an electrode needle, the multi-channel programmable electric stimulator supports multi-waveform, sweep frequency and duty cycle adjustment, and the electrode needle is provided with multiple spaced conductive rings along the length of the needle body and is connected to different stimulation channels through insulation.

[0014] The feedback control module forms a closed-loop control path with the signal acquisition module and the parameter regulation module and is used for dynamically updating the stimulation parameters.

[0015] In a possible implementation, the electrode needle has a segmented structure, the conductive segments have a width of 0.5 mm, and the intervals between adjacent conductive segments are 3 mm, and the intervals are insulated by using polyimide material.

[0016] In a possible implementation, the feedback control module includes a state priority scheduling mechanism, and when multiple abnormal states occur concurrently, a safety strategy is preferentially executed.

[0017] In a possible implementation, the feedback control module includes a four-electrode tissue impedance measurement unit, supports electric impedance imaging reconstruction of current distribution images, and assists in deep stimulation control.

[0018] In a possible implementation, the electric stimulator is provided with a 12-bit resolution FPGA waveform synthesizer, and supports switching of cluster waves, dense and sparse waves and logarithmic sweep frequency waves.

[0019] In a possible implementation, the feedback control module sets three-level safety protection logic, including an electromyography amplitude threshold current limiting mechanism, an impedance anomaly automatic power-off mechanism and a temperature anomaly warning mechanism.

[0020] In a possible implementation, the intelligent adaptive acupuncture electric stimulation system further includes:

[0021] The acupoint positioning module comprises an infrared thermal imaging unit and an ultrasonic elastography unit.

[0022] The data recognition module adopts a YOLOv5 deep learning model to perform image recognition and positioning on the target acupoint according to the established acupoint database.

[0023] The projection navigation module projects the acupoint position information on the animal body surface based on three-dimensional pose registration technology.

[0024] To solve the above technical problems, the embodiment of the present application also provides a kind of intelligent adaptive acupuncture electric stimulation method, and the following technical solutions are adopted:

[0025] A kind of intelligent adaptive acupuncture electric stimulation method is applied to the intelligent adaptive acupuncture electric stimulation system described in any possible embodiment, comprising the following steps:

[0026] Implanting electrode needle with multi-section conductive structure in target acupoint area;

[0027] Collecting multi-modal physiological signals including EMG signals, skin impedance, skin temperature and PPG signals at the depth of electrode needle;

[0028] Filtering, standardizing, wavelet denoising and feature extraction are performed on the multi-modal physiological signals to obtain time-frequency domain feature parameters reflecting nerve excitation, circulation state and tissue adaptability;

[0029] The feature parameters are input into a state recognition model, including support vector machine and long short-term memory neural network, to determine whether the target object is in a high stress, high metabolism or low reactivity state;

[0030] According to the identified state, corresponding electric stimulation parameter combination is matched in the preset stimulation parameter library to generate a control strategy, and the parameters include frequency, voltage, waveform type, pulse width and duty cycle;

[0031] The multi-channel electric pulse output device outputs the stimulation signal under the parameters to drive the conductive ring at different depths of the needle for targeted electric stimulation;

[0032] Synchronous acquisition of stimulation feedback signals, real-time adjustment of parameters based on feedback changes, and closed-loop optimization of the stimulation process.

[0033] In one possible embodiment, the state includes:

[0034] The high stress state matches low frequency single phase wave, and the frequency is 2Hz, and the voltage is less than 1.2V;

[0035] The high metabolism state matches cluster wave combination, and the frequency is 20-40Hz, and the duty cycle is 40%-60%;

[0036] Low reactivity state triggers dynamic sweep frequency wave, frequency range is 1-100Hz, step ≤2Hz.

[0037] In a possible implementation, in the step of synchronously acquiring the stimulation feedback signal, adjusting the parameters in real time based on the feedback change, and realizing closed-loop optimization of the stimulation process, the adjustment parameters include:

[0038] When the EMG response signal amplitude drops by more than 30%, the voltage is increased by 0.2V or the pulse duration is widened;

[0039] When the electrode impedance suddenly rises by more than 10Ω, the output density is reduced and the stimulation interval is lengthened;

[0040] When the skin temperature rises by more than ±1.5℃, the stimulation is automatically paused.

[0041] Compared with the prior art, the embodiments of the application have the following beneficial effects:

[0042] The intelligent adaptive acupuncture electric stimulation system disclosed in the application, by fusing multi-modal physiological signal perception, state recognition and parameter automatic matching, accurate electric stimulation execution and feedback adjustment modules, compared with traditional acupuncture and fixed parameter electric acupuncture instrument, the system not only can respond to individual physiological changes in real time, but also has the ability to adjust the stimulation strategy on demand, so as to realize higher treatment accuracy, safety and individual adaptability. BRIEF DESCRIPTION OF DRAWINGS

[0043] In order to more clearly illustrate the scheme in the application, the drawings needed in the description of the embodiments of the application will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the application, and other drawings can also be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0044] Figure 1 is a structural schematic diagram of an embodiment of the intelligent adaptive acupuncture electric stimulation system according to the application;

[0045] Figure 2 is a flowchart of an embodiment of the method of intelligent adaptive acupuncture electric stimulation according to the application;

[0046] Figure 3 is a structural schematic diagram of an embodiment of the electrode needle according to the application;

[0047] Figure 4 is an acupuncture stimulation treatment analysis diagram of the application;

[0048] Figure 5 is an analysis diagram of the application for reducing body weight and improving abnormal glucose tolerance of the guanyuan acupoint and the zusanli acupoint electric stimulation;

[0049] Figure 6 is the analysis chart of the application of Guanyuan acupoint electrical stimulation to improve the estrous cycle, follicular development and reproductive outcomes of PCOS mice;

[0050] Figure 7 is the analysis chart of the application of Guanyuan acupoint electrical stimulation to improve the serum hormone levels of PCOS mice;

[0051] Figure 8 is the analysis chart of the application of Guanyuan acupoint electrical stimulation to improve the heart rate and heart rate variability. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical scheme and advantages of the present application more clear, the present application will be further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and do not limit the present application.

[0053] Reference Figure 1 , shows the structural schematic diagram of one embodiment of the intelligent adaptive acupuncture electrical stimulation system according to the present application. The intelligent adaptive acupuncture electrical stimulation system 100 comprises:

[0054] The signal acquisition module 101 comprises electromyographic sensors, impedance sensors, temperature sensors and photoelectric plethysmographic sensors, and is used to acquire multi-modal physiological signals of the subject animal in real time;

[0055] The signal processing module 102 is connected with the signal acquisition module, and comprises a signal filtering subunit and a feature extraction subunit, and is used to filter and denoise the collected physiological signals, and extract time domain and frequency domain features;

[0056] The physiological state recognition module 103 is connected with the signal processing module, and is used to input the features to a pre-trained state recognition model, which comprises a support vector machine classifier and an LSTM network, to identify whether the current state is a stress, hyperglycemia or sympathetic nerve activation state;

[0057] The parameter matching and regulation module 104 is connected with the physiological state recognition module, and is used to automatically determine the electrical stimulation parameter combination, including frequency, voltage, waveform type and pulse width, according to the recognition result;

[0058] The electrical stimulation execution module 105 comprises a multi-channel programmable electrical stimulator and an electrode needle, the multi-channel programmable electrical stimulator supports multi-waveform, frequency sweeping and duty cycle adjustment; the electrode needle is provided with a plurality of spaced conductive rings along the length of the needle body, which are isolated by insulating materials and connected to different stimulation channels;

[0059] The feedback control module 106, together with the signal acquisition module and the parameter control module, constitutes a closed-loop control path for dynamically updating the stimulation parameters.

[0060] In this embodiment, a closed-loop treatment framework with the main line of "physiological signal acquisition-state recognition-stimulation parameter generation-current output-feedback correction" is constructed as a whole, aiming to integrate traditional acupuncture with modern electronic stimulation means, while solving the key technical bottlenecks of existing electro-acupuncture devices in terms of stimulation depth control, individualized treatment, and lack of feedback mechanism. The system contains six core modules, which work cooperatively through organic connection to realize real-time optimization of neuro-electric stimulation treatment under complex physiological conditions.

[0061] Among them, the signal acquisition module undertakes the function of system front-end perception, specifically including electromyographic sensors for recording local muscle excitation state, tissue impedance sensors for measuring tissue conductivity and electrode contact state, temperature sensors for monitoring local temperature rise in acupuncture area, and PPG sensors based on optical volume change method for reflecting blood perfusion state. Through the synergistic effect of the four types of sensors, the system can comprehensively perceive the physiological dimensions of the subject individual, such as neuromuscular, bioelectric, temperature and blood flow, forming a high-dimensional, multi-parameter input data stream.

[0062] The collected physiological signals need to be preprocessed by the signal processing module. This module is provided with a filtering subunit dedicated to eliminating non-physiological noise such as power frequency interference and motion artifacts, ensuring the stability and accuracy of the signals; at the same time, it is provided with a feature extraction subunit for extracting key features such as electromyographic root mean square, impedance change rate, spectral density, sympathetic / parasympathetic control index from time domain, frequency domain and even nonlinear dimension, laying a foundation for subsequent physiological state recognition.

[0063] The physiological state recognition module constitutes the core intelligent layer of the system, using machine learning algorithms to classify and judge the extracted features. The module combines support vector machine classifier and long short-term memory neural network model, where SVM is used to efficiently divide the boundary of multi-dimensional data, adapting to static distribution characteristics, while LSTM is good at identifying long-term dependencies in time series, adapting to individual physiological response trajectory learning, and the synergistic effect of the two can accurately judge whether the subject individual is in a typical state such as stress, hyperglycemia, and sympathetic nervousness.

[0064] According to the recognition result, the parameter matching and control module will call the preset stimulation parameter database in the system to automatically select the matching electric stimulation strategy, and the parameter combination includes stimulation frequency (such as 2Hz / 10Hz), voltage amplitude (such as 1.5V-2.5V), waveform type (such as single-phase, double-phase, cluster, sweep), pulse width (such as 100-800μs), etc., and real-time generation of execution commands, replacing the traditional manual setting mode, realizing the individualization and precision of treatment.

[0065] The execution part includes a multi-channel programmable electrical stimulator and an electrode needle. The electrical stimulator is based on an FPGA architecture, can realize multi-channel and multi-mode current output, and is suitable for intervention requirements of different acupoints and depths. The needle is designed to arrange multiple conductive rings (such as every 3 mm, width 0.5 mm) along the axial direction, the conductive sections are insulated by polyimide material, and each section corresponds to an output channel of the electrical stimulator, so that the electrical pulse can be accurately output at different depth levels, and the deep tissue targeting and electric field focusing capability are enhanced.

[0066] The feedback control module is a key link of the closed-loop control of the system, connects the acquisition end and the parameter control end, and is used for judging whether the stimulation intensity is appropriate and whether the tissue response is saturated by monitoring the physiological response after stimulation in real time, and then dynamically correcting the output parameters. For example, when the electromyographic response amplitude decreases or the impedance changes, the system can automatically adjust the stimulation frequency or voltage, realize adaptive response, and improve the consistency and safety of the treatment effect.

[0067] In summary, the embodiment clearly defines the overall structure and core functions of the six key modules, and constructs a highly integrated intelligent acupuncture electrical stimulation system, which has multi-dimensional perception ability, individualized identification and regulation ability, and dynamic feedback regulation ability, and breaks through the technical limitations of traditional electro-acupuncture in treatment accuracy, control intelligence and adaptability.

[0068] Compared with the traditional acupuncture and fixed parameter electro-acupuncture instrument, the system can not only respond to the physiological changes of individuals in real time, but also has the ability to adjust the stimulation strategy on demand, so as to realize higher treatment accuracy, safety and individual adaptability.

[0069] In a possible implementation, the electrode needle has a segmented structure, the conductive section has a width of 0.5 mm, the interval between adjacent conductive sections is 3 mm, and the interval is insulated and separated by using a polyimide material.

[0070] The stimulation signal generation module in the system is mainly used for providing an accurate and controllable electrical pulse signal. By adjusting parameters such as waveform, frequency and amplitude, selective electrical stimulation of the target tissue is realized. The signal is generated by an external control device and conducted to the inside of the tissue through the needle to realize functions such as nerve regulation, tissue activation or treatment intervention. The system supports multiple stimulation modes, including single-phase, double-phase pulse and programmable sequence, and can flexibly set the stimulation scheme according to different clinical needs.

[0071] The needle structure is as follows Figure 3As shown, the whole adopts a layered conductive design, consisting of three functional areas of needle tip 3, needle body 2 and handle 1, taking into account the electrical conductivity, biocompatibility and operational convenience. Among them, the handle 1 represents the insulating layer, the needle body 2 represents the stimulating electrode, and each component cooperates with each other in function to achieve the purpose of precise electrical stimulation conduction.

[0072] The needle tip part is made of medical stainless steel material, which has excellent electrical conductivity and tissue penetration ability, and is the main area of the electrical pulse signal acting on the target tissue. Its high strength and corrosion resistance ensure the reliability of long-term stable work in the tissue.

[0073] The middle section of the needle body is covered with a layer of polyimide insulating material, which not only has good flexibility and high temperature resistance, but also has excellent electrical insulation properties. The design of the insulating layer effectively prevents the leakage of electrical pulses in non-target areas, ensuring precise and safe electrical stimulation. In addition, this layer also provides mechanical support for the needle body, making the overall structure more stable and reliable.

[0074] Under the insulating layer, a conductive stimulating electrode is provided, which is connected to the external electrical stimulation system and is the core medium for electrical pulse conduction. Its precise position design ensures efficient transmission of stimulating signals, and the length and exposure range of the electrode can be adjusted according to different target depths, thereby achieving more detailed stimulation control.

[0075] The handle part is designed with a magnetic electrode interface for quick connection with the external electrical stimulation signal source. Magnetic connection not only makes operation simple, but also improves the efficiency of installation and removal, and ensures the stability and electrical continuity of the connection, suitable for multiple plug-in and clinical rapid wiring needs. The overall structure design not only meets the technical requirements of precise stimulation, but also has good practical operation performance, suitable for various types of neural regulation and minimally invasive treatment scenarios.

[0076] In one possible implementation, the feedback control module includes a state priority scheduling mechanism that prioritizes the execution of safety strategies when multiple abnormal states occur concurrently.

[0077] In this embodiment, the concept of "state priority scheduling mechanism" is proposed, which is to solve the problem of how the system judges which risk should be prioritized when multiple abnormal states (such as excessive temperature, sudden impedance rise, EMG mutation, etc.) occur simultaneously. In actual treatment, different physiological indicators represent different clinical risks, for example, skin temperature abnormalities may indicate the risk of burns and should be prioritized for cooling or stopping stimulation; while mild EMG fluctuations may not require immediate intervention. Therefore, this module pre-sets multiple risk levels and processing priorities, and once multiple abnormalities occur concurrently, the system prioritizes the response operation (such as power-off, current-limiting, etc.) with the highest safety level according to the predetermined weight order, thereby enhancing the safety and robustness of the treatment process.

[0078] The application can quickly make decisions in complex or emergency situations by pre-setting response priorities of different states, and preferentially execute response strategies with greater impact on patient safety (such as current limiting or power off), thereby avoiding treatment accidents caused by concurrent risks not being handled in time.

[0079] In a possible implementation, the feedback control module includes a four-electrode tissue impedance measurement unit, supports EIT reconstruction of current distribution images, and assists in deep stimulation control.

[0080] In this embodiment, impedance data of the target tissue are collected by a dual-frequency four-electrode method, three-dimensional reconstruction of the distribution of the current in the tissue can be realized, and the system can accurately identify the region where the stimulation current is concentrated in combination with the imaging results. When the EIT feedback shows that the current focus is too shallow or deviates from the target region, the system can adjust the stimulation channel selection, waveform parameters or switch the conductive segment of the needle tool accordingly to realize accurate deep targeted electrical stimulation control. This embodiment significantly improves the spatial accuracy of electrical stimulation and effectively solves the problem of poor control of deep nerves / organs by traditional electroacupuncture.

[0081] The application integrates a four-electrode impedance measurement unit and combines with electrical impedance tomography technology, so that the system can realize spatial visualization and depth evaluation of the current distribution.

[0082] In a possible implementation, the electrical stimulator has a 12-bit resolution FPGA waveform synthesizer, which supports switching of cluster waves, dense and sparse waves and logarithmic sweep waves.

[0083] In this embodiment, compared with conventional circuits, the FPGA can realize high programmability and multi-channel output, support real-time switching and dynamic adjustment of the electrical pulse waveform. The supported waveforms include:

[0084] Cluster wave: The number of pulses and the interval between groups can be controlled, which is suitable for improving the intensity of nerve stimulation;

[0085] Dense and sparse wave: The duty cycle is adjusted to adapt to different tissue tolerances;

[0086] Logarithmic sweep wave: Nonlinear increment / decrement can be performed within a certain frequency range, which is beneficial for nerve pathway awakening and adaptive training. These highly adjustable waveform capabilities enable the system to select the most suitable stimulation strategy according to different treatment stages and physiological differences of the subject.

[0087] The application uses a FPGA waveform synthesizer with high resolution, so that the system has higher precision and richer waveform output capability. The cluster wave, dense and sparse wave and sweep wave modes supported by the system not only cover various treatment needs, but also can adjust the output characteristics in real time through parameters such as frequency and duty cycle, thereby optimizing the neural response and tissue tolerance.

[0088] In a possible implementation, the feedback control module sets a three-level safety protection logic, including an EMG amplitude threshold current limiting mechanism, an impedance anomaly automatic power-off mechanism, and a temperature anomaly early warning mechanism.

[0089] In this embodiment, the EMG current limiting mechanism: when the EMG amplitude exceeds a set threshold (such as 150% of the baseline), it may indicate that the nerves are over-excited or the muscles are abnormally contracted, and the system immediately reduces the output current; the impedance anomaly power-off mechanism: poor electrode contact or abnormal tissue reaction often leads to rapid impedance change, and the system immediately stops stimulation when a rapid impedance rise / fall is monitored; the temperature early warning mechanism: if the local temperature changes abnormally, it may cause heat injury or over-stimulation, and should be immediately downshifted. The above three interlocking mechanisms combine the three physiological parameters of electricity, life, and heat to ensure that the system can quickly and safely respond in various emergency situations.

[0090] The present application establishes a physiological safety protection system from three dimensions of EMG, electrical impedance, and temperature through a three-level safety linkage mechanism. Once an anomaly is detected, the system can automatically respond and adjust or suspend stimulation, effectively preventing potential risks such as over-stimulation, heat injury, and needle loss.

[0091] In a possible implementation, the intelligent adaptive acupuncture electric stimulation system further comprises:

[0092] An acupoint positioning module, including an infrared thermal imaging unit and an ultrasonic elastography unit;

[0093] A data recognition module adopts a YOLOv5 deep learning model to perform image recognition and positioning on the target acupoint based on the established acupoint database;

[0094] A projection navigation module projects acupoint position information on the animal body surface based on three-dimensional pose registration technology.

[0095] In this embodiment, the acupoint positioning module: uses infrared thermal imaging and ultrasonic elastography technology to detect the temperature distribution and tissue hardness distribution on the body surface, respectively, to identify pathological abnormal areas;

[0096] The data recognition module: inputs the above image data into a trained and optimized YOLOv5 deep learning model to quickly identify the position of the target acupoint, with an accuracy of over 95%;

[0097] The projection navigation module: uses three-dimensional registration and augmented reality technology to project the identified acupoint coordinates in real time to the animal or patient body surface, providing intuitive and accurate acupuncture positioning reference for the operator.

[0098] The present application realizes automatic, standardized, and visual acupoint recognition and navigation through the fusion of infrared thermal imaging, ultrasonic elastography, and YOLOv5 target detection algorithm.

[0099] Reference Figure 2 As shown in the embodiments of the present application, a method for intelligent adaptive acupuncture electric stimulation is also provided, which adopts the technical solutions as described below:

[0100] A method for intelligent adaptive acupuncture electric stimulation, applied to the intelligent adaptive acupuncture electric stimulation system as described in any possible embodiment above, comprises the following steps:

[0101] S201, implanting an electrode needle with a multi-section conductive structure in a target acupoint area;

[0102] S202, collecting multi-modal physiological signals including electromyographic signals, electrode impedance, skin temperature, and PPG signals;

[0103] S203, filtering, standardizing, wavelet denoising, and feature extraction on the multi-modal physiological signals to obtain time-frequency domain feature parameters reflecting neural excitation, circulatory state, and tissue adaptability;

[0104] S204, inputting the feature parameters into a state recognition model including a support vector machine and a long short-term memory neural network to determine whether a target object is in a high-stress, high-metabolism, or low-reactivity state;

[0105] S205, matching corresponding electric stimulation parameter combinations in a preset stimulation parameter library according to the recognized state to generate a control strategy, wherein the parameters include frequency, voltage, waveform type, pulse width, and duty cycle;

[0106] S206, outputting stimulation signals under the parameters through a multi-channel electric pulse output device to drive the conductive rings at different depths of the needle to perform targeted electric stimulation;

[0107] S207, synchronously acquiring stimulation feedback signals, and adjusting the parameters in real time based on feedback changes to realize closed-loop optimization of the stimulation process.

[0108] In this embodiment, first, an electrode needle with a multi-section conductive structure is implanted in the target acupoint area. The structure of the needle is based on the segmented conductive design defined in the system (such as the interval and width of the conductive sections, the insulating material), so that the conductive sections at different depths can be independently connected to different stimulation channels, thereby achieving precise spatial focusing of electric current in the deep tissue, solving the problem of the traditional electro-acupuncture that cannot control the penetration depth of electric current. Then, a set of key physiological signals are collected, including electromyography (EMG), electrode impedance, skin temperature, and PPG signals. These signals can perceive the local physiological state of the subject from the aspects of electrophysiology, electrical contact, heat conduction, and microcirculation, and are the basis for building a dynamic feedback mechanism. The collected signals need to go through filtering, standardization, wavelet denoising, and feature extraction steps. Filtering and standardization ensure the cleanliness and consistency of the signals, wavelet denoising provides a more suitable denoising method for transient analysis of biological signals, and the feature extraction stage converts the original signals into time-frequency domain feature indicators that can reflect neural excitation, blood flow regulation, and tissue resistance changes. The obtained multi-modal feature parameters are input into a physiological state recognition model, which combines support vector machines (SVM) and long short-term memory networks (LSTM). The former is used to process high-dimensional classification problems in static space, and the latter is good at learning the long-term trends of individuals from time series data, so as to identify whether the target object is in a high stress (sympathetic excitation) state, a high metabolic state, or a neural hypo-reactivity state, etc. According to the recognition result, further match the appropriate electrostimulation parameter combination from the parameter library, and generate a control strategy including frequency, voltage, waveform, pulse width, and duty cycle. This strategy generation method completely gets rid of the traditional acupuncture "manual setting" idea, and realizes automatic adaptation according to the state of the subject. Then, through the multi-channel electric pulse output device in the control system, the electrostimulation signal under the generated parameters is applied to the corresponding conductive section of the electrode needle, realizing precise stimulation of the target acupoint. The current output of different conductive rings can be accurately controlled according to the depth of the tissue, thereby optimizing the stimulation efficiency and targeting. Finally, while the stimulation is being performed, the system continuously synchronously collects physiological feedback signals, evaluates the stimulation effectiveness in real time according to the signal changes, and adjusts the control parameters appropriately, such as reducing the frequency or pausing the output, forming a complete, logically consistent, and agile self-adaptive closed-loop stimulation process.

[0109] The present application integrates multi-modal signal perception, AI recognition, parameter control, and execution feedback into a continuous process, not only solving the problems of traditional electro-acupuncture methods such as "uncontrollable control, unpredictable response, and uncontrollable reaction", but also improving the targeting and dynamic matching ability of treatment through adaptive adjustment, having significant intelligent upgrading value and clinical universality.

[0110] In one possible implementation, the state includes:

[0111] High stress state matches low frequency single-phase wave, frequency is 2Hz, voltage is less than 1.2V;

[0112] High metabolic state matches cluster wave combination, frequency is 20-40Hz, duty cycle is 40%-60%;

[0113] Low reactivity state triggers dynamic sweep wave, frequency range is 1-100Hz, step is less than or equal to 2Hz.

[0114] In the embodiment, when the high stress state is identified (such as increased sympathetic nerve activity, increased EMG discharge, and narrowed PPG blood flow), the matched stimulation strategy is low frequency (2Hz) single-phase wave, and the voltage is set to be lower than 1.2V to simulate the effect of traditional acupuncture in calming, soothing the nerves and relieving sympathetic hyperactivity. When the high metabolic state is identified (such as increased blood flow and temperature reflected by PPG, etc.), the system suggests to use a medium-high frequency cluster wave combination, the frequency is 20-40Hz, and the duty cycle is controlled between 40%-60%. This mode has a good metabolic regulation effect in neuromuscular stimulation. If the nerve low reactivity state is identified (such as slow EMG response, stable impedance but invalid output), the system automatically switches to the dynamic sweep wave output strategy, the frequency is increased between 1-100Hz, and the step is not greater than 2Hz, which aims to wake up the tired or low-sensitive nerve pathways through stimulation frequency scanning.

[0115] The application establishes a "mapping table" of state-parameters, so that the system can quickly call the optimal strategy according to the identification result. This setting not only improves the response speed of the stimulation matching, but also improves the physiological rationality of the output effect, and guarantees the path dependence rationality of the treatment strategy and the consistency of the treatment target.

[0116] In a possible implementation, in the step of synchronously acquiring a stimulation feedback signal, adjusting parameters in real time based on feedback changes, and realizing closed-loop optimization of the stimulation process, the adjusting parameters include:

[0117] When the EMG response signal amplitude decreases by more than 30%, the voltage is increased by 0.2V or the pulse duration is widened;

[0118] When the electrode impedance suddenly rises by more than 10Ω, the output density is reduced and the stimulation interval is prolonged;

[0119] When the skin temperature rises by more than ±1.5℃, the stimulation is automatically paused.

[0120] In this embodiment, if the system detects that the EMG response amplitude drops by more than 30% over a period of time (indicating a weakening of the tissue stimulation response), the output voltage is automatically increased by 0.2V, or the pulse width is extended to increase the stimulation energy; if the electrode impedance is detected to rise suddenly by more than 10Ω (possibly caused by tissue reaction, poor conductivity, needle dislodgment, etc.), the stimulation density is reduced (such as reducing the number of pulses per second) and the stimulation interval is extended to avoid tissue overheating or ineffective stimulation; if the skin temperature continues to rise and exceeds ±1.5°C (such as reaching the dangerous value of local thermal effect), the stimulation is immediately terminated to ensure tissue safety.

[0121] This application uses quantitative judgment of parameters such as EMG response, impedance fluctuation, and temperature anomaly to enable the system to accurately fine-tune strategies or perform security interruptions, thereby improving the system's fine-grained adaptability.

[0122] For example, Figure 4 As shown in the figure, based on the traditional acupuncture points commonly used in the treatment of polycystic ovary syndrome, the five most frequently used acupuncture points in clinical practice were selected: Guanyuan, Zusanli, Sanyinjiao, Uterus, and Tianshu. To investigate whether different acupuncture depths at the same acupoint produce different acupuncture effects, three mice were assigned to each stimulation site. Needling depths at each site were gradually increased from 1 mm to 5 mm. Because lower impedance indicates greater sensitivity to electrical stimulation and a more pronounced therapeutic effect, we measured and recorded the impedance of the electrodes at the stimulation sites. Experimental findings revealed that impedance significantly decreased after piercing tissue barriers (particularly the skin barrier). After transdermal penetration, the impedance measured at different insertion depths at the same acupoint showed minimal difference, indicating that the insertion depth after transdermal penetration had little impact on the experimental results. In summary, as long as the needle penetrates the skin at the insertion site, therapeutic effects can be achieved regardless of the depth of the needle. Furthermore, to identify a safe and effective voltage intensity, three mice were assigned to each stimulation site. The voltage intensity was gradually increased from 0 V until the voltage at which the mouse just convulsed was observed, and the voltage was recorded. Experimental results show that when subjected to electrical stimulation of around 2.5V, muscles generate an electrical signal that can cause mice to twitch. Therefore, when the voltage is less than 2.5V, mice receiving electrical stimulation are in a safe state. We set 2.5V as the high voltage for mice, and set the high, medium, and low stimulation intensities to 2.5V, 1.5V, and 0.5V, respectively. To further verify that 2.5V is indeed a safe high voltage that mice can tolerate, we simultaneously stimulated five acupoints of mice with 2.5V for 30 minutes. After the stimulation stopped, the mice's electrocardiograms were observed for 30 minutes. It was found that when the stimulation ended, the mice's heart rate dropped significantly compared to before the stimulation; about 10 minutes after the stimulation ended, the heart rate began to gradually return to the mice's pre-stimulation heart rate.

[0123] Example 1: Electroacupuncture at Guanyuan and Zusanli acupoints can reduce body weight and improve glucose tolerance in PCOS mice

[0124] C57BL / 6J mice were chosen for 3-4 weeks to establish a PCOS mouse model Figure 5 A) The effect of different numbers of acupuncture points on improving body weight was first evaluated in a polycystic ovary syndrome (PCOS) model. All mice were divided into a control group (Control), a PCOS model group (PCOS), a PCOS+4 needle group (PCOS+4 needles), a PCOS+2 needle group (PCOS+2 needles), and a PCOS+1 needle group (PCOS+1 needle) Figure 5 B) Compared with the PCOS model group, the 4 needle group and the 1 needle group showed a trend of weight loss in mice after treatment Figure 5 C) In the 1 needle treatment group, the effect of different acupuncture points on improving body weight was further evaluated in a polycystic ovary syndrome (PCOS) model. All mice were divided into a control group (Control), a PCOS model group (PCOS), a PCOS+Guanyuan group (CV4), and a PCOS+Zusanli group (ST36). In the 1 needle treatment group, both the Guanyuan group (CV4) and the Zusanli group (ST36) showed a trend of weight loss in mice after treatment Figure 5 D)

[0125] Through intravenous glucose tolerance test (IGTT), it was observed that the blood glucose levels and the area under the blood glucose curve at 30 and 60 min time points in the Guanyuan group (CV4) and the Zusanli group (ST36) were lower than those in the PCOS group, and the Zusanli group (ST36) was more effective than the Guanyuan group (CV4) in improving glucose tolerance in PCOS mice Figure 5 E, F)

[0126] Example 2: Electroacupuncture at Guanyuan acupoint can improve estrous cycle, follicular development, and reproductive outcomes in PCOS mouse models

[0127] The experimental protocol is shown in Figure 6 A, and the estrous cycle was evaluated according to the cell types observed in vaginal smears. Similar to the menstrual cycle in women, a normal estrous cycle is one of the main observations of normal ovarian function in female mice. The control group had a regular estrous cycle, with each cycle lasting 4-5 days. The distribution of mice in each stage was well matched with the typical duration of each stage Figure 6 B) Compared with the PCOS group, Guanyuan acupoint electroacupuncture treatment increased the average number of P days by 0.6875 days, and increased the average number of E days by 0.53 days, and reduced the duration of M / D by 1.27 days. These results indicate that Guanyuan acupoint electroacupuncture treatment can restore the disrupted estrous cycle in PCOS and increase the proportion of mice in estrus Figure 6C).

[0128] By hematoxylin and eosin (HE) staining, it was observed that the follicle distribution at each stage after CV4 treatment showed signs of recovery Figure 6 D). Notably, the appearance of mature follicles after CV4 treatment Figure 6 E) highlights the potential of CV4 treatment in promoting follicle maturation.

[0129] We then conducted a breeding experiment on mice, allowing male and female mice to mate freely, and then recording the pregnancy rate and litter size in two consecutive breeding cycles to assess the reproductive capacity of mice. In two breeding cycles, the pregnancy rate of PCOS mice treated with CV4 was higher than that of the PCOS group (PCOS), about 37.5% Figure 6 F). The number of offspring of PCOS mice treated with CV4 was similar to the control group (Control), with an average of 6.67 Figure 6 G). Our research results show that CV4 treatment can improve the reproductive potential of PCOS mice.

[0130] Example 3: CV4 acupoint electrical stimulation can improve serum hormone levels in PCOS mouse model

[0131] The levels of testosterone (T), luteinizing hormone (LH) and follicle-stimulating hormone (FSH) were evaluated, all of which are related to reproductive endocrine. Compared with the PCOS group, the level of testosterone (T) in the electrical stimulation group decreased significantly, and the effect of CV4 treatment was obvious Figure 7 A, B, C). Further evaluation of the effect of CV4 treatment on reproductive endocrine showed that compared with the PCOS group, the LH content of CV4 group (CV4) decreased significantly, and the LH / FSH ratio decreased significantly Figure 7 D).

[0132] Example 4: CV4 acupoint electrical stimulation improves PCOS mice through autonomic nervous regulation

[0133] Modern neurobiological research shows that the onset of acupuncture is essentially a systemic regulation process mediated by nerves. In order to study whether CV4 acupoint electrical stimulation can improve PCOS mice through autonomic nervous regulation, we conducted a series of bioelectrical physiological experiments to evaluate heart rate changes and heart rate variability. The heart electrical physiology of mice was monitored and recorded 30 minutes before and after electrical stimulation Figure 8 A). The data showed that CV4 acupoint electrical stimulation can reduce the heart rate of mice Figure 8B). Further according to the heart rate variability analysis of electrocardiogram, the results show that, compared with before treatment, the LF / HF ratio decreases, the sympathetic nerve ratio decreases, and the sympathetic vagus balance index decreases after the electric stimulation treatment, indicating that the vagus nerve is dominant, and the electric stimulation treatment can relieve the hyperactivity of the sympathetic nerve Figure 8 C-E).

[0134] The embodiments of the present application can acquire and process related data based on artificial intelligence technology. Among them, artificial intelligence (AI) is to use digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0135] The basic technology of artificial intelligence generally includes technologies such as sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction system, mechatronics, etc. The software technology of artificial intelligence mainly includes computer vision technology, robot technology, biometric technology, speech processing technology, natural language processing technology, and machine learning / deep learning, etc.

[0136] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by computer readable instructions instructing related hardware, and the computer readable instructions can be stored in a computer readable storage medium. When the program is executed, it can include the processes of the above-mentioned embodiments of each method. Among them, the storage medium can be a non-volatile storage medium such as a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0137] It should be understood that, although each step in the flowchart of the accompanying drawings is displayed in sequence according to the direction of the arrow, these steps are not necessarily executed in sequence according to the direction of the arrow. Unless otherwise stated herein, the execution of these steps is not strictly limited in sequence, and they can be executed in other orders. Moreover, at least part of the steps in the flowchart of the accompanying drawings can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence is not necessarily sequential, but can be executed alternately or alternately with at least part of other steps or sub-steps or stages of other steps.

[0138] The above are only the preferred embodiments of the present application and are not used to limit the present application. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. An intelligent adaptive acupuncture electro-stimulation system, characterized in that, The system comprises: a signal acquisition module comprising an electromyography sensor, an impedance sensor, a temperature sensor, and a photoplethysmography sensor, for acquiring multi-modal physiological signals of a subject in real time; a signal processing module connected to the signal acquisition module, comprising a signal filtering subunit and a feature extraction subunit, for filtering and denoising the collected physiological signals, and extracting time-domain and frequency-domain features; a physiological state recognition module connected to the signal processing module, for inputting the features into a pre-trained state recognition model comprising a support vector machine classifier and an LSTM network, to identify whether the subject is in a stress, hyperglycemia, or sympathetic nerve activation state; a parameter matching and regulation module connected to the physiological state recognition module, for automatically determining an electric stimulation parameter combination including frequency, voltage, waveform type, and pulse width according to the recognition result; an electric stimulation execution module comprising a multi-channel programmable electric stimulator and an electrode needle, the multi-channel programmable electric stimulator supporting multi-waveform, frequency sweeping, and duty cycle adjustment, and the electrode needle being provided with multiple spaced conductive rings along the length of the needle body, insulated by an insulating material, and connected to different stimulation channels; a feedback control module forming a closed-loop control path with the signal acquisition module and the parameter regulation module, for dynamically updating the stimulation parameters.

2. The intelligent adaptive acupuncture electro-stimulation system according to claim 1, wherein, The electrode needle has a segmented structure, with a conductive segment width of 0.5 mm and a spacing of 3 mm between adjacent conductive segments, and the spacing is insulated by a polyimide material.

3. The intelligent adaptive acupuncture electro-stimulation system according to claim 1, wherein, The feedback control module comprises a state priority scheduling mechanism, which prioritizes the execution of safety strategies when multiple abnormal states occur simultaneously.

4. The intelligent adaptive acupuncture electro-stimulation system according to claim 3, wherein, The feedback control module comprises a four-electrode tissue impedance measurement unit, which supports electric impedance imaging to reconstruct current distribution images and assist in deep stimulation control.

5. The intelligent adaptive acupuncture electro-stimulation system according to claim 1, wherein, The electric stimulator is equipped with a 12-bit resolution FPGA waveform synthesizer, which supports switching between cluster waves, dense and sparse waves, and logarithmic frequency sweeping waves.

6. The intelligent adaptive acupuncture electro-stimulation system according to claim 1, wherein, The feedback control module sets up a three-level safety protection logic, including an electromyography amplitude threshold current limiting mechanism, an impedance anomaly automatic power-off mechanism, and a temperature anomaly warning mechanism.

7. The intelligent adaptive acupuncture electro-stimulation system according to claim 1, wherein, The intelligent adaptive acupuncture electric stimulation system further comprises: an acupoint positioning module comprising an infrared thermal imaging unit and an ultrasonic elastography unit; a data recognition module using a YOLOv5 deep learning model to perform image recognition and positioning on target acupoints based on an established acupoint database; a projection navigation module projecting acupoint position information on the surface of the animal based on three-dimensional pose registration technology.

8. A method of intelligent adaptive acupuncture electrostimulation, characterized by, The system is applied to an intelligent adaptive acupuncture electric stimulation system as claimed in any one of claims 1 to 7, comprising the following steps: implanting an electrode needle with a multi-segment conductive structure in the target acupoint region; acquiring multi-modal physiological signals including EMG signals, skin impedance at the depth of the electrode needle, skin temperature, and PPG signals; filtering, standardizing, wavelet denoising, and feature extraction on the multi-modal physiological signals to obtain time-frequency domain feature parameters reflecting neural excitation, circulatory state, and tissue adaptability; inputting the feature parameters into a state recognition model comprising a support vector machine and a long short-term memory neural network to determine whether the target object is in a high stress, high metabolism, or low reactivity state; According to the identification state, a corresponding electric stimulation parameter combination is matched in a preset stimulation parameter library to generate a control strategy, the parameters including frequency, voltage, waveform type, pulse width and duty cycle; By controlling a multi-channel electric pulse output device to output a stimulation signal under the parameters, a conductive ring at different depths of a needle is driven to perform targeted electric stimulation; Synchronous acquisition of a stimulation feedback signal, real-time adjustment of parameters based on feedback changes, and closed-loop optimization of the stimulation process.

9. The method of intelligent adaptive acupuncture electrostimulation according to claim 8, characterized by, In the state: The high stress state matches a low-frequency single-phase wave, the frequency is 2 Hz, and the voltage is less than 1.2 V; The high metabolism state matches a cluster wave combination, the frequency is 20-40 Hz, and the duty cycle is 40%-60%; The low reactivity state triggers a dynamic sweep wave, the frequency range is 1-100 Hz, and the step is less than or equal to 2 Hz.

10. The method of intelligent adaptive acupuncture electrical stimulation of claim 8, wherein, In the step of synchronous acquisition of a stimulation feedback signal, real-time adjustment of parameters based on feedback changes, and closed-loop optimization of the stimulation process, the adjustment of parameters includes: When the EMG response signal amplitude decreases by more than 30%, the voltage is increased by 0.2 V or the pulse duration is adjusted; When the electrode impedance suddenly rises by more than 10 Ω, the output density is reduced and the stimulation interval is extended; When the skin temperature rises by more than ±1.5℃, the stimulation is automatically paused.

Images