Neural regulation and control system for degenerative disease treatment based on multi-modal physiological feedback

Abstract

The invention relates to the technical field of nervous system dysfunction, in particular to a nerve regulation and control system for degenerative disease treatment based on multi-modal physiological feedback, which comprises a central processing unit, an optical radiation applicator, a mechanical vibration applicator, an integrated biological signal sensing array and a man-machine interaction module, cooperative stimulation is applied by using a multispectral light source and a broadband vibrator, and multi-dimensional physiological signals such as heart rate variability, myoelectricity, galvanic skin and the like are monitored in real time through a sensor array. A multi-parameter adaptive control algorithm built in the central processing unit can dynamically and intelligently adjust stimulation parameters based on the feedback signals to form an accurate personalized treatment closed loop. Meanwhile, the safety monitoring module based on the biological thermal model ensures the safety boundary of the treatment process. According to the invention, intelligent, self-adaptive and non-invasive treatment of nerve dysfunction is realized.

Description

Technical Field

[0001] This invention relates to the field of nervous system dysfunction technology, specifically to a neuromodulation system for the treatment of degenerative diseases based on multimodal physiological feedback. Background Technology

[0002] Currently, neurodegenerative diseases such as Alzheimer's disease and Parkinson's disease have become major health problems worldwide, seriously affecting the quality of life. These diseases are usually accompanied by the gradual loss of neurons and dysfunction of neural networks, clinically manifesting as impairments in multiple aspects of function, including cognition, motor function, and autonomic nervous system regulation. While existing drug treatments can slow disease progression to some extent, they still have limitations, significant side effects, and large individual differences, failing to achieve a fundamental cure.

[0003] In recent years, neuromodulation technology, as a non-invasive or minimally invasive treatment method, has gradually shown its potential in intervening in neurological dysfunction. For example, transcranial magnetic stimulation (TMS), transcranial direct current stimulation (tDCS), and deep brain stimulation (DBS) have been used in clinical practice, but their modulation precision, targeting, and adaptability still have considerable room for improvement. Especially in terms of real-time response and individualized regulation of multimodal physiological states, most existing systems are still in the open-loop or simple feedback stage, making it difficult to achieve truly "adaptive" treatment.

[0004] On the other hand, photobiological modulation (PBM) and mechanical vibration stimulation, as two emerging neuromodulation modes, have shown potential in improving cellular energy metabolism, promoting nerve regeneration, and regulating nerve excitability, respectively. However, current technologies mostly use these two stimuli separately, lacking a systematic multimodal synergistic strategy and failing to form a closed-loop interaction with real-time physiological signal feedback. Therefore, their therapeutic efficacy and safety have not yet been optimally realized.

[0005] Furthermore, although some devices have incorporated certain physiological signal monitoring functions (such as heart rate and electromyography), they remain relatively weak in multi-dimensional, multi-parameter fusion analysis and intelligent decision-making. In particular, there is a lack of high-order algorithm systems capable of dynamically adjusting stimulation parameters based on the patient's real-time status during treatment, resulting in a low level of personalization and adaptability in the treatment process, which limits their application prospects in long-term, home-based management of degenerative diseases.

[0006] Therefore, there is an urgent need for a closed-loop neuromodulation system that can integrate multimodal neural stimulation with high-precision physiological signal feedback and is supported by intelligent algorithms, in order to improve the treatment effect and patient adaptability for degenerative diseases. Summary of the Invention

[0007] The purpose of this invention is to provide a neuromodulation system for the treatment of degenerative diseases based on multimodal physiological feedback. By integrating photoradiation and mechanical vibration stimulation and performing adaptive closed-loop regulation based on real-time physiological signals, it can be used to safely and accurately treat neurodegenerative diseases.

[0008] To achieve the above-mentioned technical objectives and effects, the present invention is implemented through the following technical solution: A neuromodulation system for treating degenerative diseases based on multimodal physiological feedback, comprising: The system includes a central processing unit; at least one light radiation applicator for applying light energy stimulation to nerve tissue; at least one mechanical vibration applicator for applying mechanical vibration stimulation to the skeletal system or deep tissues; and an integrated biosignal sensing array and human-computer interaction and data management module for real-time acquisition of multi-dimensional physiological signals; the central processing unit is electrically connected to and interacts with the light radiation applicator, the mechanical vibration applicator, the biosignal sensing array, and the human-computer interaction and data management module. The optical radiation applicator is a multispectral light source array, which consists of multiple light-emitting diode units with independently controllable wavelength and power. It can accurately emit blue light (λ1) with a center wavelength in the range of 410 nm to 475 nm, red light (λ2) with a center wavelength in the range of 610 nm to 685 nm, and near-infrared light (λ3) with a center wavelength in the range of 800 nm to 850 nm. The optical power output ratio coefficients k1, k2, and k3 of each band can be dynamically adjusted by pulse width modulation or current amplitude modulation, and satisfy the normalization constraint condition k1 + k2 + k3 ≡ 1. The mechanical vibration applicator is a wideband, high-fidelity electromechanical transducer element, whose frequency response function H v (f) from extremely low frequency f L (where f) L It exhibits a flat amplitude-frequency response over a wide frequency range from ≤ 35Hz to 1000Hz, satisfying |H v (f) | ≥ -3dB; The integrated biosignal sensing array includes at least a photoelectric pulse wave sensor for monitoring cardiovascular activity, a skin conductance sensor for monitoring sympathetic nerve excitability, a surface electromyography sensor for monitoring muscle tension, and a high-precision temperature sensor for real-time monitoring of temperature changes at the treatment site. The central processing unit includes: The signal generation and modulation algorithm module is used to generate composite drive signals modulated by a specific rhythm; The multi-parameter adaptive control algorithm module performs intelligent decision-making and parameter adjustment based on multi-source physiological feedback signals; The safety monitoring and early warning algorithm module is used to ensure that all energy output is strictly within the biosafety range; The signal generation and modulation algorithm module generates a fundamental frequency signal: Where the fundamental frequency f b It can be continuously adjusted or swept according to a preset mode within the physiologically effective range of 8 Hz to 42 Hz, and the base frequency signal is used to control a source signal S. m (t) undergoes complex nonlinear modulation, ultimately outputting a composite driving signal S. d (t); The multi-parameter adaptive control algorithm module receives, fuses, and analyzes multi-channel physiological signals from the biosignal sensing array in real time, and dynamically and adaptively adjusts the driving signal S based on its built-in multiple judgmental logic rules (including but not limited to negative feedback stabilization logic, positive feedback enhancement logic, and state machine switching logic). d The gain G(t) of (t), the fundamental frequency f b (t), the optical power scaling factors k1(t), k2(t), k3(t), and the source signal S m (t) type selection; the safety monitoring and early warning algorithm module is configured to be based on the photothermal transport dynamics model of biological tissue, and to calculate and monitor in real time the irradiance E generated by the light radiation applicator on the surface of the target tissue. avg（λ，t） And ensure that the cumulative energy deposition at any time point t satisfies the following biosafety inequality constraint: Where μ eff (λ) is the effective attenuation coefficient of biological tissue to light radiation of a specific wavelength λ, Φ max (λ) is the maximum safe cumulative energy density threshold preset according to tissue type and illumination time.

[0009] Furthermore, it also includes a respiratory information acquisition module, which is integrated into the biosignal sensor array to simultaneously acquire two key physiological parameters: respiratory rate and blood oxygen saturation. This provides more comprehensive feedback on respiratory and oxygenation status for the central processing unit's multi-parameter adaptive control algorithm. Simultaneously, an intelligent database module is added, which is electrically connected to and works collaboratively with the central processing unit and the human-machine interaction and data management module. Its functions include: Record, store, and modify complete medical record data generated during treatment performed by this instrument at any time (randomly); It can import treatment cases obtained from other similar instruments or different channels for data integration and comparative analysis; It supports importing and exporting treatment plans for different patient groups, facilitating standardized management and personalized application of treatment plans; It integrates a professional citation index database, which includes scientific literature related to neuromodulation and the treatment of degenerative diseases, aiming to provide timely and reliable academic support for treatment decisions and scientific research analysis.

[0010] The beneficial effects of this invention are: This invention achieves precise and dynamic regulation of neural function through the synergistic stimulation of two modes—photoradiation and mechanical vibration—and the deep integration of multidimensional physiological signals. Traditional neuromodulation devices often employ a single stimulus source and lack effective feedback channels, making it difficult to address the complex and variable physiological states in degenerative diseases. The multispectral light source array (covering blue, red, and near-infrared light bands) used in this invention can target neural tissue at different depths, utilizing its unique photobiological regulatory effects to regulate cellular energy metabolism, oxidative stress levels, and neural excitability. Meanwhile, the broadband mechanical vibration applicator acts on deep tissues through mechanisms such as bone conduction, regulating local blood circulation and nerve rhythms. The integrated biosignal sensing array captures multidimensional physiological state information in real time, including cardiovascular (photopulse wave), autonomic nervous system (skin conductance response), muscular system (surface electromyography), and local temperature.

[0011] The system of this invention includes negative feedback stabilization logic, positive feedback reinforcement logic, and state machine switching logic. By analyzing indicators such as heart rate variability (RMSSD) in real time, it determines whether the user is in a state of sympathetic over-excitation or parasympathetic dominance, and accordingly dynamically adjusts the overall output energy and the optical power proportional coefficient and fundamental frequency through a proportional-integral (PI) control law. When an expected rapid decrease in electromyographic and electrodermal signals is detected after stimulation, the currently effective stimulation parameters are appropriately enhanced to reward and consolidate the correct physiological response. The entire treatment process includes initialization, main treatment, maintenance stabilization, and safe extinction stages. The system intelligently switches control strategies and parameter sets according to preset conditions at different stages, ensuring that the system can proactively adapt to the patient's real-time state changes.

[0012] The safety monitoring and early warning algorithm module of this invention is based on the Pennes biothermal equation, a validated model of photothermal transport dynamics in biological tissues, and constructs a real-time computing engine. It can simulate and predict the energy deposition and temperature rise process of light radiation within tissues in real time, calculate its irradiance, and ensure that the cumulative energy deposition always meets strict biosafety inequalities. Intervening before thermal damage occurs fundamentally eliminates the risk of tissue damage due to excessive energy accumulation, providing crucial safety assurance for the system's long-term and home applications.

[0013] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0014] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a schematic diagram of the system modules of the present invention; Figure 2 This is a schematic diagram of the treatment process of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Example 1 A neuromodulation system for treating degenerative diseases based on multimodal physiological feedback, as described in this embodiment, includes: The system includes a central processing unit; at least one light radiation applicator for applying light energy stimulation to nerve tissue; at least one mechanical vibration applicator for applying mechanical vibration stimulation to the skeletal system or deep tissues; and an integrated biosignal sensing array and human-computer interaction and data management module for real-time acquisition of multi-dimensional physiological signals; the central processing unit is electrically connected to and interacts with the light radiation applicator, the mechanical vibration applicator, the biosignal sensing array, and the human-computer interaction and data management module. The optical radiation applicator is a multispectral light source array, which consists of multiple light-emitting diode units with independently controllable wavelength and power. It can accurately emit blue light (λ1) with a center wavelength in the range of 410 nm to 475 nm, red light (λ2) with a center wavelength in the range of 610 nm to 685 nm, and near-infrared light (λ3) with a center wavelength in the range of 800 nm to 850 nm. The optical power output ratio coefficients k1, k2, and k3 of each band can be dynamically adjusted by pulse width modulation or current amplitude modulation, and satisfy the normalization constraint condition k1 + k2 + k3 ≡ 1. The mechanical vibration applicator is a wideband, high-fidelity electromechanical transducer element, whose frequency response function H v (f) from extremely low frequency f L (where f) LIt exhibits a flat amplitude-frequency response over a wide frequency range from ≤ 35Hz to 1000Hz, satisfying |H v (f) | ≥ -3dB; The integrated biosignal sensing array includes at least a photoelectric pulse wave sensor for monitoring cardiovascular activity, a skin conductance sensor for monitoring sympathetic nerve excitability, a surface electromyography sensor for monitoring muscle tension, and a high-precision temperature sensor for real-time monitoring of temperature changes at the treatment site. The central processing unit includes: The signal generation and modulation algorithm module is used to generate composite drive signals modulated by a specific rhythm; The multi-parameter adaptive control algorithm module performs intelligent decision-making and parameter adjustment based on multi-source physiological feedback signals; The safety monitoring and early warning algorithm module is used to ensure that all energy output is strictly within the biosafety range; The signal generation and modulation algorithm module generates a fundamental frequency signal: Where the fundamental frequency f b It can be continuously adjusted or swept according to a preset mode within the physiologically effective range of 8 Hz to 42 Hz, and the base frequency signal is used to control a source signal S. m (t) undergoes complex nonlinear modulation, ultimately outputting a composite driving signal S. d (t); The multi-parameter adaptive control algorithm module receives, fuses, and analyzes multi-channel physiological signals from the biosignal sensing array in real time, and dynamically and adaptively adjusts the driving signal S based on its built-in multiple judgmental logic rules (including but not limited to negative feedback stabilization logic, positive feedback enhancement logic, and state machine switching logic). d The gain G(t) of (t), the fundamental frequency f b (t), the optical power scaling factors k1(t), k2(t), k3(t), and the source signal S m (t) type selection; the safety monitoring and early warning algorithm module is configured to be based on the photothermal transport dynamics model of biological tissue, and to calculate and monitor in real time the irradiance E generated by the light radiation applicator on the surface of the target tissue. avg（λ，t） And ensure that the cumulative energy deposition at any time point t satisfies the following biosafety inequality constraint: Where μ eff (λ) is the effective attenuation coefficient of biological tissue to light radiation of a specific wavelength λ, Φ max(λ) is the maximum safe cumulative energy density threshold preset according to tissue type and illumination time.

[0018] In this embodiment, in the signal generation and modulation algorithm module, the composite driving signal S d The generation of (t) is precisely defined by the following nonlinear time-varying integral equation with memory effect: Wherein: G(t) is a time-varying adaptive gain coefficient, the value of which is calculated and output in real time by the multi-parameter adaptive control algorithm module; DC offset This is a DC bias value used to prevent signal overmodulation during the modulation process; α is a modulation depth coefficient, whose value is strictly limited to between 0 and 1, i.e. 0 < α ≤ 1; Env() is an envelope extraction operator used to extract the fundamental frequency signal S. b（t） Extract its instantaneous temporal envelope information; * indicates the convolution integral operator; K(t) is a kernel function of exponential decay form, and its specific mathematical expression is as follows: Where τ is the decay time constant used to control the duration of the system memory effect; the introduction of this kernel function makes the strength of the current driving signal depend not only on the current fundamental frequency envelope, but also on its historical state over a period of time, thereby simulating the cumulative response characteristics of biological nervous systems to stimuli.

[0019] In this embodiment, the multi-parameter adaptive control algorithm module calculates the heart rate variability time-domain index RMSSD (root mean square of the difference between adjacent cardiac cycles) extracted from the photoelectric pulse wave signal in real time. If the moving average of RMSSD(t) is detected to be continuously lower than a preset lower threshold TH... low (For example, 20ms) exceeds the time window T window1 (For example, 60 seconds), the system determines that the user is in a state of excessive sympathetic nervous system excitation or high stress; at this time, the adaptive control algorithm will activate an inhibitory negative feedback regulation strategy, which dynamically reduces the overall output energy of the system through the following proportional-integral (PI) control law: Among them, U suppress (t) = [G(t), P total (t)] T The control vector to be suppressed mainly includes the driving signal gain and total optical power; K p1 and K i1These are the proportional gain coefficient and integral gain coefficient for this suppressive adjustment, respectively. Simultaneously, the algorithm also coordinates the ground-level dimming power proportional coefficients k2(t) (red light) and / or k3(t) (infrared light), and sets the fundamental frequency f... b（t） Adjusting to a lower frequency range (e.g., the 8-12 Hz alpha wave range) aims to utilize the biostimulatory effects of long-wavelength light waves and the soothing properties of low-frequency rhythms to induce physiological relaxation.

[0020] Conversely, if RMSSD(t) is detected to be continuously higher than the preset upper threshold TH, high (For example, 50ms) exceeds the time window T window2 If the user is in a state of excessive inhibition that is not in line with the therapeutic goal (such as drowsiness), the algorithm will initiate an enhanced negative feedback regulation strategy. Through the inverse operation of the aforementioned control law, it will gently increase the overall output energy intensity, pulling the user's physiological state back to the target therapeutic range. low TH high ]Inside.

[0021] In this embodiment, the root mean square value (RMS) of the surface electromyography (EMG) signal is monitored in real time. RMS (t) and skin conductance level SCL(t). When the multi-parameter adaptive control algorithm module detects that within a short time delay Δt (e.g., 10-30 seconds) after the application of the current stimulation parameters, EMG RMS The values ​​of SCL(t) and SCL(t) both show a significant and rapid decreasing trend, i.e., their time derivatives: and Where D neg and D` neg If the threshold for negative rate of change is reached, then the current combination of stimulation parameters is deemed to have effectively induced the expected relaxation response, and the treatment window is open. At this point, the algorithm will initiate a short-term, small-scale positive feedback reinforcement mechanism to consolidate and amplify the therapeutic effect. Where γ is a positive feedback gain coefficient much smaller than 1 (e.g., 0.05-0.15), β is the attenuation constant, and t trigger The moment that triggers positive feedback. This aims to "reward" the correct physiological response, utilizing the principle of neural plasticity to enhance the consolidation of therapeutic effects. The duration T of this positive feedback loop. boost The system is strictly limited by an internal timer (e.g., 60-120 seconds). After the timer expires, the system automatically switches back to a steady-state control mode dominated by negative feedback to prevent system divergence.

[0022] In this embodiment, the multi-parameter adaptive control algorithm module is based on the state machine switching logic of the treatment process—the system maintains a finite state machine internally, whose states include the initialization and baseline calibration phase, the main treatment phase, the stability maintenance phase, and the safety extinction phase.

[0023] The transition condition from the initialization phase to the main treatment phase is: the system successfully collects and processes sustained T data. initial Resting-state physiological data (e.g., 180 seconds) were collected, and personalized target state interval thresholds were calculated based on this data. low TH high And the initial parameters of the control algorithm.

[0024] The transition condition from the primary treatment phase to the maintenance phase is: the calculated physiological state index (PSI(t)) (calculated by fusing multiple indicators such as HRV, EMG, and SCL) remains consistently and stably within the target range. target low PSI target high [More than T] stable Time (e.g., 300 seconds). During the maintenance phase, the algorithm will automatically decrease the proportional gain coefficient K. p (For example, reducing it to 50% of the original level), thereby reducing the system's regulatory amplitude and response speed, making the output stimulus more stable and gentle, aiming to maintain the current good physiological state.

[0025] The trigger condition for switching from any activity phase to the safe decay phase is: the total treatment duration reaches the preset single treatment time T. session Alternatively, the safety monitoring algorithm module may issue a high-priority intervention command. During the decay phase, all output parameters (gain, power, etc.) no longer follow the aforementioned control law, but instead follow a preset exponential decay curve: The stimulation is smoothly and gradually reduced to zero over several tens of seconds, thus avoiding any discomfort or rebound effect on the user caused by the sudden termination of stimulation.

[0026] In this embodiment, the safety monitoring and early warning algorithm module has the highest system priority, surpassing all control logic. Internally, it runs the Pennes biothermal equation real-time calculation engine to simulate the energy deposition and temperature rise process of light radiation within tissues; the simplified form of this equation is as follows: Where ρ is the tissue density, c p For specific heat capacity, k t μ is the thermal conductivity. a (λ) is the absorption coefficient, μ eff (λ) is the effective attenuation coefficient, Q met For metabolic heat production, Qperf To account for heat loss due to blood perfusion, the system solves the model in real time using the finite difference method or a pre-computed lookup table. If the predicted temperature T(r,t) at a point within the tissue exceeds the safe temperature threshold T... safe Or the rate of temperature rise dT / dt exceeds the maximum allowable rate of temperature rise R. max The module will immediately issue a safety interrupt flag, which will directly override all instructions output by the adaptive control algorithm module and trigger an emergency negative feedback loop. The proportional gain K of this loop... p emergency The gain value is much greater than that of the normal mode, forcing the system to reduce the output power to a safe level or shut down the output completely in a very short time (e.g., within 3-5 seconds).

[0027] In this embodiment, the human-computer interaction and data management module provides a graphical user interface, allowing users or physicians to select from multiple predefined Digital Treatment Protocols (DTPs). Each DTP is a structured data file that fully defines the target physiological state range, initial control parameters, and adaptive algorithm parameters (K parameters for each layer). p K i The system includes all operating parameters such as γ, various thresholds, safety thresholds, treatment phase duration, and photoacoustic coordination mode (synchronous, alternating, random). After the user selects these parameters, the system enters fully automatic operation and encrypts and records all physiological data and control parameter change logs throughout the treatment process for subsequent efficacy evaluation and treatment plan optimization.

[0028] In this embodiment, the system has a hardware self-identification function, which can automatically identify the specific physical form (such as earplug, headgear, or neck collar) of the currently connected light radiation applicator and mechanical vibration applicator, and automatically load the corresponding default treatment protocol and working mode accordingly, thereby achieving seamless switching of application scenarios.

[0029] On the other hand, the present invention proposes an intelligent neural modulation method using the above-mentioned system, comprising the following steps: Before treatment: Perform personalized baseline calibration to establish individualized target state parameters; During treatment: The multi-level adaptive control algorithm runs in real time, dynamically and intelligently adjusting all energy output parameters based on the positive and negative feedback information of physiological feedback signals, and intelligently switching control strategies at different treatment stages; Post-treatment: Automatically generates a comprehensive report containing detailed physiological data trend charts, parameter adjustment records, and safety logs; Among them, the single treatment time Tsession The recommended duration is 20 to 30 minutes. For significant and sustained intervention effects, the cumulative treatment time T should be [calculated]. total It should be no less than 90 hours.

[0030] In this embodiment, the method, through its precise closed-loop feedback and adaptive capabilities, can be widely applied in the following fields: adjunctive treatment and auditory rehabilitation for sudden deafness, sound therapy and habituation therapy for chronic subjective tinnitus, prevention and acute relief of migraines, neuromodulation of anxiety and insomnia, and as a non-pharmacological intervention and function maintenance means in the comprehensive management of neurodegenerative diseases such as Alzheimer's disease and Parkinson's disease.

[0031] On the other hand, the present invention proposes a treatment device based on the above system, including: an earplug-type treatment unit and a headgear-type treatment unit; The earplug-type treatment unit includes: The outer shell is ergonomically designed to conform to the shape of the human ear and ear canal. A light radiation applicator is disposed on the area of ​​the earphone shell facing the conchae, intertragus notch and / or the opening of the external auditory canal, for targeted irradiation of the auricular branch of the vagus nerve, the terminal area of ​​the auriculotemporal nerve and / or the skin of the external auditory canal; A mechanical vibration applicator is embedded in the earphone shell and fits tightly against the tragus, earlobe, or mastoid process to transmit mechanical vibrations to the temporal bone and inner ear structures. The electrodes of the photoelectric pulse wave sensor and the skin conductance sensor of the biosignal sensing array are integrated on the surface of the earphone that contacts the skin of the auricle. The electrodes of the surface electromyography sensor are arranged at the part of the earphone that contacts the mastoid muscle or the auricular muscle behind the ear. The temperature sensor is attached to the inner surface of the earphone shell.

[0032] The headgear-style treatment unit includes: A flexible helmet that can fit snugly around the head or a rigid helmet with an adjustable internal structure; The light radiation applicator is embedded in the cap liner in a distributed two-dimensional array, and its spatial arrangement covers the scalp areas corresponding to the key brain functional areas of Fp1 / Fp2 (frontal lobe), T3 / T4 (temporal lobe) and / or O1 / O2 (occipital lobe) in the international 10-20 system EEG electrode placement standard. Mechanical vibration applicators are fixed in an array on the inside of the cap, corresponding to the areas with thicker bone in the parietal and temporal bones or the area around the foramen magnum of the occipital bone. The various sensors in the biosignal sensing array include photoelectric pulse wave sensors, skin conductance response sensors, surface electromyography sensors, and temperature sensors, which are integrated into the cap liner in a multi-point array manner to ensure reliable contact with the scalp surface for collecting physiological signals.

[0033] In this embodiment, the central processing unit may be placed inside the body of the earbud-type treatment unit or the headgear-type treatment unit, or in an external independent host box that is connected to the wearable body via wired / wireless means.

[0034] Example 2 In this embodiment, to enhance system functionality and expand its application scenarios, the integrated biosignal sensing array may further include attitude and motion sensors, and an inertial measurement unit (IMU) integrating a three-axis accelerometer and a three-axis gyroscope. This is installed within the headgear-style treatment unit to monitor the user's head posture and body movement in real time.

[0035] In this embodiment, the central processing unit is further configured to: receive and fuse data from the posture and motion sensors, photoelectric pulse wave sensors, skin conductance response sensors, and surface electromyography sensors; based on a pre-stored algorithm model, for example by analyzing body movement characteristics, heart rate variability derived indicators, and skin conductance signal levels, to segment and identify the user's sleep state, wherein the sleep state includes at least the wakefulness stage, REM sleep stage, and non-REM sleep stage; and to specifically identify sleep events such as snoring and sleep apnea.

[0036] The multi-parameter adaptive control algorithm module can dynamically adjust the output parameters of the light radiation applicator and the mechanical vibration applicator based on the identified sleep stages or events. When the user is detected to have entered a deep sleep stage, the current stimulation parameters are maintained or fine-tuned to consolidate sleep quality; when a snoring event is detected, a preset low-frequency mechanical vibration mode designed to change head posture or stimulate the opening of the upper airway can be triggered, thereby achieving closed-loop neural regulation based on sleep physiological state.

[0037] Example 3 In this embodiment, to enhance the personalization and precision of treatment, the human-computer interaction and data management module provides an advanced parameter programming interface. Users or physicians can use this interface to finely and customarily set and adjust the output parameters of the light radiation applicator and the mechanical vibration applicator. The adjustable parameters include, but are not limited to: Stimulation intensity corresponds to optical power or vibration amplitude; The stimulation frequency corresponds to the light modulation frequency fb or the dominant vibration frequency. The time pattern includes the duration of a single stimulus, the interval between stimuli, and the total treatment time; The stimulation site, for an applicator array containing multiple independent controllable units, allows for selective activation of applicators at specific sites; and the application sequence and coordination patterns between different modal stimuli, including synchronous, alternating, or sequential triggering.

[0038] In this embodiment, the adjustment mode supports purely manual setting, automatic optimization within a preset safety range, or fully automatic dynamic adjustment based on real-time physiological feedback signals or preset trigger conditions by the multi-parameter adaptive control algorithm module. This gives the stimulation mode of this system programmability comparable to deep brain stimulation (DBS) systems, enabling flexible adaptation to diverse clinical protocols and individual needs.

[0039] Example 4 In this embodiment, to expand the treatment coverage and explore new treatment mechanisms, the hardware implementation of the system of the present invention can be expanded to include a neck brace treatment unit and a shawl treatment unit.

[0040] In this embodiment, the neck-wrap unit has a flexible structure adapted to the curve of the human neck. The inner side integrates the array unit of the light radiation applicator and the mechanical vibration applicator. The arrangement position corresponds to the key anatomical areas of the neck, including the suboccipital region of the posterior neck, the carotid triangle on both sides of the neck, and the sternocleidomastoid muscle region, in order to provide targeted stimulation to the vagus nerve, carotid sinus, jugular vein, and deep lymphatic vessels and lymph nodes passing through the region.

[0041] In this embodiment, the shawl-like unit covers the user's shoulders and upper back area, and its inner side also integrates a stimulation applicator array, acting on muscle groups such as the trapezius muscle and surrounding neural networks. The central processing unit is configured to coordinate and control the applicators on the headgear, neckband, and shawl units, achieving synergistic stimulation in space and time. By applying synergistic stimulation with specific parameters to this complex area, the aim is to promote the return of cervical lymph and venous blood, regulate cerebrospinal fluid circulation dynamics, thereby assisting in enhancing the metabolic waste clearance function of the brain's glymphatic system, providing an innovative non-pharmacological intervention for the treatment of neurodegenerative diseases such as Alzheimer's disease.

[0042] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A neuroregulatory system for treating degenerative disorders based on multi-modal physiological feedback, characterized in that, Comprise: A central processing unit; At least one light radiation applicator for applying light energy stimulation to nervous tissue; At least one mechanical vibration applicator for applying mechanical vibration stimulation to the skeletal system or deep tissue; also comprising an integrated biological signal sensing array for real-time acquisition of multi-dimensional physiological signals, and a human-computer interaction and data management module; the central processing unit is electrically connected and data interacts with the light radiation applicator, the mechanical vibration applicator, the biological signal sensing array and the human-computer interaction and data management module; The light radiation applicator is a multi-spectral light source array composed of multiple light-emitting diode units with independently controllable wavelength and power, which can accurately emit blue light band λ1 with central wavelength in the range of 410-475 nm, red light band λ2 with central wavelength in the range of 610-685 nm, and near-infrared light band λ3 with central wavelength in the range of 800-850 nm, and the light power output proportion coefficients k1, k2, k3 of each band can be dynamically adjusted by pulse width modulation or current amplitude modulation, and satisfy the normalization constraint condition k1+ k2+ k3≡1; The mechanical vibration applicator is a wide-band, high-fidelity electro-mechanical transducer element, whose frequency response function H v (f) has a flat amplitude-frequency characteristic in a wide frequency interval from an extremely low frequency f L v (f) |H(f)| ≥ -3dB;​ The integrated biological signal sensing array includes a photoelectric pulse wave sensor for monitoring cardiovascular activity status, a galvanic skin response sensor for monitoring sympathetic nervous excitability, a surface electromyography sensor for monitoring muscle tension, and a high-precision temperature sensor for real-time monitoring of temperature changes at the treatment site; The central processing unit comprises: A signal generation and modulation algorithm module for generating a composite driving signal modulated by a specific rhythm; A multi-parameter adaptive control algorithm module for intelligent decision-making and parameter adjustment based on multi-source physiological feedback signals; A safety monitoring and early warning algorithm module for ensuring that all energy outputs are strictly within the biological safety range; The signal generation and modulation algorithm module generates a base frequency signal: wherein the fundamental frequency f b The signal S(t) can be continuously tunable in the physiologically effective range of 8 to 42 Hz or swept in a preset pattern and modulated with the fundamental signal S m (t) with complex nonlinear modulation, resulting in a final output of a composite drive signal S d (t). The multi-parameter adaptive control algorithm module receives, fuses and analyzes the multi-channel physiological signals from the biosignal sensing array in real time, and adjusts the driving signal S d (t) according to the gain G(t), the fundamental frequency f b (t), the light power proportionality coefficient k1(t), k2(t), k3(t) and the type selection of the source signal S m (t) according to the judgment logic rules built in the module. The safety monitoring and warning algorithm module is configured to calculate and monitor in real time the irradiance E generated by the optical radiation applicator on the target tissue surface based on a photothermal transport dynamics model of biological tissue avg（λ，t） and ensure that its cumulative energy deposition at any time point t satisfies the following biological safety inequality constraint: where μ eff (λ) is the effective attenuation coefficient of the biological tissue to the optical radiation of a specific wavelength λ, Φ max (λ) is the maximum safe cumulative energy density threshold value preset according to the tissue type and the illumination time.

2. The multi-modal physiologic feedback based neuroregulation system for degenerative disorder treatment as claimed in claim 1, wherein: In the signal generation and modulation algorithm module, the complex driving signal S d The generation of (t) is precisely defined by the following nonlinear time-varying integral equation with memory effect: Where: G(t) is a time-varying adaptive gain coefficient, whose value is calculated and output in real time by the multi-parameter adaptive control algorithm module; DC offset is a direct current bias quantity for preventing signal overmodulation in the modulation process; Alpha is a modulation depth coefficient, whose value is strictly limited between 0 and 1, i.e. 0 < alpha ≤ 1; Env(·) is an envelope extraction operator, which is used to extract the instantaneous time-domain envelope information of the fundamental signal S b（t） from the fundamental signal S * represents the convolution integral operator; K(t) is an exponential decay kernel function, whose specific mathematical expression is: Where τ is the decay time constant for controlling the length of the system's memory effect; the introduction of this kernel function makes the intensity of the current driving signal not only dependent on the current base frequency envelope, but also on its historical state over a period of time, thereby simulating the cumulative response characteristics of biological nervous systems to stimulation.

3. The multi-modal physiologic feedback based neuroregulation system for degenerative disorder treatment as claimed in claim 1 wherein: The multi-parameter adaptive control algorithm module calculates in real time the heart rate variability time domain index RMSSD extracted from the photoelectric pulse wave signal; if the sliding average value of RMSSD(t) is continuously lower than the preset lower threshold TH low over the time window T window1 , the system determines that the user is in a sympathetic nervous system overexcitation or high stress state; at this time, the adaptive control algorithm will start an inhibitory negative feedback regulation strategy, and dynamically down-regulate the overall output energy of the system through the following proportional-integral control law: wherein U suppress (t) = [G(t), P total (t)] T represents the control vector to be suppressed, mainly including the driving signal gain and total optical power; K p1 and K i1 are the proportional gain coefficient and integral gain coefficient for the suppression regulation, respectively; at the same time, the algorithm also cooperatively increases the optical power proportional coefficient k2(t) and / or k3(t), and adjusts the fundamental frequency f b（t） to the lower frequency interval of the alpha wave range of 8-12 Hz, in order to induce the relaxation of physiological state by using the biological stimulation effect of long wavelength light wave and the soothing characteristics of low frequency rhythm; Conversely, if RMSSD(t) is detected to be persistently higher than a preset upper threshold TH high for a time window T window2 , it is determined that the user might have entered an over-inhibited state beyond the therapeutic target; in this case, the algorithm will initiate an enhanced negative feedback regulation strategy to boost the overall output energy intensity by reversing the operation of the control law, and pull the user's physiological state back to the target therapeutic interval [TH low , TH high ].

4. The multi-modal physiologic feedback based neuroregulation system for degenerative disorder treatment as claimed in claim 1, wherein: Real-time monitoring of the root mean square value of the surface electromyographic signal EMG RMS (t) and the skin conductance level SCL(t); when the multi-parameter adaptive control algorithm module detects a significant and rapid downward trend in the values of EMG RMS (t) and SCL(t) simultaneously, i.e. in their time derivatives: and where D neg and D` neg is a negative rate of change threshold, then the current stimulation parameter combination is determined to effectively induce the expected relaxation response, and the therapeutic window is in an open state. At this time, the algorithm will start a positive feedback reinforcement mechanism to consolidate and expand the efficacy: where γ is a positive feedback gain coefficient much smaller than 1, β is a decay constant, t trigger is the time to trigger positive feedback; aims to "reward" the correct physiological response, using the principle of plasticity of the nervous system to enhance the consolidation of the therapeutic effect; the duration T of this positive feedback loop boost is strictly limited by an internal timer, after which the system automatically switches back to the steady-state control mode dominated by negative feedback, to prevent divergence of the system.

5. The multi-modal physiologic feedback based neuroregulation system for degenerative disorder treatment as claimed in claim 1, wherein: The multi-parameter adaptive control algorithm module is based on the state machine switching logic of the treatment process - a finite state machine is maintained inside the system, whose states include initialization and baseline calibration phase, main treatment phase, stable maintenance phase and safety regression phase; The transition condition from the initialization phase to the main therapy phase is that the system successfully acquires and processes resting physiological data for a duration T initial and computes the personalized target state interval threshold [TH low , TH high ] and the initial parameters of the control algorithm from it. The transition condition from the main treatment phase to the maintenance phase is that the physiological state index PSI(t) calculated from the fusion of HRV, EMG, SCL remains stably within the target interval [PSI target low , PSI target high ] for more than T stable time. In the maintenance phase, the algorithm will automatically reduce the proportional gain coefficient K p Thus, the adjustment range and response speed of the system are reduced, and the output stimulation is more stable. The triggering condition to switch from any active phase to the safety fade-out phase is: the total treatment duration reaches a preset single treatment time T session , or the safety monitoring algorithm module issues a highest priority intervention instruction; in the fade-out phase, all output parameters no longer follow the above control law, but follow a preset exponential decay curve: It is smoothly and progressively zeroed within tens of seconds, thereby avoiding any discomfort or rebound effect on the user caused by sudden termination of stimulation.

6. The multi-modal physiologic feedback based neuroregulation system for degenerative disorder treatment as claimed in claim 1, wherein: The safety monitoring and early warning algorithm module has the highest system priority beyond all control logic, and has a Pennes bio-heat equation real-time calculation engine inside for simulating the energy deposition and temperature rise process of optical radiation in tissues; the simplified form of the equation is as follows: where p is the tissue density, c p is the specific heat capacity, k t is the thermal conductivity, p a (λ) is the absorption coefficient, p eff (λ) is the effective attenuation coefficient, Q met is the metabolic heat production, Q perf is the heat loss due to blood perfusion; The system solves the model in real time by finite difference method or pre-computed look-up table; once the real-time calculated prediction of the temperature T(r,t) at a certain point inside the tissue exceeds the safety temperature threshold T safe , or the temperature rise rate dT / dt exceeds the maximum allowed temperature rise rate R max , the module will immediately issue a safety interrupt flag, which will directly override all instructions output by the adaptive control algorithm module and trigger an emergency negative feedback loop with a proportional gain K p emergency much larger than the gain value in normal mode, forcing the system to rapidly reduce the output power to a safe level or completely shut down the output in a very short time.

7. A therapeutic device based on the system of any one of claims 1-6, characterized in that: Comprise: Earplug-type treatment unit and headgear-type treatment unit; The earplug-type treatment unit comprises: A shell shaped according to ergonomic design of human ear and ear canal; Optical radiation applicator arranged on the earphone housing surface facing the region of the cymba concha, intertragic notch and / or the opening of the external auditory canal for targeted irradiation of the auricular branch of the vagus nerve, the region of the auriculotemporal nerve endings and / or the skin of the external auditory canal; Mechanical vibration applicator embedded in the earphone housing and closely fitted with the tragus, the antitragus or the mastoid process for transmitting mechanical vibrations to the temporal bone and the inner ear structures; The electrodes of the photoplethysmogram sensor and the galvanic skin response sensor of the biological signal sensing array are integrated on the surface of the earphone that contacts the pinna skin, and the electrodes of the surface electromyography sensor are arranged on the part of the earphone that contacts the mastoid muscle behind the ear or the auricularis superior muscle; The headgear-type treatment unit comprises: A flexible cap body that can be wrapped around the head or a hard helmet with adjustable internal structure; Optical radiation applicators are distributedly embedded in the lining of the cap body in a two-dimensional array, and the spatial arrangement covers the scalp regions corresponding to the key brain functional areas of the frontal lobe (Fp1 / Fp2), the temporal lobe (T3 / T4) and / or the occipital lobe (O1 / O2) in the international 10-20 system of electroencephalogram electrode placement standard; Mechanical vibration applicators are distributedly fixed on the inner side of the cap body in an array corresponding to the regions with thicker bone mass of the parietal bone, the temporal bone or the periphery of the foramen magnum of the occipital bone; The various sensors of the biological signal sensing array, including the photoplethysmogram sensor, the galvanic skin response sensor, the surface electromyography sensor and the temperature sensor, are integrated in the lining of the cap body in a multi-point array to ensure reliable contact with the scalp surface for physiological signal acquisition.

Images