Tremor stimulation closed-loop control method and computer program product

Abstract

The application discloses a tremor stimulation closed-loop control method and a computer program product, relates to the technical field of biomedical engineering and neural electrical stimulation regulation, and can accurately identify pathological tremors requiring intervention by continuously judging the tremor main frequency and tremor amplitude of a patient in a preset window; the safety and comfort of electrical stimulation treatment and the precise effectiveness of dynamic correction of pathological tremors are taken into account by determining the tremor symptom subtype category in an initial starting state, performing gradient electrical stimulation intervention treatment, and performing layered intervention of gradient electrical stimulation smooth adaptation transition and dynamic closed-loop real-time tremor correction, thereby improving the adaptability and comfort of patients in the initial treatment stage; in a non-initial starting state, real-time correction intervention treatment is performed, the stimulation intervention intensity is dynamically adapted according to the real-time fluctuation of the tremor main frequency and tremor amplitude of the patient, and the intelligent, efficient and comfortable intervention treatment purpose of essential tremor is achieved.

Description

Technical Field

[0001] This disclosure generally relates to the fields of biomedical engineering and neuroelectric stimulation modulation technology, and in particular to a tremor stimulation closed-loop control method and computer program product. Background Technology

[0002] Traditional transcutaneous electrical nerve stimulation (TENS) therapy typically uses an open-loop control approach to intervene in essential tremor, which means that fixed stimulation parameters such as stimulation frequency, stimulation pulse width, and stimulation amplitude are preset to implement continuous electrical stimulation therapy.

[0003] However, because the tremor symptoms of patients with essential tremor are prone to significant fluctuations with their movements and emotional states, using a fixed-parameter stimulation mode for electrical stimulation therapy can easily lead to problems such as overstimulation causing muscle fatigue when the tremor is mild, or insufficient stimulation intensity and poor tremor suppression effect when the tremor is severe. This approach cannot adapt to the real-time changes in the patient's tremor state, resulting in poor treatment suitability and intervention effectiveness. At the same time, the above-mentioned open-loop control method usually uses a single stimulation mode and lacks a smooth transition start-up control logic. Directly applying a fixed intensity of electrical stimulation can easily cause discomfort such as muscle spasms and tingling on the skin surface, resulting in a poor treatment experience for the patient. Summary of the Invention

[0004] In view of the aforementioned defects or deficiencies in the existing technology, it is desirable to provide a closed-loop control method and computer program product for tremor stimulation. By continuously judging the patient's tremor dominant frequency and tremor amplitude within a current preset window, the method can accurately identify pathological tremors requiring intervention. In the initial startup state, the method determines the tremor symptom subtype and performs gradual electrical stimulation intervention. It also employs a tiered intervention approach, first using gradient-decreasing gradual electrical stimulation for a smooth transition, followed by dynamic closed-loop real-time tremor correction. This approach balances the safety and comfort of electrical stimulation therapy with the precise effectiveness of dynamic correction of pathological tremors, improving patient adaptability and comfort in the initial treatment phase. In the non-initial startup state, the method performs real-time tremor correction intervention, dynamically adapting the stimulation intensity according to the patient's real-time fluctuating tremor dominant frequency and amplitude, thereby achieving a more intelligent, efficient, and comfortable intervention for idiopathic tremor.

[0005] Firstly, this application provides a tremor stimulation closed-loop control method, applied to a tremor stimulation closed-loop control system. The method includes: Obtain the dominant frequency and amplitude of tremor in patients with essential tremor within the current preset window; If, within the current preset window, the dominant frequency of the tremor remains within the preset pathological frequency band of essential tremor and the amplitude of the tremor remains above the preset non-pathological physiological baseline threshold, then it is determined whether the system is currently in the initial startup state. If the system is currently in the initial startup state, the first tremor symptom subtype category to which the essential tremor patient currently belongs is determined. According to the first stimulation parameter corresponding to the first tremor symptom subtype category, the patient with essential tremor is subjected to progressive electrical stimulation intervention treatment until the actual output stimulation parameter increases to the benchmark anchoring parameter corresponding to the first tremor symptom subtype category. Then, the patient with essential tremor is subjected to real-time tremor correction intervention treatment. If the system is not currently in an initial startup state, then real-time tremor correction intervention is performed on the patient with essential tremor.

[0006] In conjunction with the first aspect, in one possible implementation, the real-time tremor correction intervention for the patient with essential tremor includes: Determine the tremor frequency band energy of the patient with essential tremor within the current preset window; the tremor frequency band energy is the integral result of the energy or amplitude within the frequency band corresponding to the dominant tremor frequency within the current preset window; If the energy of the tremor frequency band is higher than the preset energy upper limit threshold, the stimulation amplitude in the first stimulation parameter is adjusted in a forward stepping manner, and the actual stimulation amplitude output to the idiopathic patient is determined based on the upper limit extreme value of the stimulation amplitude of the first tremor symptom subtype category and the stimulation amplitude after forward stepping adjustment. If the energy of the tremor frequency band remains below the preset lower energy threshold for a first preset duration, the stimulation amplitude in the first stimulation parameter is adjusted in a negative stepping manner, and the actual stimulation amplitude output to the idiopathic patient is determined based on the lower limit extreme value of the stimulation amplitude of the first tremor symptom subtype and the stimulation amplitude after negative stepping adjustment.

[0007] In conjunction with the first aspect, in one possible implementation, determining the actual stimulus amplitude output to the idiopathic patient based on the upper limit extreme value of the stimulus amplitude for the first tremor symptom subtype category and the positively stepped-adjusted stimulus amplitude includes: If the positively stepped adjustment of the stimulation amplitude exceeds the upper limit of the stimulation amplitude of the first tremor symptom subtype, then the upper limit of the stimulation amplitude is determined to be the actual stimulation amplitude. If the positively stepped stimulation amplitude does not exceed the upper limit of the stimulation amplitude for the first tremor symptom subtype, then the positively stepped stimulation amplitude is determined to be the actual stimulation amplitude.

[0008] In conjunction with the first aspect, in one possible implementation, the step of performing gradual electrical stimulation intervention on the patient with essential tremor according to the first stimulation parameter corresponding to the first tremor symptom subtype category includes: Based on the preset step time interval and the first adaptive ramp rate in the first stimulation parameters, the step stimulation amplitude output to the patient with essential tremor at the current step time is determined; the current step time is the sum of the current time and m preset step time intervals, the current time belongs to the current preset window, and the initial value of m is 1; Based on the upper limit of the stimulation amplitude in the first stimulation parameter and the step stimulation amplitude, the actual step stimulation amplitude output to the idiopathic patient is determined. If the actual step stimulation amplitude does not rise to the first reference stimulation amplitude in the first stimulation parameter and the patient's tremor symptoms are not eliminated, then the value of m is incremented by 1, and the above steps are repeated.

[0009] In conjunction with the first aspect, in one possible implementation, the method further includes: If the first tremor symptom subtype category is updated to the second tremor symptom subtype category within the current preset window, and the patient's tremor symptoms have not been eliminated, then the current output stimulation parameters are controlled to gradually decrease to the preset safety benchmark level or zero with a preset downward slope. Then, according to the second benchmark stimulation amplitude and the second adaptive ramp rate corresponding to the second tremor symptom subtype category, a new gradual electrical stimulation intervention is performed on the essential tremor patient.

[0010] In conjunction with the first aspect, in one possible implementation, determining the first tremor symptom subtype category to which the essential tremor patient currently belongs includes: Obtain the multidimensional physiological characteristics of limb tremor of the patient with essential tremor within the current preset window, and perform feature standardization and dimensionality reduction on the multidimensional physiological characteristics of limb tremor. Based on the matching relationship between the multidimensional physiological characteristics of limb tremor after dimensionality reduction and various tremor symptom types, the first tremor symptom subtype category to which the patient with essential tremor currently belongs is determined.

[0011] In conjunction with the first aspect, in one possible implementation, the method further includes: If, within the current preset window, the dominant frequency of the tremor is not continuously within the preset pathological frequency band and / or the amplitude of the tremor is not continuously exceeding the non-pathological physiological baseline threshold, then the variance of the composite acceleration signal and the variance of the composite angular velocity signal of the multi-axis acceleration signal within the current preset window are determined. If, within a preset resting time, the variance of the synthetic acceleration signal remains below a preset resting acceleration threshold and the variance of the synthetic angular velocity signal remains below a preset resting angular velocity threshold, then the patient with essential tremor is determined to be in a limb resting state and enters a low-power standby mode or a low-intensity maintenance stimulation state.

[0012] In conjunction with the first aspect, in one possible implementation, obtaining the dominant frequency and amplitude of tremor in a patient with essential tremor within the current preset window includes: When the patient with essential tremor wears a tremor stimulation wearable device and the device base plate integrates an M-axis inertial sensor, the device pitch angle and device roll angle are determined, and the projection vector of the tremor plane in the local coordinate system of the sensor is determined based on the device pitch angle and the device roll angle. Based on the projection vector, the flutter capture contribution of the M-axis inertial sensor is analyzed, and the candidate redundant axis with the lowest flutter capture contribution is selected. The variance of the angular velocity signal of the candidate redundant axis within a preset observation window is determined. If the variance of the angular velocity signal is continuously lower than a preset minimum threshold within the preset observation window, the candidate redundant axis is masked to obtain an N-axis inertial sensor for feature extraction. The N-axis sensor is the remaining inertial sensor after removing the candidate redundant axis from the M-axis inertial sensor. M > N, and both M and N are positive integers greater than 0. The inertial motion data of the N-axis inertial sensor is periodically collected within the current preset window. The inertial motion data is subjected to spectrum analysis to extract the main frequency of the vibration and the vibration amplitude.

[0013] In conjunction with the first aspect, in one possible implementation, the method further includes: During the execution of the gradual electrical stimulation intervention or the real-time tremor correction intervention, the impedance of the stimulation electrode is collected in real time using a preset detection current. If the detected impedance of the stimulation electrode is lower than a preset short-circuit threshold or higher than a preset detachment abnormality threshold, the electrical stimulation treatment will be forcibly terminated through a preset hardware interruption.

[0014] Secondly, this application also provides a computer-readable storage medium. This computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the tremor stimulation closed-loop control method described in the first aspect.

[0015] This application provides a closed-loop control method and computer program product for tremor stimulation. The closed-loop control method acquires the dominant frequency and amplitude of tremor within a preset window and makes continuous judgments based on these values. This accurately identifies pathological tremors requiring intervention, avoiding misjudgments of normal physiological activities or occasional tremors found in traditional methods. In the initial startup state, it not only introduces tremor symptom subtypes and implements a gradual electrical stimulation intervention strategy, allowing for a smooth transition from low to high stimulation intensity, but also introduces a layered intervention strategy: first, a gradual, gradient-decreasing electrical stimulation for smooth adaptation, followed by dynamic closed-loop real-time tremor correction. This approach effectively addresses the problems of traditional methods, such as sudden parameter changes during the cold start phase causing patient discomfort and poor treatment adaptability and stability due to the lack of operational condition differentiation. It balances the safety and comfort of electrical stimulation therapy with the precise and effective dynamic correction of pathological tremor, significantly improving patient adaptability and comfort in the initial treatment phase. Furthermore, the strategy of introducing real-time tremor correction intervention in non-initial start states dynamically adapts the stimulation intensity based on the patient's real-time fluctuating tremor dominance and amplitude. This overcomes the technical problems of traditional open-loop fixed-parameter therapy, which cannot adapt to dynamic tremor changes, has unstable tremor suppression effects, and poor adaptability, achieving continuous, precise, and dynamic closed-loop suppression therapy for pathological tremor. Thus, through refined tremor status assessment, personalized subtype matching, smooth transition start strategies, and real-time closed-loop feedback control, it overcomes the limitations of traditional open-loop fixed-parameter electrical stimulation therapy in terms of adaptability, intervention effect, and patient experience, achieving a more intelligent, efficient, and comfortable intervention for essential tremor. Attached Figure Description

[0016] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is one of the flowcharts illustrating a tremor stimulation closed-loop control method in one embodiment; Figure 2 This is a second schematic flowchart of a tremor stimulation closed-loop control method in one embodiment; Figure 3 This is the third flowchart of a tremor stimulation closed-loop control method in one embodiment; Figure 4 This is the fourth flowchart of a tremor stimulation closed-loop control method in one embodiment; Figure 5 This is the fifth flowchart of a tremor stimulation closed-loop control method in one embodiment; Figure 6 This is the sixth flowchart of a tremor stimulation closed-loop control method in one embodiment; Figure 7This is the seventh flowchart of a tremor stimulation closed-loop control method in one embodiment; Figure 8 This is the eighth flowchart of a tremor stimulation closed-loop control method in one embodiment; Figure 9 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0017] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0018] It should be noted that, unless otherwise specified, the embodiments and features described in the embodiments of this invention can be combined with each other. The invention will now be described in detail with reference to the accompanying drawings and embodiments. Furthermore, the term "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. The terms "first" and "second," etc., in the specification and claims of the embodiments of this invention are used to distinguish different objects, not to describe a specific order of objects.

[0019] Traditional TENS typically employs an open-loop control approach to intervene in essential tremor, which involves pre-setting fixed stimulation parameters such as stimulation frequency, pulse width, and amplitude to implement continuous electrical stimulation therapy.

[0020] However, because the tremor symptoms of patients with essential tremor are prone to significant fluctuations with their movements and emotional states, using a fixed-parameter stimulation mode for electrical stimulation therapy can easily lead to problems such as overstimulation causing muscle fatigue when the tremor is mild, or insufficient stimulation intensity and poor tremor suppression effect when the tremor is severe. This approach cannot adapt to the real-time changes in the patient's tremor state, resulting in poor treatment suitability and intervention effectiveness. At the same time, the above-mentioned open-loop control method usually uses a single stimulation mode and lacks a smooth transition start-up control logic. Directly applying a fixed intensity of electrical stimulation can easily cause discomfort such as muscle spasms and tingling on the skin surface, resulting in a poor treatment experience for the patient.

[0021] While traditional TENS (Temperature-Assisted Neurostimulation) devices have employed adaptive closed-loop control methods to intervene in essential tremor, they heavily rely on external sensors, making it difficult to achieve low-power, low-latency closed-loop control on wristwatch-level devices. Furthermore, existing adaptive closed-loop control methods often employ a single linear feedback adjustment approach. When patients initially wear the tremor stimulation wearable device for a cold start, or when limb movement changes abruptly, a considerable trial-and-error time is required to converge to the optimal stimulation parameters. This not only increases the computational burden but also easily causes discomfort such as muscle spasms and tingling in patients. Cold start (initial patient wear) refers to the initial operating condition of the device when the patient first puts on the tremor stimulation wearable device, the tremor stimulation closed-loop control system completes power-on initialization, and formal tremor intervention treatment has not yet commenced.

[0022] To address the aforementioned technical issues, this application proposes a closed-loop control method and computer program product for tremor stimulation. By continuously assessing the patient's tremor dominance frequency and amplitude within a preset window, it can accurately identify pathological tremors requiring intervention. In the initial startup state, it determines the tremor symptom subtype and executes gradual electrical stimulation intervention. Furthermore, it employs a tiered intervention approach: first, a gradient-decreasing gradual electrical stimulation for a smooth transition, followed by dynamic closed-loop real-time tremor correction. This approach balances the safety and comfort of electrical stimulation therapy with the precise effectiveness of dynamic correction of pathological tremors, improving patient adaptability and comfort in the initial treatment phase. In the non-initial startup state, it executes real-time tremor correction intervention, dynamically adapting the stimulation intensity based on the patient's real-time fluctuating tremor dominance frequency and amplitude, thereby achieving a more intelligent, efficient, and comfortable intervention for idiopathic tremor.

[0023] The following is combined with Figures 1 to 9 This application describes a tremor stimulation closed-loop control method and computer program product. The tremor stimulation closed-loop control method is applied to a tremor stimulation closed-loop control system, which typically includes a sensor module (such as a multi-axis inertial sensor), a processing module, and a stimulation output module. The execution entity of the tremor stimulation closed-loop control method is the processing module within the tremor stimulation closed-loop control system. Specifically, the processing module may be a main controller, a microcontroller unit (MCU), an embedded control unit, or a main control processing unit, etc. This application does not specifically limit this. The following description uses the example of the MCU built into the tremor stimulation closed-loop control system as the execution entity of the tremor stimulation closed-loop control method to illustrate this approach.

[0024] To facilitate understanding of the tremor stimulation closed-loop control method provided in this application, the following examples will provide a detailed description of the tremor stimulation closed-loop control method. It is understood that these examples can be combined with each other, and similar concepts or processes may not be repeated in some embodiments.

[0025] Reference Figure 1 This is a schematic flowchart of the tremor stimulation closed-loop control method provided in the embodiments of this application, as shown below. Figure 1 As shown, the tremor stimulation closed-loop control method includes the following steps 101 to 104.

[0026] Step 101: Obtain the dominant frequency and amplitude of tremor in patients with essential tremor within the current preset window.

[0027] Among them, the dominant frequency of tremor is the number of times a patient's tremor limb completes a full vibration cycle per unit time when the tremor limb makes regular reciprocating shaking movements within a preset window. It is used to characterize the speed of limb tremor (that is, to reflect the frequency of tremor attacks) and is the core frequency characteristic parameter for distinguishing between physiological micromovements and pathological tremors.

[0028] The tremor amplitude is the maximum deviation of the vibration trajectory from the static equilibrium baseline position during the reciprocating shaking motion of the patient's tremor limb within the same preset window. It is used to characterize the intensity and amplitude of limb tremor and reflect the strength of tremor.

[0029] Specifically, after a patient with essential tremor wears a tremor stimulation wearable device (such as a wearable wristband), the inertial motion sensor / inertial measurement unit integrated inside the device (such as the base plate of the wearable wristband) continuously collects the original simulated spatial vibration signals of the patient's tremor limbs in real time at a pre-set high-frequency sampling frequency. The sampling frequency is set to >60Hz to ensure complete capture of the subtle tremor motion trajectory.

[0030] Meanwhile, the MCU pre-sets a 2-second duration as a data statistics preset window, capturing a complete segment of the original simulated spatial vibration signal within the preset window every 2 seconds. The MCU first performs low-pass filtering on each segment of the original simulated spatial vibration signal to remove low-frequency interference signals generated by large limb movements such as walking, raising arms, and turning over, while also filtering out high-frequency electromagnetic noise signals from the circuit, retaining only the regular oscillating waveform signal caused by pathological tremor. Subsequently, the MCU performs Fast Fourier Transform spectral analysis on the remaining regular oscillating waveform signal after filtering, extracting the frequency value corresponding to the main peak of the waveform, which can then be determined as the dominant tremor frequency of the essential tremor patient within the current preset window. Then, by calculating the difference between the peak value of the waveform vibration and the baseline steady-state value, the amplitude fluctuation is statistically obtained and determined as the tremor amplitude of the essential tremor patient within the current preset window.

[0031] Alternatively, after a patient with essential tremor wears a tremor stimulation wearable device (such as a wearable wristband), the device's built-in motion sensing unit (such as the wristband's base plate) collects temporal motion data of the patient's limb shaking at a preset high-frequency sampling frequency. The MCU is pre-set to a 2-second analysis window and extracts the temporal motion data segment by segment. The MCU first performs smoothing filtering on the extracted temporal motion data to remove irregular noise caused by daily micro-movements, postural shifts, etc., retaining only the periodic reciprocating shaking waveform. For the remaining periodic reciprocating shaking waveform after filtering, within the corresponding preset window, the number of complete reciprocating vibrations of the shaking waveform per unit time is counted, and this is used to calculate the tremor dominant frequency of the patient with essential tremor within the current preset window. Then, the difference between the maximum forward offset and the maximum reverse offset of the vibration waveform within the corresponding preset window is calculated to obtain the tremor amplitude of the patient with essential tremor within the current preset window.

[0032] It should be noted that the wearable wristband houses a hardware pulse generator, specifically an electrical stimulation output circuit or an analog front end (AFE), which is electrically connected to the MCU. The hardware pulse generator receives digital parameters (i.e., target current amplitude, pulse width, and frequency, etc.) calculated and clamped by the MCU's closed-loop algorithm. Through an internal digital-to-analog converter (DAC) and constant current / constant voltage output stage circuit, the digital signal is converted into nerve electrical stimulation pulses with corresponding energy characteristics, which are ultimately applied to the patient via skin electrodes.

[0033] Step 102: If the dominant frequency of tremor is continuously within the preset pathological frequency band of essential tremor within the current preset window, and the tremor amplitude continuously exceeds the preset non-pathological physiological baseline threshold, then determine whether the system is currently in the initial startup state.

[0034] The initial startup state refers to the initial working condition of the device when the patient first puts on the tremor stimulation wearable device, the tremor stimulation closed-loop control system has completed power-on initialization, and formal tremor intervention treatment has not yet been carried out; or, the stage when the tremor stimulation closed-loop control system is turned on for the first time or reactivated after a long period of inactivity, requiring a smooth transitional stimulation intervention.

[0035] The pre-defined pathological frequency band range for essential tremor is a specific frequency range determined based on medical research and clinical experience, in which the dominant tremor frequency of patients with essential tremor typically falls. When the dominant tremor frequency falls within this range, it indicates that the patient may have a pathological tremor.

[0036] The preset non-pathological physiological baseline threshold is the tremor amplitude limit used to distinguish between pathological tremor and normal physiological tremor. When the tremor amplitude consistently exceeds this non-pathological physiological baseline threshold, it indicates that essential tremor may require clinical intervention.

[0037] For example, the preset pathological frequency band range for essential tremor is 4Hz to 12Hz.

[0038] Specifically, for patients with essential tremor within the current preset window, the MCU first determines whether the tremor dominant frequency and tremor amplitude simultaneously meet the frequency domain determination condition, the energy domain determination condition, and the time domain anti-accidental touch condition. The frequency domain determination condition is that the tremor dominant frequency is within the preset pathological frequency band of essential tremor. The energy domain determination condition is that the tremor amplitude exceeds the preset non-pathological physiological baseline threshold. The time domain anti-accidental touch condition is that the above-mentioned dual over-limit state in the frequency domain and energy domain continues to exist within the current preset window.

[0039] Only when the frequency domain determination condition, energy domain determination condition, and time domain anti-accidental trigger condition are simultaneously met—a combined determination condition—can it be determined that the patient with essential tremor currently does not have non-pathological high-frequency noise and low-frequency voluntary movements. This confirms an action tremor episode and triggers subsequent dimensionality reduction and treatment procedures, i.e., it is determined to be a valid tremor event. Through this combined determination method, the system can initially identify pathological tremors that require intervention, avoiding unnecessary interventions to normal physiological activities or occasional tremors. Under the premise that a valid tremor event is established, further system state detection is performed to determine whether the current system state is in the initial startup state.

[0040] Step 103: If the system is currently in the initial startup state, determine the first tremor symptom subtype category to which the essential tremor patient belongs. Based on the first stimulation parameter corresponding to the first tremor symptom subtype category, perform progressive electrical stimulation intervention on the essential tremor patient until the actual output stimulation parameter increases to the baseline anchoring parameter corresponding to the first tremor symptom subtype category. Then, perform real-time tremor correction intervention on the essential tremor patient.

[0041] Among them, the benchmark anchoring parameter corresponding to the first tremor symptom subtype category can be the lower limit extreme value of the stimulus amplitude or the lower limit extreme value of the stimulus intensity corresponding to the first tremor symptom subtype category.

[0042] It should be noted that if the system is currently in its initial startup state, a series of specific intervention strategies need to be implemented. Specifically, the first step is to determine the primary tremor symptom subtype to which the essential tremor patient currently belongs. This can be done by inquiring about the patient's medical history and symptom description, or by having a doctor make a diagnosis based on clinical experience, thereby determining which tremor subtype the patient belongs to, i.e., determining the primary tremor symptom subtype to which the essential tremor patient currently belongs.

[0043] Based on this, gradual electrical stimulation intervention is performed on patients with essential tremor according to the first stimulation parameters corresponding to the first tremor symptom subtype category. For example, a list of stimulation parameters can be preset, which contains multiple tremor symptom subtype categories, and each tremor symptom subtype category corresponds to a set of stimulation parameters; the MCU can determine the first stimulation parameters corresponding to the first tremor symptom subtype category by querying this list of stimulation parameters.

[0044] Based on the first stimulation parameters corresponding to the first tremor symptom subtype, gradual electrical stimulation intervention is performed on patients with essential tremor. This gradual electrical stimulation intervention can be implemented in various ways. For example, a fixed stimulation amplitude increase step size can be set, with the amplitude increase step size increased at intervals until the target stimulation intensity is reached. Alternatively, the stimulation intensity can be gradually increased according to a preset curve showing a linear increase in stimulation intensity over time. This gradual electrical stimulation intervention process continues until the actual output stimulation parameters increase to the benchmark anchoring parameters corresponding to the first tremor symptom subtype. For example, a fixed lower limit for stimulation intensity can be set; when the stimulation intensity during the gradual process exceeds this lower limit, the gradual electrical stimulation intervention is considered complete. Subsequently, real-time tremor correction intervention is performed on patients with essential tremor, such as dynamically adjusting the electrical stimulation parameters according to the patient's real-time tremor status to continuously and effectively suppress tremor.

[0045] Step 104: If the system is not currently in an initial startup state, perform real-time tremor correction intervention on patients with essential tremor.

[0046] Specifically, if the system is not currently in its initial startup state, it will directly perform real-time tremor correction intervention on patients with essential tremor. This means that the system has completed the initial startup and adaptation process and can directly enter the normalized tremor suppression mode. In this tremor suppression mode, the system will continuously monitor the patient's tremor status and adjust the stimulation parameters based on real-time feedback to maintain the best tremor suppression effect.

[0047] For example, after completing the gradual electrical stimulation intervention, the MCU immediately transitions to real-time tremor correction intervention for a patient with essential tremor (e.g., patient A). During this phase, the system continuously monitors patient A's tremor dominance frequency and amplitude, and dynamically adjusts the stimulation parameters based on these real-time data. For instance, if an increase in patient A's tremor amplitude is detected, the MCU will fine-tune the stimulation amplitude accordingly to enhance the inhibitory effect; if the tremor amplitude decreases, the MCU will correspondingly reduce the stimulation amplitude to avoid overstimulation. This real-time tremor correction intervention strategy ensures that, even in the non-initiation phase, the system can rapidly respond to changes in the patient's tremor state, providing continuous and effective intervention.

[0048] The tremor stimulation closed-loop control method provided in this application obtains the dominant frequency and amplitude of tremor within the current preset window and makes continuous judgments based on the dominant frequency and amplitude. This accurately identifies pathological tremors requiring intervention, avoiding misjudgments of normal physiological activities or occasional tremors in traditional methods. In the initial startup state, it not only introduces tremor symptom subtype categories and implements a strategy of gradual electrical stimulation intervention, allowing the stimulation intensity to smoothly transition from low to high, but also introduces a layered intervention strategy of first using gradient-decreasing gradual electrical stimulation for smooth adaptation and then dynamic closed-loop real-time tremor correction. This effectively solves the problem of the coldness of traditional solutions. The sudden parameter changes during the initiation phase can easily cause patient discomfort, and the lack of differentiation between operating conditions leads to poor treatment adaptability and stability. This approach balances the safety and comfort of electrical stimulation therapy with the precise effectiveness of dynamic correction of pathological tremors, significantly improving patient adaptability and comfort in the initial treatment phase. Furthermore, the strategy of introducing real-time tremor correction intervention in non-initial initiation states dynamically adapts the stimulation intensity based on the patient's real-time fluctuating tremor dominance frequency and amplitude. This solves the technical problems of traditional open-loop fixed-parameter therapy, which cannot adapt to dynamic tremor changes, has unstable tremor suppression effects, and poor adaptability, achieving continuous, precise, and dynamic closed-loop suppression therapy for pathological tremors. Thus, through refined tremor status assessment, personalized subtype matching, a smooth transition initiation strategy, and real-time closed-loop feedback control, the limitations of traditional open-loop fixed-parameter electrical stimulation therapy in terms of adaptability, intervention effect, and patient experience are overcome, achieving a more intelligent, efficient, and comfortable intervention for essential tremor.

[0049] Based on the above Figure 1 In one example embodiment of the method shown, step 103 involves performing real-time tremor correction intervention on a patient with essential tremor. The specific process of this step can be achieved through… Figure 2 Steps 201 to 203 shown are implemented.

[0050] Step 201: Determine the tremor frequency band energy of patients with essential tremor within the current preset window; the tremor frequency band energy is the integral result of the energy or amplitude within the frequency band corresponding to the dominant tremor frequency within the current preset window.

[0051] Step 202: If the tremor frequency band energy is higher than the preset energy upper limit threshold, the stimulation amplitude in the first stimulation parameter is adjusted in a forward stepping manner, and the actual stimulation amplitude output to the idiopathic patient is determined based on the upper limit extreme value of the stimulation amplitude of the first tremor symptom subtype and the stimulation amplitude after forward stepping adjustment.

[0052] Step 203: If the tremor frequency band energy remains below the preset lower energy threshold for a first preset duration, the stimulation amplitude in the first stimulation parameter is adjusted in a negative stepping manner, and the actual stimulation amplitude output to the idiopathic patient is determined based on the lower limit extreme value of the stimulation amplitude of the first tremor symptom subtype and the stimulation amplitude after negative stepping adjustment.

[0053] Specifically, by integrating the energy or amplitude within the frequency band corresponding to the dominant tremor frequency within the current preset window, the tremor frequency band energy of the patient with essential tremor within the current preset window can be obtained. This tremor frequency band energy can be denoted as... This avoids the drawback of instantaneous amplitude being easily affected by random noise, and obtains the most realistic and stable pathological tremor intensity.

[0054] At this point, determine the energy of the tremor frequency band. With preset energy upper limit threshold The size relationship between them, if > If the tremor is not completely suppressed, the stimulation amplitude in the first stimulation parameter is further adjusted in a positive stepping manner. That is, the stimulation amplitude update value to be increased is calculated based on the adaptive algorithm (such as amplitude step +0.1mA). Each time, a fixed stimulation amplitude update value is added, and the stimulation amplitude in the first stimulation parameter is adjusted in a positive stepping manner.

[0055] For example, the updated stimulus amplitude value for each increase is denoted as . It can be calculated using equations (1) to (2).

[0056] (1) (2) In equations (1) to (2), This represents the update value of the stimulus amplitude at the current time t, and its initial value. Pre-defined and known; This represents the difference between the actual tremor amplitude detected at the current time t and the target tremor suppression threshold. This represents the difference between the actual tremor amplitude detected at the previous moment and the target tremor suppression threshold. Represents the proportional gain coefficient. Represents the differential gain coefficient. This represents the actual flutter amplitude detected at the current time t. This indicates the preset target tremor suppression threshold. Both the current time and the previous time belong to the current preset window.

[0057] To prevent a sudden increase in current due to sudden violent movements by the patient (causing a surge in e(t)), before actual delivery, it is necessary to force the current to be clamped under the dynamic boundary constraints corresponding to the first tremor symptom subtype category. The safe boundary amplitude after clamping is then used as the actual stimulus amplitude output to the essential tremor patient.

[0058] Conversely, by comparing the energy of the tremor frequency bands With preset energy upper limit threshold Determine the size relationship between them. ≤ If the duration of the state exceeds or falls below a preset duration (e.g., 3 seconds), it is determined that the tremor has been significantly relieved. At this time, in order to delay neural adaptation and reduce system power consumption, the MCU adjusts the stimulation amplitude in the first stimulation parameter in a negative step manner, that is, it calculates the stimulation amplitude update value to be reduced based on the adaptive algorithm (e.g., amplitude step -0.1mA), and reduces it by a fixed stimulation amplitude update value each time, thereby adjusting the stimulation amplitude in the first stimulation parameter in a negative step manner. The calculation process involved is referred to in equations (1) to (2).

[0059] Similarly, dynamic boundary constraints are applied to the calculated proposed reduction in stimulus amplitude update value each time. That is, if the calculated proposed reduction in stimulus amplitude update value exceeds the lower limit of stimulus amplitude for the first tremor symptom subtype (or the maximum tolerance threshold set by the patient), the actual issued parameter is locked at the lower limit of stimulus amplitude. If the calculated proposed reduction in stimulus amplitude update value does not exceed the lower limit of stimulus amplitude for the first tremor symptom subtype, the degradation step is performed according to the calculated stimulus amplitude update value.

[0060] The tremor stimulation closed-loop control method provided in this application introduces tremor frequency band energy as a more comprehensive indicator of tremor severity. Combined with dynamic step adjustment and symptom subtype extreme value limits, this allows electrical stimulation intervention to precisely adapt to the real-time changes in the patient's tremor state. This not only avoids muscle fatigue and discomfort caused by overstimulation in mild tremors but also solves the problems of insufficient stimulation intensity and poor tremor suppression in severe tremors. Furthermore, considering the physiological tolerance and treatment needs of patients with different tremor symptom subtypes, the setting of upper and lower amplitude extreme values ​​enables more personalized and safer treatment, significantly improving the adaptability, effectiveness, and patient experience of the treatment.

[0061] In one example embodiment, step 202 determines the actual stimulus amplitude output to the idiopathic patient based on the upper limit of the stimulus amplitude for the first tremor symptom subtype and the positively stepped-adjusted stimulus amplitude. The specific process in this embodiment can be achieved through… Figure 3 Steps 301 and 302 shown are implemented.

[0062] Step 301: If the stimulus amplitude after positive step adjustment exceeds the upper limit of the stimulus amplitude for the first tremor symptom subtype, then the upper limit of the stimulus amplitude is determined to be the actual stimulus amplitude.

[0063] Step 302: If the stimulus amplitude after positive step adjustment does not exceed the upper limit of the stimulus amplitude of the first tremor symptom subtype, then the stimulus amplitude after positive step adjustment is determined to be the actual stimulus amplitude.

[0064] It should be noted that the stimulus amplitude after positive step adjustment is specifically the updated stimulus amplitude value for each increase calculated by equations (1) to (2). .

[0065] The MCU makes a judgment every preset period of time (e.g., 500ms). As long as the tremor does not subside, it will calculate the proposed increase in stimulus amplitude update value and attempt to increase the step size. However, if the calculated stimulus amplitude update value is greater than or equal to the upper limit of the stimulus amplitude for the first tremor symptom subtype, the upper limit of the stimulus amplitude will be output to the patient and no further increase will be made. Conversely, if the calculated stimulus amplitude update value is less than the upper limit of the stimulus amplitude for the first tremor symptom subtype, the calculated stimulus amplitude update value will be output to the patient until the tremor is relieved.

[0066] For example, update the value of each calculated stimulus amplitude. The restrictions are set within the safety boundaries of the first tremor symptom subtype category, which is specifically accomplished by equation (3).

[0067] (3) In equation (3), Indicates the stimulus amplitude update value Safety boundary amplitude after clamping judgment processing Indicates the subtype of the first tremor symptom. The effective lower limit of therapeutic efficacy for fine-tuning the stimulation amplitude (current) is... Indicates the subtype of the first tremor symptom. The maximum safe upper limit of the stimulation amplitude (current) fine-tuning is defined by max, which indicates the operation of taking the maximum value and min, which indicates the operation of taking the minimum value.

[0068] In other words, the safety boundary amplitude obtained after clamping judgment processing by equation (3) The actual stimulus amplitude can be output to patients with essential tremor.

[0069] The tremor stimulation closed-loop control method provided in this application introduces a refined safety guarantee and effect optimization mechanism into the positive step adjustment of stimulation amplitude. This ensures that when dynamically adjusting the stimulation amplitude to adapt to changes in tremor intensity, the individual differences and safety tolerance of patients are always taken into account. This not only effectively suppresses tremor but also significantly improves the safety, comfort, and patient compliance of treatment, thereby optimizing the patient's treatment experience while ensuring treatment effectiveness.

[0070] In one example embodiment, step 103 involves performing gradual electrical stimulation intervention on patients with essential tremor based on the first stimulation parameters corresponding to the first tremor symptom subtype. The specific process in this embodiment can be achieved through… Figure 4 Steps 401 to 403 shown are implemented.

[0071] Step 401: Determine the step stimulation amplitude output to the essential tremor patient at the current step time based on the preset step time interval and the first adaptive ramp rate associated with the first tremor symptom subtype category; the current step time is the sum of the current time and m preset step time intervals; the initial value of m is 1, and the current time belongs to the current preset window.

[0072] Step 402: Determine the actual step stimulation amplitude output to the idiopathic patient based on the upper limit of the stimulation amplitude and the step stimulation amplitude of the first tremor symptom subtype category.

[0073] Step 403: If the actual step stimulation amplitude does not rise to the first reference stimulation amplitude in the first stimulation parameter and the patient's tremor symptoms are not eliminated, then increment the value of m by 1 and repeat the above steps.

[0074] The first baseline stimulus amplitude is the lower limit of the stimulus amplitude or the lower limit of the stimulus intensity corresponding to the first tremor symptom subtype category.

[0075] It should be noted that, for the initial startup state, an adaptive gradual rise process can be executed upon the first trigger. That is, when the MCU first recognizes the tremor from the resting state and completes subtype matching (such as the first tremor symptom subtype category)... After that, instead of immediately sending the amplitude reference parameter from the first stimulus parameter to the patient, the quantitative steps of calculating the step slope, executing the step output, and real-time feedback verification are executed sequentially.

[0076] To calculate the step slope, you can call the category corresponding to the first tremor symptom subtype. Associated climbing time constant For example, for highly sensitive individuals, = For strong earthquakes, = For step-by-step output, the current amplitude can be gradually increased according to equation (4) based on timer interrupts (e.g., every 100ms interval) to obtain the step-by-step stimulation amplitude output to the patient with essential tremor at the current step-by-step time. .

[0077] (4) In equation (4), This represents the update value of the stimulus amplitude at the current time t, and its initial value. Pre-defined and known; This indicates the preset step time interval, such as 100ms.

[0078] For real-time feedback verification, during the gradual increase, it can be monitored whether the step stimulation amplitude calculated by formula (4) rises to the first reference stimulation amplitude in the first stimulation parameter. If the patient's tremor symptoms have disappeared before the step stimulation amplitude rises to the first reference stimulation amplitude, the climbing can be stopped in advance and locked at the current effective intensity. Conversely, if the patient's tremor symptoms have not disappeared before the step stimulation amplitude has risen to the first reference stimulation amplitude, the value of m is increased by 1, the step stimulation amplitude is recalculated according to formula (4), and feedback verification is performed again until the climbing is stopped.

[0079] The tremor stimulation closed-loop control method provided in this application calculates the stimulation amplitude at each moment based on a fixed step time interval and an adaptive ramp rate corresponding to the tremor subtype. At the same time, it uses the upper limit of the amplitude to constrain the output range, preventing the stimulation intensity from exceeding the limit. It continuously iterates and adjusts the stimulation intensity while the stimulation amplitude has not fallen back to the baseline level and the tremor symptoms have not been eliminated. This allows the electrical stimulation intensity to decrease gradually, enabling the human nerves and muscles to gradually adapt to the stimulation effect. It avoids the discomfort caused by sudden changes in intensity, thereby achieving an organic unity of stability, safety and effectiveness in the initial gradual intervention treatment.

[0080] In one example embodiment, considering that the tremor symptoms of patients with essential tremor are prone to change with their movements, emotions, and other states during the treatment process, and the tremor symptom subtype may also be updated accordingly, if there is no logic for adjusting the stimulation parameters after the subtype is updated, and the stimulation parameters corresponding to the original tremor symptom subtype are still used, it will be impossible to adapt to the patient's changed tremor symptoms, and the problem of inappropriate stimulation intensity may occur. This will not only reduce the effect of tremor suppression treatment, but may also cause unnecessary discomfort to the patient. Therefore, this application solves the problem of stimulation adaptation adjustment after the tremor symptom subtype is updated by the following steps in the embodiment.

[0081] If the first tremor symptom subtype category in the current preset window is updated to the second tremor symptom subtype category, and the patient's tremor symptoms have not been eliminated, the current output stimulation parameters are controlled to gradually decrease to the preset safety benchmark level or zero at the preset downward slope. Then, based on the second benchmark stimulation amplitude and the second adaptive ramp rate corresponding to the second tremor symptom subtype category, a new gradual electrical stimulation intervention is performed on the essential tremor patient.

[0082] Specifically, within the current preset window, the tremor symptom subtype category can be determined at regular intervals. Thus, when a sudden change in patient movement is detected (e.g., from rest to raising a glass), the first tremor symptom subtype category can be switched to the most recently updated second tremor symptom subtype category. To avoid startle responses caused by parameter abrupt changes, the following process of decreasing followed by increasing or cross-fading is performed: controlled gradual decrease of the old parameter (i.e., the first stimulus parameter corresponding to the first tremor symptom subtype category), adaptive gradual increase of the new parameter, and transitional masking.

[0083] The old parameters are controlled to decrease gradually, that is, initially with a relatively high downward slope. (like The stimulation intensity (i.e., stimulation amplitude) of the first tremor symptom subtype is reduced to a preset safety baseline level or zero.

[0084] The new parameters are adaptively and gradually increased. That is, after confirming that the old stimulation intensity (the stimulation intensity corresponding to the first tremor symptom subtype) has been safely withdrawn, the real-time tremor correction intervention process is frozen. Then, the gradual increase procedure for the second tremor symptom subtype is immediately started. Using the second baseline stimulation amplitude and the second adaptive ramp rate corresponding to the second tremor symptom subtype, a new gradual electrical stimulation intervention is performed on patients with essential tremor until the parameters are pushed up to the new anchor point.

[0085] Transitional state shielding means that during the entire parameter transition period (usually 1-2 seconds), the tremor real-time correction intervention treatment algorithm (second-stage closed-loop fine-tuning) is temporarily frozen, and the second-stage closed-loop fine-tuning is restarted after the new parameters stabilize.

[0086] The tremor stimulation closed-loop control method provided in this application introduces a tremor symptom subtype update detection mechanism and combines it with a logic for smooth transition of stimulation parameters. This enables the system to switch to a more suitable stimulation strategy in a timely and safe manner when the patient's tremor symptom subtype changes. Specifically, after confirming the subtype update, the original stimulation parameters are reduced to a safe baseline threshold to avoid discomfort caused by sudden changes in stimulation intensity. Subsequently, based on the new tremor symptom subtype category, a new gradual electrical stimulation intervention is performed using its corresponding baseline stimulation amplitude and adaptive ramp rate. This ensures that the stimulation intensity accurately matches the patient's current updated tremor symptom needs, significantly improving the suitability and effectiveness of tremor suppression therapy. Furthermore, the smooth stimulation switching process greatly improves the patient's treatment experience and reduces adverse reactions such as muscle spasms and surface tingling, thereby achieving a dual optimization of treatment effect and patient comfort.

[0087] In one example embodiment, step 103 involves determining the first tremor symptom subtype category to which the essential tremor patient currently belongs. The specific process for this in this embodiment can be achieved through… Figure 5 Steps 501 and 502 shown are implemented.

[0088] Step 501: Obtain the multidimensional physiological characteristics of limb tremor in patients with essential tremor within the current preset window, and perform feature standardization and dimensionality reduction on the multidimensional physiological characteristics of limb tremor.

[0089] Step 502: Based on the matching relationship between the multidimensional physiological characteristics of limb tremor after dimensionality reduction and various tremor symptom types, determine the first tremor symptom subtype category to which the patient with essential tremor currently belongs.

[0090] Among them, the multidimensional physiological characteristics of limb tremor include multiaxial inertial motion characteristics and optional multidimensional physiological characteristics of limb tremor.

[0091] For example, multi-axis inertial motion characteristics include tremor dominant frequency, tremor power spectral density / tremor amplitude, harmonic energy ratio, and attitude angle change rate; the multidimensional physiological characteristics of limb tremor are electromyographic envelope symmetry.

[0092] The preset types of tremor symptoms are clustered into N essential tremor subtypes.

[0093] It should be noted that before the system detects a tremor and triggers a stimulus, it first captures a very short time window (e.g., 1-2 seconds) of multi-axis sensor data for initial screening. Specifically, within the current preset window, a raw feature vector is constructed by extracting k time-frequency features from the multi-axis sensing data acquired from the multi-axis inertial sensor and the electromyography signal. , Original feature vector The formula is shown in equation (5).

[0094] (5) In formula (5), represents the main tremor frequency,[[]] represents the tremor power spectral density / tremor amplitude,[[]] represents the proportion of harmonic energy,[[]] represents the rate of change of the attitude angle,[[]] represents the electromyogram envelope amplitude,[[]] T represents the transpose operation.[[]]

[0095] To eliminate the influence of different dimensions, for the original feature vector shown in formula (5)[[]] feature standardization (such as Z-score standardization) is performed to obtain the standardized high-dimensional feature vector[[]] ; where the original feature vector[[]] in the i th original feature[[]] The process of feature standardization is shown in formula (6).[[]]

[0096] (6)[[]] In formula (6), represents the feature obtained after performing feature standardization on the i th original feature[[]] , represents the mean of the pre-stored large amount of clinical sample feature data,[[]] represents the standard deviation of the pre-stored large amount of clinical sample feature data.[[]]

[0097] Subsequently, using the pre-trained offline principal component projection matrix[[]] (usually d < k, and d = 2 or 3 is usually taken to adapt to edge computing), the high-dimensional feature vector[[]] is mapped to a low-dimensional feature space to obtain the principal component feature vector after dimensionality reduction[[]] , and the principal component feature vector after dimensionality reduction[[]] is the multi-dimensional physiological feature of limb tremor after dimensionality reduction processing; its dimensionality reduction process is shown in formula (7).[[]]

[0098]

[0099] Finally, subtype matching is performed, that is, calculating the Euclidean distance (or Mahalanobis distance) between the current principal component feature vector after dimensionality reduction[[]] and the clustering center of the L-type idiopathic tremor subtype, where the current principal component feature vector after dimensionality reduction[[]] and the b [[]]th clustering center of the idiopathic tremor subtype[[]] The Euclidean distance between them is denoted as[[]] and its calculation formula is shown in formula (8).[[]]

[0100] (8) In equation (8), , Represents the principal component eigenvector The Middle j The principal component features, that is, the current principal component feature vector after dimensionality reduction. In the j The projected coordinate values ​​(i.e., principal component scores) on each principal component axis have the actual physical meaning of the comprehensive feature components in the dimension-reduced direction after orthogonal transformation of the current patient's multidimensional initial tremor features. Indicates the first b Cluster center of essential tremor subtype In the j The standard reference coordinate values ​​on the principal component axes, that is, the pre-trained and calibrated first principal component axis, represent the first reference coordinate values. Cluster center of essential tremor subtype In the j Standard reference coordinate values ​​on each principal component axis; This represents the total number of principal component feature dimensions after dimensionality reduction (e.g., ...). ).

[0101] The cluster center corresponding to the shortest distance among the calculated L Euclidean distances is selected as the first tremor symptom subtype to which the essential tremor patient currently belongs. The process is shown in equation (9).

[0102] (9) In equation (9), argmin represents the independent variable that minimizes the value of the function.

[0103] It should be noted that, due to the principal component linear combination, the standard reference coordinate value represents the typical position of the subtype in the dimension-reduced feature space (latent space), rather than a single original physical quantity (such as absolute frequency or absolute amplitude).

[0104] The tremor stimulation closed-loop control method provided in this application acquires multidimensional physiological characteristics of limb tremor in patients with essential tremor, directly quantifying the physiological state of tremor. Through feature standardization and dimensionality reduction, it effectively eliminates dimensional differences between features, removes redundant information, and significantly improves the efficiency and accuracy of classification. Finally, based on the matching relationship between the processed high-quality features and various tremor symptom types, it can accurately and quickly determine the current tremor symptom subtype category of the patient with essential tremor. This accurate subtype classification provides a solid foundation for the subsequent closed-loop control system to match corresponding stimulation parameters according to different subtypes, enabling the stimulation parameters to more accurately adapt to the patient's current actual tremor condition, thereby significantly improving the accuracy of closed-loop regulation, intervention effect, and patient treatment experience.

[0105] In one example embodiment, considering that the dominant tremor frequency is not consistently within the preset pathological frequency band and / or the tremor amplitude does not consistently exceed the non-pathological physiological baseline threshold within the current preset window, if there is a lack of further judgment logic on the patient's actual limb state and no corresponding step to adjust the stimulation parameters, it will be impossible to adjust the stimulation intensity in a timely manner when the patient has actually entered a state of limb rest without obvious tremor. This can easily lead to unnecessary overstimulation, causing muscle fatigue, increasing ineffective energy consumption, reducing treatment comfort, and failing to adapt to the resting state after the tremor disappears, thereby affecting the treatment experience. Therefore, this embodiment addresses this issue by... Figure 6 Steps 601 and 602 shown resolve this issue.

[0106] Step 601: If the dominant tremor frequency does not remain within the preset pathological frequency band and / or the tremor amplitude does not exceed the non-pathological physiological baseline threshold within the current preset window, then determine the variance of the composite acceleration signal and the variance of the composite angular velocity signal of the multi-axis acceleration signal within the current preset window.

[0107] Step 602: If the variance of the synthesized acceleration signal is consistently lower than the preset acceleration resting threshold and the variance of the synthesized angular velocity signal is consistently lower than the preset angular velocity resting threshold within the preset resting time, then the patient with essential tremor is determined to be in a limb resting state and enters a low-power standby mode or a low-intensity maintenance stimulation state.

[0108] Among them, the multi-axis acceleration signal is the dynamic signal component of multi-axis acceleration within the current preset window, such as a 3-axis acceleration signal, and each axis acceleration signal can provide a high contribution to tremor capture. The multi-axis angular velocity signal is the dynamic signal component of multi-axis angular velocity within the same current preset window, such as a 2-axis angular velocity signal, and each axis angular velocity signal can also provide a high contribution to tremor capture.

[0109] The preset rest period is the preset stabilization duration, for example, 3 seconds.

[0110] It should be noted that when the MCU determines that the dominant tremor frequency within the current preset window is not continuously within the preset pathological frequency band and / or the tremor amplitude is not continuously exceeding the non-pathological physiological baseline threshold, it does not rely on the dominant tremor frequency and amplitude at a single instant. Instead, it uses the raw signals from the multi-axis inertial sensors continuously acquired within the current preset window to quantitatively determine the resting / non-tremor state. Specifically, firstly, eigenvalue calculation is performed; that is, the MCU extracts the dynamic signal components of the 3-axis acceleration within the current preset window and calculates the variance of the composite acceleration signal of the three-dimensional composite acceleration. Simultaneously, dynamic signal components of the two-axis angular velocities within the same current preset window are acquired, and the variance of the composite angular velocity signal is calculated. These two signal variances represent the absolute intensity of the patient's hand movements.

[0111] Then, a double threshold comparison is performed, that is, the variance of the synthesized acceleration signal is compared. Compared with the preset resting acceleration threshold The variance of the synthesized angular velocity signal is compared. Compared with the preset resting threshold of angular velocity Compare; acceleration resting threshold and angular velocity resting threshold This is to define the calibrated environmental background noise and the upper limit of physiological micro-vibration. And... At this point, time jitter stabilization is confirmed based on the comparison results; that is, if and only if ≤ and ≤ Furthermore, only when this dual below-threshold state continues for more than the preset anti-shake maintenance time (e.g., 3 seconds) can the patient's target limb be finally determined to be in a true "resting state" or the patient with essential tremor be finally determined to be in a limb resting state. At this time, the system will either enter low-power standby or smoothly switch to the minimum maintenance stimulus (Sub-threshold).

[0112] For example, for high-frequency, low-amplitude tremors with high nerve sensitivity, a gentler incline (e.g., a step size limited to 0.05 mA / second) is used to ensure the patient's nerve fibers adapt and avoid muscle jerking caused by sudden high current. Conversely, for low-frequency, high-amplitude, severe tremors, a steeper incline (e.g., 0.2 mA / second) is used to maximize onset speed while ensuring safety. The stimulation pulse generator has a time constant... Under the constraints, the output amplitude or pulse width register is updated gradually through microsecond-level timer interrupts until it smoothly approaches or exits the target parameter space.

[0113] The tremor stimulation closed-loop control method provided in this application, when the dominant tremor frequency of an essential tremor patient is not continuously within the preset pathological frequency band and / or the tremor amplitude is not continuously exceeding the non-pathological physiological baseline threshold, the system no longer simply maintains or abruptly stops stimulation. Instead, it determines the variance of the synthesized acceleration signal and the variance of the synthesized angular velocity signal of the multi-axis acceleration signal, and combines this with continuous monitoring within a preset resting judgment period to accurately determine whether the patient is truly in a limb resting state. This continuous judgment mechanism based on multi-dimensional inertial signal variance significantly improves the accuracy and reliability of limb state judgment and effectively avoids errors caused by accidental activity or single... Overstimulation caused by signal misinterpretation; in addition, when the patient is confirmed to be in a resting state, the system will smoothly reduce the stimulation parameters instead of abruptly stopping the stimulation. This greatly improves the patient's treatment comfort, avoids discomfort caused by sudden interruption of stimulation, and avoids continuing to apply unnecessary stimulation when the patient's limbs are at rest, reducing muscle fatigue and ineffective energy consumption. This improves the logic of closed-loop control, allowing the stimulation intensity to be dynamically adapted to the patient's actual limb tremor and movement state throughout the process. Especially in the resting state after tremor subsidence, it provides a more refined and humane intervention strategy, significantly improving the adaptability, comfort and effectiveness of treatment.

[0114] In one example embodiment, step 101 involves obtaining the tremor dominant frequency and tremor amplitude of a patient with essential tremor within the current preset window. The specific process in this embodiment can be achieved through… Figure 7 Steps 701 to 704 shown are implemented.

[0115] Step 701: When a patient with essential tremor wears a tremor stimulation wearable device and the device base plate integrates an M-axis inertial sensor, determine the device pitch angle and device roll angle, and determine the projection vector of the tremor plane in the local coordinate system of the sensor based on the device pitch angle and device roll angle.

[0116] Step 702: Analyze the contribution of the M-axis inertial sensor to flutter capture based on the projection vector, and select the candidate redundant axis with the lowest contribution to flutter capture.

[0117] Step 703: Determine the variance of the angular velocity signal of the candidate redundant axis within the preset observation window. If the variance of the angular velocity signal is continuously lower than the preset minimum threshold within the preset observation window, the candidate redundant axis is masked to obtain the N-axis inertial sensor for feature extraction. The N-axis sensor is the remaining inertial sensor after removing the candidate redundant axis from the M-axis inertial sensor. M > N, and both M and N are positive integers greater than 0.

[0118] Step 704: Periodically collect inertial motion data from the N-axis inertial sensor within the current preset window, perform spectrum analysis on the inertial motion data, and extract the dominant tremor frequency and tremor amplitude of the patient with essential tremor within the current preset window.

[0119] Among them, patients with essential tremor can wear tremor stimulation wearable devices by wearing them on their left hand, right hand, or both hands simultaneously.

[0120] Candidate redundant axes are sensing axes that contribute very little to the effective capture of tremor motion and have very weak signal fluctuations. The number of candidate redundant axes is usually 1.

[0121] The variance of angular velocity signals can be used to characterize the intensity of fluctuations in single-axis motion signals; the smaller the value, the lower the motion activity.

[0122] It should be noted that, considering that tremor stimulation wearable devices typically integrate multi-axis inertial sensors to collect motion data, the contribution of motion signal data collected by different axis sensors to tremor capture varies. If inertial motion data from all axes is used directly to extract the tremor dominant frequency and tremor amplitude, it will not only generate a large amount of redundant data and increase the computational burden on the device, but also reduce the accuracy of tremor feature extraction due to the interference of redundant data, thereby affecting the precision of subsequent stimulation control and failing to provide reliable basic data for tremor stimulation closed-loop control.

[0123] To address these issues, this embodiment employs sensor input and dynamic dimensionality reduction determination, posture calculation, and redundant axis determination and shielding. Specifically, hand movement data is acquired at high frequencies (e.g., >60Hz) using an M-axis inertial sensor (such as a 3-axis accelerometer and a 3-axis angular velocity sensor) integrated into the base plate of a tremor stimulation wearable device (such as a wearable wristband) worn by patients with essential tremor. To reduce the power consumption of edge computing, the system performs an axial dynamic shielding process based on rotation angle upon startup or when the patient's arm posture changes significantly.

[0124] For attitude calculation, the MCU first extracts the static gravity component of the 3-axis accelerometer and calculates the current rotation angle of the wearable wristband in three-dimensional space (including pitch and roll angles), thereby anchoring the real-time spatial attitude of the patient's wrist.

[0125] For example, the specific calculation process of the rotation angle matrix includes: processing the acquired 3-axis acceleration signals... Low-pass filtering (LPF) is performed to filter out high-frequency vibration interference and extract the static gravity vector component.

[0126] Let the gravitational acceleration when the wearable wristband's coordinate system coincides with the geographic coordinate system be... Then, the real-time attitude angle is calculated using equations (10) to (11), which is also the calculation of the wearable wristband's pitch angle. and the roll angle of wearable wristbands .

[0127] (10) (11) In equations (10) to (11), arctan represents the operation of obtaining the arctangent function value.

[0128] Wearable wristband tilt angle Represents the rotation angle of the wearable wristband around the Y-axis, i.e., the angle between the X-axis and the horizontal plane; the roll angle of the wearable wristband. This represents the rotation angle of the wearable hand around the X-axis, which is the angle between the Y-axis and the horizontal plane.

[0129] For redundant axis determination, the pitch angle calculated by combining equations (10) to (11) is used. and roll angle The algorithm calculates the projection of the primary motion plane of essential tremor (such as the flexion-extension or pronation-supination plane) onto the local coordinate system of the sensor, and uses an algorithm to determine the specific single gyroscope (i.e., the candidate redundant axis with extremely low contribution to tremor capture) whose projection tends to be perpendicular to the primary motion plane of essential tremor or whose angular velocity signal energy variance is continuously lower than a preset minimum threshold within a set observation window.

[0130] For example, the redundant axis determination process includes: based on the wearable wristband pitch angle calculated by equations (10) to (11). and the roll angle of wearable wristbands Construct a rotation transformation matrix from the human anatomical reference coordinate system (Global / Anatomical Frame) to the wearable wristband sensor local coordinate system (Sensor Frame). The specific matrix is ​​shown in equation (12).

[0131] (12) In equation (12), cos represents the operation of obtaining the cosine function value, and sin represents the operation of obtaining the sine function value.

[0132] Considering that the main pathological wrist movements in essential tremor are typically flexion-extension or pronation-supination, a rotation transformation matrix is ​​defined. The main movement pattern, in the anatomical reference coordinate system, has a typical tremor rotation axis unit vector as follows: (For example, the rotation axis of simple flexion-extension tremor is often perpendicular to the sagittal plane of the forearm). This can be achieved through a rotational transformation matrix. Mapping it to the local coordinate system of the wearable wristband sensor yields the projection vector. The specific vector is referenced in equation (13). This indicates that... (13) In equation (13), This represents the theoretical weighting of the tremor angular velocity along the X-axis of the local coordinate system of the wearable wristband sensor. This represents the theoretical weighting of the vibration angular velocity along the Y-axis of the wearable wristband sensor's local coordinate system. This represents the theoretical weighting of the tremor angular velocity on the Z-axis of the wearable wristband sensor's local coordinate system. T This indicates the transpose operation.

[0133] By taking the absolute value of the weights assigned to the three theories mentioned above and finding the minimum value, the specific single gyroscope axis with the lowest contribution to tremor detection was determined. The determination process is shown in equation (14). The single gyroscope axis determined here... This is an N-axis redundant inertial sensor, where N is 1.

[0134] (14) In equation (14), argmin represents the independent variable that minimizes the value of the function.

[0135] To prevent theoretical projection errors caused by abnormal patient posture, the MCU simultaneously opens a set time observation window (e.g., 2 seconds) to calculate candidate redundant axes. The variance of the actual angular velocity signal within the preset time observation window If and only if the variance of the actual angular velocity signal... If the value remains below the preset minimum threshold within the preset time window, the candidate redundant axis is finally confirmed and blocked.

[0136] The process then proceeds to dimensionality reduction and data acquisition. In the subsequent closed-loop fine-tuning phase, the MCU actively goes into sleep mode or disables the candidate redundant axis. The data acquisition uses only the remaining "3-axis acceleration + 2-axis angular velocity" to form the 5-axis high-frequency feature extraction axis input closed-loop algorithm. Five-axis inertial motion data is periodically acquired according to the current preset window, filtered, and feature vectors are calculated. These feature vectors include the dominant tremor frequency (typically in the 3-12Hz range), tremor power spectral density / tremor amplitude, harmonic energy ratio, attitude angle change rate, and optional electromyographic envelope symmetry. The dominant tremor frequency and tremor amplitude are then further extracted from this feature vector.

[0137] The tremor stimulation closed-loop control method provided in this application first calculates the tremor motion plane projection vector based on the device attitude angle, evaluates the contribution of each sensor axis signal, filters and shields redundant axes with weak fluctuations, and simplifies the effective sensor acquisition channels; then, it acquires motion data based on the retained sensor axes and performs spectrum analysis to extract tremor features, which not only reduces invalid data interference and computational load, but also effectively improves the accuracy of tremor main frequency and amplitude parameter extraction.

[0138] In one example embodiment, considering that the stimulating electrodes may experience short circuits or detachment during treatment, without a real-time monitoring and emergency response mechanism for these abnormal electrode conditions, continued electrical stimulation under abnormal conditions would not only fail to achieve the expected tremor suppression effect but also pose unnecessary safety risks to the patient. Therefore, this embodiment addresses this issue by... Figure 8 Steps 801 and 802 shown resolve this issue.

[0139] Step 801: During the implementation of gradual electrical stimulation intervention or real-time tremor correction intervention, the impedance of the stimulation electrode is collected in real time using a preset detection current.

[0140] Step 802: If the detected impedance of the stimulation electrode is lower than the preset short-circuit threshold or higher than the preset detachment abnormal threshold, the electrical stimulation treatment will be forcibly terminated through a preset hardware interruption method.

[0141] When the detected electrode impedance is lower than a preset short-circuit threshold, it usually indicates that a short circuit has occurred between the electrodes, or that the contact area between the electrode and the patient's skin is too large and the contact resistance is abnormally low. This preset short-circuit threshold can be set according to the electrode material, skin characteristics, stimulation parameters, and clinical experience, for example, it can be set to tens of ohms.

[0142] When the detected electrode impedance is higher than a preset abnormal detachment threshold, it usually indicates poor contact between the electrode and the patient's skin, partial detachment, or complete detachment. This preset abnormal detachment threshold can be set based on the electrode material, skin characteristics, stimulation parameters, and clinical experience; for example, it can be set to several thousand ohms.

[0143] Specifically, by using a preset detection current to collect the impedance of the stimulation electrode in real time, quantitative data on the contact state between the stimulation electrode and the skin can be continuously obtained. Subsequently, the detected stimulation electrode impedance is compared with a preset short-circuit threshold and a preset detachment abnormality threshold. Once the detected stimulation electrode impedance is lower than the preset short-circuit threshold (indicating a short circuit) or higher than the preset detachment abnormality threshold (indicating detachment), it will be immediately judged as an electrode abnormality.

[0144] Considering that abnormal stimulation electrodes may lead to poor stimulation effects or even harm to the patient, the MCU will forcibly terminate the electrical stimulation therapy through a preset hardware interrupt. This can be achieved through a dedicated hardware circuit, such as a comparator circuit, that monitors the impedance of the stimulation electrodes in real time. Once the interrupt condition is met, an interrupt signal is immediately triggered, which directly controls the switch of the stimulation output circuit, causing it to disconnect. Alternatively, the MCU can receive abnormal signals from the impedance monitoring module through an external interrupt pin. Upon receiving an abnormal signal, the MCU immediately executes a preset interrupt service routine, which directly shuts down the stimulation output channel. This minimizes the safety risks to the patient caused by abnormal states and ensures that the patient's treatment safety is effectively monitored and guaranteed throughout the entire closed-loop vibration suppression therapy process, regardless of how the stimulation parameters are dynamically adjusted. This allows advanced closed-loop control methods to be applied more safely and reliably in clinical practice.

[0145] The tremor stimulation closed-loop control method provided in this application introduces a real-time electrode abnormality safety guarantee mechanism during the electrical stimulation treatment process for idiopathic tremor. It can promptly detect abnormal states such as short circuits or detachment of the stimulation electrodes and forcibly terminate the electrical stimulation treatment with a rapid hardware interruption response. This effectively avoids the safety risks to patients caused by abnormal electrical stimulation. Especially in closed-loop treatment scenarios where stimulation parameters are dynamically adjusted, such as gradual electrical stimulation intervention and real-time tremor correction intervention, this solution ensures comprehensive safety monitoring at all electrical stimulation output stages, improves the safety and reliability of the entire treatment process, and allows patients to receive efficient tremor suppression treatment in a safer environment.

[0146] Through the aforementioned method embodiments, the tremor stimulation closed-loop control method provided in this application has the following technical effects: (1) Two-stage rapid response: Based on onboard IMU data, an innovative two-stage strategy of "subtype classification to set the baseline + limited fine-tuning" is introduced to break the limitations of traditional closed-loop step-by-step trial and error, realize millisecond-level tremor feature recognition and parameter response, greatly shorten the time to find the optimal parameters, realize the smooth adjustment of stimulation parameters, avoid the discomfort caused by drastic current adjustment, and prevent abnormal actions from causing algorithm divergence.

[0147] (2) Precise control: By extracting the features of tremor amplitude and frequency, the stimulation intensity (amplitude / pulse width) is automatically adjusted to achieve "on-demand stimulation", extend battery life and reduce neural adaptation. Hardware collaboration and low power consumption precise control: Using a specially trimmed 5-axis motion sensor (removing redundant axial data), combined with principal component analysis for dimensionality reduction, while achieving "on-demand stimulation", the edge computing overhead is significantly reduced, extending battery life and reducing neural adaptation.

[0148] (3) Safety and comfort: The introduction of a delayed rise and fall mechanism and impedance monitoring ensures safety during the closed-loop regulation process.

[0149] (4) Safe and comfortable dynamic boundary: introduce parameter fine-tuning boundary limits based on different tremor subtypes, and combine delay rise and fall mechanism with impedance monitoring to ensure absolute safety and foolproof capability in the closed-loop regulation process.

[0150] The following is for reference. Figure 9 It shows a schematic diagram of the structure of a computer system 900 suitable for implementing computer devices or servers in the embodiments of this application.

[0151] like Figure 9 As shown, the computer system 900 includes a central processing unit (CPU) 901, which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) 902 or programs loaded from storage section 908 into random access memory (RAM) 903. The RAM 903 also stores various programs and data required for the operation of the system 900. The CPU 901, ROM 902, and RAM 903 are interconnected via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0152] The following components are connected to I / O interface 905: an input section 906 including a keyboard, mouse, etc.; an output section 907 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 908 including a hard disk, etc.; and a communication section 909 including a network interface card such as a LAN card, modem, etc. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to I / O interface 905 as needed. A removable medium 911, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 910 as needed so that computer programs read from it can be installed into storage section 908 as needed.

[0153] Specifically, according to embodiments of this application, the above references Figure 1 The described process can be implemented as a computer software program. For example, embodiments of this application include a computer program product comprising a computer program tangibly embodied on a machine-readable medium, the computer program containing instructions for performing... Figure 1 The program code for the method. In such an embodiment, the computer program can be downloaded and installed from a network via communication section 909, and / or installed from removable media 911.

[0154] It should be noted that the computer-readable medium shown in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0155] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0156] The units or modules described in the embodiments of this application can be implemented in software or hardware. The described units or modules can also be located in a processor. The names of these units or modules do not, in certain circumstances, constitute a limitation on the unit or module itself.

[0157] On the other hand, this application also provides a computer-readable storage medium, which may be included in the computer device described in the above embodiments, or may exist independently and not assembled into the computer device. The aforementioned computer-readable storage medium stores one or more programs that, when used by one or more processors, execute the methods described in this application. For example, it may execute... Figure 1 The steps of the method shown.

[0158] This application provides a computer program product including instructions that, when executed, cause the method described in this application to be performed. For example, it can execute... Figure 1 The steps of the method shown.

[0159] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0160] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A closed-loop control method for tremor stimulation, characterized in that, Applied to a tremor stimulation closed-loop control system; the method includes: Obtain the dominant frequency and amplitude of tremor in patients with essential tremor within the current preset window; If, within the current preset window, the dominant frequency of the tremor remains within the preset pathological frequency band of essential tremor and the amplitude of the tremor remains above the preset non-pathological physiological baseline threshold, then it is determined whether the system is currently in the initial startup state. If the system is currently in the initial startup state, the first tremor symptom subtype category to which the essential tremor patient currently belongs is determined. According to the first stimulation parameter corresponding to the first tremor symptom subtype category, the patient with essential tremor is subjected to progressive electrical stimulation intervention treatment until the actual output stimulation parameter increases to the benchmark anchoring parameter corresponding to the first tremor symptom subtype category. Then, the patient with essential tremor is subjected to real-time tremor correction intervention treatment. If the system is not currently in an initial startup state, then real-time tremor correction intervention is performed on the patient with essential tremor.

2. The method according to claim 1, characterized in that, The real-time tremor correction intervention for the patients with essential tremor includes: Determine the tremor frequency band energy of the patient with essential tremor within the current preset window; the tremor frequency band energy is the integral result of the energy or amplitude within the frequency band corresponding to the dominant tremor frequency within the current preset window; If the energy of the tremor frequency band is higher than the preset energy upper limit threshold, the stimulation amplitude in the first stimulation parameter is adjusted in a forward stepping manner, and the actual stimulation amplitude output to the idiopathic patient is determined based on the upper limit extreme value of the stimulation amplitude of the first tremor symptom subtype category and the stimulation amplitude after forward stepping adjustment. If the energy of the tremor frequency band remains below the preset lower energy threshold for a first preset duration, the stimulation amplitude in the first stimulation parameter is adjusted in a negative stepping manner, and the actual stimulation amplitude output to the idiopathic patient is determined based on the lower limit extreme value of the stimulation amplitude of the first tremor symptom subtype and the stimulation amplitude after negative stepping adjustment.

3. The method according to claim 2, characterized in that, The step of determining the actual stimulation amplitude output to the idiopathic patient based on the upper limit of the stimulation amplitude of the first tremor symptom subtype and the positively stepped stimulation amplitude includes: If the positively stepped adjustment of the stimulation amplitude exceeds the upper limit of the stimulation amplitude of the first tremor symptom subtype, then the upper limit of the stimulation amplitude is determined to be the actual stimulation amplitude. If the positively stepped stimulation amplitude does not exceed the upper limit of the stimulation amplitude for the first tremor symptom subtype, then the positively stepped stimulation amplitude is determined to be the actual stimulation amplitude.

4. The method according to claim 1, characterized in that, The step of performing gradual electrical stimulation intervention on the essential tremor patient according to the first stimulation parameter corresponding to the first tremor symptom subtype category includes: Based on the preset step time interval and the first adaptive ramp rate in the first stimulation parameters, the step stimulation amplitude output to the patient with essential tremor at the current step time is determined; the current step time is the sum of the current time and m preset step time intervals, the current time belongs to the current preset window, and the initial value of m is 1; Based on the upper limit of the stimulation amplitude in the first stimulation parameter and the step stimulation amplitude, the actual step stimulation amplitude output to the idiopathic patient is determined. If the actual step stimulation amplitude does not rise to the first reference stimulation amplitude in the first stimulation parameter and the patient's tremor symptoms are not eliminated, then the value of m is incremented by 1, and the above steps are repeated.

5. The method according to claim 1, characterized in that, The method further includes: If the first tremor symptom subtype category is updated to the second tremor symptom subtype category within the current preset window, and the patient's tremor symptoms have not been eliminated, then the current output stimulation parameters are controlled to gradually decrease to the preset safety benchmark level or zero with a preset downward slope. Then, according to the second benchmark stimulation amplitude and the second adaptive ramp rate corresponding to the second tremor symptom subtype category, a new gradual electrical stimulation intervention is performed on the essential tremor patient.

6. The method according to claim 1, characterized in that, The determination of the first tremor symptom subtype category currently belonging to the essential tremor patient includes: Obtain the multidimensional physiological characteristics of limb tremor of the patient with essential tremor within the current preset window, and perform feature standardization and dimensionality reduction on the multidimensional physiological characteristics of limb tremor. Based on the matching relationship between the multidimensional physiological characteristics of limb tremor after dimensionality reduction and various tremor symptom types, the first tremor symptom subtype category to which the patient with essential tremor currently belongs is determined.

7. The method according to claim 1, characterized in that, The method further includes: If, within the current preset window, the dominant frequency of the tremor is not continuously within the preset pathological frequency band and / or the amplitude of the tremor is not continuously exceeding the non-pathological physiological baseline threshold, then the variance of the composite acceleration signal and the variance of the composite angular velocity signal of the multi-axis acceleration signal within the current preset window are determined. If, within a preset resting time, the variance of the synthetic acceleration signal remains below a preset resting acceleration threshold and the variance of the synthetic angular velocity signal remains below a preset resting angular velocity threshold, then the patient with essential tremor is determined to be in a limb resting state and enters a low-power standby mode or a low-intensity maintenance stimulation state.

8. The method according to claim 1, characterized in that, The process of obtaining the dominant frequency and amplitude of tremor in patients with essential tremor within the current preset window includes: When the patient with essential tremor wears a tremor stimulation wearable device and the device base plate integrates an M-axis inertial sensor, the device pitch angle and device roll angle are determined, and the projection vector of the tremor plane in the local coordinate system of the sensor is determined based on the device pitch angle and the device roll angle. Based on the projection vector, the flutter capture contribution of the M-axis inertial sensor is analyzed, and the candidate redundant axis with the lowest flutter capture contribution is selected. The variance of the angular velocity signal of the candidate redundant axis within a preset observation window is determined. If the variance of the angular velocity signal is continuously lower than a preset minimum threshold within the preset observation window, the candidate redundant axis is masked to obtain an N-axis inertial sensor for feature extraction. The N-axis sensor is the remaining inertial sensor after removing the candidate redundant axis from the M-axis inertial sensor. M > N, and both M and N are positive integers greater than 0. The inertial motion data of the N-axis inertial sensor is periodically collected within the current preset window. The inertial motion data is subjected to spectrum analysis to extract the main frequency of the vibration and the vibration amplitude.

9. The method according to claim 1, characterized in that, The method further includes: During the execution of the gradual electrical stimulation intervention or the real-time tremor correction intervention, the impedance of the stimulation electrode is collected in real time using a preset detection current. If the detected impedance of the stimulation electrode is lower than a preset short-circuit threshold or higher than a preset detachment abnormality threshold, the electrical stimulation treatment will be forcibly terminated through a preset hardware interruption.

10. A computer program product, characterized in that, The computer program product includes instructions that, when executed, cause the method as described in any one of claims 1-9 to be implemented.

Images