Neural rehabilitation method based on ultrasonic waves

Abstract

The invention relates to the technical field of neural rehabilitation, in particular to a neural rehabilitation method based on ultrasonic waves, which comprises the following steps: acquiring a real-time rehabilitation effect index when a target neural area excites nerve excitability to the maximum, and performing fusion analysis on the real-time rehabilitation effect index based on a pre-trained rehabilitation evaluation model to obtain a neural rehabilitation result. A neural function recovery degree quantitative score is obtained; and dynamically comparing the quantitative score of the neural function recovery degree with a preset rehabilitation threshold value, and when the quantitative score of the neural function recovery degree does not reach the preset rehabilitation threshold value, automatically triggering an optimization compensation mechanism of the ultrasonic stimulation. By setting the preset rehabilitation threshold value, when the effect does not reach the standard, an optimized compensation mechanism of ultrasonic stimulation is automatically triggered, the need of manual intervention is reduced, and the continuity and effect of treatment are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of nerve rehabilitation, and particularly relates to a method for nerve rehabilitation based on ultrasonic waves. Background Art

[0002] Traditional methods usually rely on fixed treatment plans and may not be able to be adjusted according to the patient's immediate physiological responses; traditional methods usually lack a precise optimization process when selecting nerve stimulation parameters and may rely on experience or preset rules; traditional methods usually adopt a one-size-fits-all treatment plan and it is difficult to carry out personalized treatment according to the specific conditions of each patient; traditional methods may not be able to precisely control the mechanical vibration effect of ultrasonic stimulation, thereby affecting the degree of excitation of nerve excitability; the rehabilitation effect evaluation of traditional methods is usually a posteriori, which may lead to the lag of evaluation and affect the timely adjustment of treatment strategies; when the traditional nerve rehabilitation method finds that the rehabilitation effect does not meet the standard, manual intervention is usually required for adjustment. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to overcome the above-mentioned disadvantages of the prior art and provide a method for nerve rehabilitation based on ultrasonic waves.

[0004] The technical solution adopted to solve the above technical problem is: A method for nerve rehabilitation based on ultrasonic waves, comprising:

[0005] Obtaining a feedback signal sequence of ultrasonic stimulation of a target nerve region, and performing joint time-frequency domain analysis on the feedback signal sequence to obtain feedback signal characteristic parameters;

[0006] Based on the feedback signal characteristic parameters, obtaining the optimal value of the parameters of the ultrasonic stimulation, wherein the ultrasonic stimulation parameters include pulse width, duty cycle and sound intensity gradient;

[0007] Based on the PID closed-loop control algorithm and the optimal parameter value, performing real-time adaptation of the ultrasonic stimulation on the physiological state of the target nerve region, so that the mechanical vibration effect of the ultrasonic stimulation maximally stimulates nerve excitability;

[0008] Obtaining a real-time rehabilitation effect index when the nerve excitability of the target nerve region is maximally stimulated, and performing fusion analysis on the real-time rehabilitation effect index based on a pre-trained rehabilitation evaluation model to obtain a quantitative score of nerve function recovery;

[0009] Dynamically comparing the quantitative score of nerve function recovery with a preset rehabilitation threshold, and when the quantitative score of nerve function recovery does not reach the preset rehabilitation threshold, automatically triggering an optimization compensation mechanism for the ultrasonic stimulation.

[0010] Preferably, a feedback signal sequence of ultrasonic stimulation of a target nerve region is obtained, including:

[0011] Based on the ultrasonic probe emitting an ultrasonic signal, applying the ultrasonic signal to the target nerve region;

[0012] Based on the piezoelectric ceramic sensor of the ultrasonic probe, the ultrasonic feedback signal of the target nerve region is collected in real time to obtain a feedback signal sequence of the ultrasonic stimulation of the target nerve region.

[0013] Preferably, a time-frequency domain joint analysis is performed on the feedback signal sequence to obtain feedback signal characteristic parameters, including:

[0014] Based on the feedback signal sequence, a signal waveform diagram changing with time is obtained;

[0015] Based on Fourier transform, a frequency domain transformation is performed on the signal waveform diagram to obtain a signal frequency domain diagram corresponding to the signal waveform diagram;

[0016] Based on the signal waveform diagram and the signal frequency domain diagram, a time-frequency domain joint analysis is performed on the feedback signal sequence to obtain the feedback signal characteristic parameters.

[0017] Preferably, based on the feedback signal characteristic parameters, an optimal value of the parameters of the ultrasonic stimulation is obtained, including:

[0018] Based on the feedback signal characteristic parameters, a target optimization function corresponding to the ultrasonic stimulation is constructed, where the target optimization function is as follows:

[0019]

[0020] Wherein, L represents a measurement coefficient of the ultrasonic stimulation, F eff represents the effective feedback intensity of the feedback signal sequence, F max represents the maximum feedback intensity of the feedback signal sequence, represents the sound intensity gradient, the safety threshold of the sound intensity gradient, D represents the duty cycle, τ represents the pulse width, T represents the feedback time of the feedback signal sequence, and α, β, γ represent weight coefficients, satisfying α + β + γ = 1;

[0021] Based on the target optimization function, the optimal value of the parameters of the ultrasonic stimulation is obtained.

[0022] Preferably, based on the PID closed-loop control algorithm and the optimal value of the parameters, real-time adaptation of the ultrasonic stimulation to the physiological state of the target nerve region is performed, including:

[0023] Adjust the parameters of the ultrasonic stimulation based on the PID closed-loop control algorithm to obtain the PID control function corresponding to the physiological state, where the PID control function is as follows:

[0024]

[0025] Among them, u represents the PID output value, K p represents the proportional coefficient, e(t) represents the physiological state error at the current timestamp, K i represents the integral coefficient, represents the historical integral of the physiological state error, K d represents the differential coefficient, represents the change rate of the physiological state error;

[0026] Map the PID output value to the adjustment amount of the parameters of the ultrasonic stimulation to obtain the real-time adaptation of the optimal parameter value to the physiological state, so as to maximize the mechanical vibration effect of the ultrasonic stimulation to stimulate nerve excitability.

[0027] Preferably, mapping the PID output value to the adjustment amount of the parameters of the ultrasonic stimulation includes:

[0028] The adjustment mapping formula for the pulse width is as follows:

[0029]

[0030] τ represents the optimal value of the pulse width, τ opt represents the reference value of the pulse width, sat(·) represents the saturation function, u(t) represents the PID output value, u mxa represents the PID output threshold;

[0031] The adjustment mapping formula for the sound intensity gradient is as follows:

[0032] I = I opt +(1 + k·u(t));

[0033] I represents the optimal value of the sound intensity gradient, I opt represents the reference value of the sound intensity gradient, k represents the sound intensity gradient adjustment coefficient, u(t) represents the PID output value;

[0034] The adjustment mapping formula for the duty cycle is as follows:

[0035]

[0036] D represents the optimal value of the duty cycle, D opt represents the reference value of the duty cycle, u(t) represents the PID output value, u refRepresents the PID reference output value.

[0037] Preferably, the rehabilitation evaluation model includes an input layer for the characteristics vector of the rehabilitation effect index and an output layer for the quantified score of the nerve function recovery degree.

[0038] Preferably, based on the pre-trained rehabilitation evaluation model, the real-time rehabilitation effect indexes are fused and analyzed to obtain the quantified score of the nerve function recovery degree, including:

[0039] The real-time rehabilitation effect indexes are standardized to obtain the corresponding standardized indexes of the real-time rehabilitation effect indexes;

[0040] Feature extraction is performed on the standardized indexes to obtain the corresponding time-frequency feature vectors of the standardized indexes;

[0041] The time-frequency feature vectors are input into the rehabilitation evaluation model to obtain the quantified score of the nerve function recovery degree.

[0042] Preferably, when the quantified score of the nerve function recovery degree does not reach the preset rehabilitation threshold, an optimization compensation mechanism for the ultrasonic stimulation is automatically triggered, including:

[0043] When the quantified score of the nerve function recovery degree does not reach the preset rehabilitation threshold, the frequency, amplitude and duration of the ultrasonic stimulation will be automatically adjusted;

[0044] When the quantified score of the nerve function recovery degree reaches the preset rehabilitation threshold, the adjustment of the frequency, amplitude and duration of the ultrasonic stimulation is automatically stopped.

[0045] The beneficial effects of the present invention are as follows: (1) By acquiring the real-time feedback signal of the target nerve area and combining time-frequency domain analysis and the PID control algorithm, the present invention can adjust the parameters of the ultrasonic stimulation in real time to ensure that the stimulation effect always remains in the optimal state; (2) By analyzing the characteristic parameters of the feedback signal, the present invention dynamically optimizes the parameters of the ultrasonic stimulation (such as pulse width, duty cycle and sound intensity gradient), thereby improving the treatment effect; (3) By dynamically adjusting the treatment plan based on the characteristic parameters of the feedback signal, the present invention enables each patient to receive a customized treatment, improving the rehabilitation effect; (4) By using the PID closed-loop control algorithm, the present invention maximizes the mechanical vibration effect of the stimulation to ensure that the stimulation effect reaches the best nerve excitability; (5) By obtaining the rehabilitation effect indexes in real time and combining with the analysis of the pre-trained rehabilitation evaluation model, the present invention can immediately evaluate the treatment effect and then make corresponding adjustments quickly, which has great contribution value to the nerve rehabilitation in the biomedical engineering industry; (6) By setting a preset rehabilitation threshold and automatically triggering the optimization compensation mechanism of the ultrasonic stimulation when the effect does not meet the standard, the present invention reduces the need for manual intervention and improves the continuity and effect of the treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a schematic diagram of the step flow of the overall method in an embodiment proposed by the present invention; DETAILED DESCRIPTION OF THE INVENTION

[0047] Embodiment 1, as Figure 1 shown, a method for nerve rehabilitation based on ultrasonic waves proposed by the present invention includes:

[0048] S1. Obtain the feedback signal sequence of ultrasonic stimulation of the target nerve area, and perform joint time-frequency domain analysis on the feedback signal sequence to obtain the feedback signal characteristic parameters;

[0049] S2. Obtain the optimal value of the ultrasonic stimulation parameters based on the feedback signal characteristic parameters, where the ultrasonic stimulation parameters include pulse width, duty cycle, and sound intensity gradient;

[0050] S3. Based on the PID closed-loop control algorithm and the optimal parameter value, perform real-time adaptation of ultrasonic stimulation on the physiological state of the target nerve area, so that the mechanical vibration effect of ultrasonic stimulation maximally stimulates nerve excitability;

[0051] S4. Obtain the real-time rehabilitation effect index when the nerve excitability of the target nerve area is maximally stimulated, and perform fusion analysis on the real-time rehabilitation effect index based on the pre-trained rehabilitation evaluation model to obtain the quantitative score of nerve function recovery;

[0052] S5. Dynamically compare the quantitative score of nerve function recovery with the preset rehabilitation threshold. When the quantitative score of nerve function recovery does not reach the preset rehabilitation threshold, automatically trigger the optimization compensation mechanism of ultrasonic stimulation.

[0053] In the present invention, the feedback signal sequence of ultrasonic stimulation refers to the sequence of electrical signals or physiological responses generated by nerve responses after ultrasonic stimulation of the target nerve region. These feedback signals reflect the excitability and response of the nerves; the joint time-frequency domain analysis refers to the analysis of signals in both the time domain and the frequency domain simultaneously. Through this method, time-domain features (such as waveform changes) and frequency-domain features (such as signal frequency components) in the signal can be extracted, so as to obtain more comprehensive signal features; the feedback signal characteristic parameters refer to the key parameters extracted from the feedback signals, which are used to describe the nature of nerve responses, such as the frequency, amplitude, waveform, duration, etc. of the signal; the optimal values of ultrasonic stimulation parameters refer to these parameters including pulse width, duty cycle and sound intensity gradient, which are the key control parameters of ultrasonic stimulation. The optimal value refers to the stimulation parameters that can maximize nerve excitability and rehabilitation effect; the pulse width refers to the duration of each pulse of the ultrasonic stimulation signal, the duty cycle refers to the ratio of the "on" and "off" states in the ultrasonic stimulation signal, and the sound intensity gradient refers to the rate of change of sound intensity with space during ultrasonic stimulation. The PID closed-loop control algorithm is a feedback control mechanism used to automatically adjust the parameters of ultrasonic stimulation. PID respectively represents proportional (P), integral (I) and derivative (D) control, which can dynamically adjust the control parameters during the stimulation process to achieve the best effect. Nerve excitability refers to the response ability of nerve cells to stimulation. Ultrasonic stimulation aims to maximize nerve excitability through mechanical vibration, thereby promoting nerve recovery; the real-time rehabilitation effect index refers to the real-time measurement value of nerve recovery, usually involving the evaluation of indicators such as nerve function, sensation, and motor ability; the rehabilitation evaluation model refers to the model trained by historical data, which is used to evaluate the effect of nerve recovery; the quantitative score of nerve function recovery degree refers to the quantitative score of nerve function recovery obtained by combining the real-time rehabilitation effect index and the rehabilitation evaluation model, which is usually used to evaluate the quality of the treatment effect; the preset rehabilitation threshold refers to a standard value set according to clinical experience or scientific research, representing the minimum requirement for treatment success. If the quantitative score of nerve function recovery degree is lower than this threshold, the treatment needs to be optimized; the optimization compensation mechanism means that when the quantitative score of nerve function recovery degree does not reach the preset rehabilitation threshold, the system will automatically adjust the frequency, amplitude and duration of ultrasonic stimulation to improve the treatment effect.

[0054] Example 2. A method for nerve rehabilitation based on ultrasonic waves proposed by the present invention. Compared with Example 1, this example further includes:

[0055] A1. Based on the ultrasonic probe, emit ultrasonic signals and apply the ultrasonic signals to the target nerve region;

[0056] A2. Based on the piezoelectric ceramic sensor of the ultrasonic probe, collect the ultrasonic feedback signals of the target nerve region in real time to obtain the feedback signal sequence of ultrasonic stimulation of the target nerve region.

[0057] In this embodiment, an ultrasonic probe refers to a device for transmitting and receiving ultrasonic waves; the target nerve area refers to a specific nerve or nerve structure selected for ultrasonic stimulation during treatment or experiments; a piezoelectric ceramic sensor is a sensor that can convert mechanical energy (such as pressure, vibration) into electrical energy. When an ultrasonic signal acts on the target area, the piezoelectric ceramic sensor can capture these reflected ultrasonic signals and convert them into electrical signals for further processing; an ultrasonic feedback signal refers to the signal generated after the ultrasonic signal acts on the target nerve area and is reflected back or through tissue changes.

[0058] In an alternative embodiment, a time-frequency domain joint analysis is performed on the feedback signal sequence to obtain feedback signal characteristic parameters, including:

[0059] B1. Obtain a signal waveform diagram that changes with time based on the feedback signal sequence;

[0060] B2. Perform a frequency domain transformation on the signal waveform diagram based on the Fourier transform to obtain the signal frequency domain diagram corresponding to the signal waveform diagram;

[0061] B3. Perform a time-frequency domain joint analysis on the feedback signal sequence based on the signal waveform diagram and the signal frequency domain diagram to obtain the feedback signal characteristic parameters.

[0062] It should be noted that the signal waveform diagram is an image of the signal on the time axis, showing the change of the signal over time; the Fourier transform is a mathematical method that converts the signal from the time domain to the frequency domain, which can reveal the contribution of each frequency component in the signal. In the frequency domain, the signal is decomposed into sine waves of different frequencies, and the Fourier transform can help analyze the periodicity and spectral characteristics of the signal; the signal frequency domain diagram is an image of the signal on the frequency axis, showing the spectral information of the signal. The signal frequency domain diagram obtained through the Fourier transform can reflect the amplitude and phase of each frequency component of the signal.

[0063] In an alternative embodiment, the optimal value of the ultrasonic stimulation parameter is obtained based on the feedback signal characteristic parameters, including:

[0064] C1. Construct a target optimization function corresponding to the ultrasonic stimulation based on the feedback signal characteristic parameters, where the target optimization function is as follows:

[0065]

[0066] Among them, L represents the measurement coefficient of the ultrasonic stimulation, F eff represents the effective feedback intensity of the feedback signal sequence, F max represents the maximum feedback intensity of the feedback signal sequence, represents the sound intensity gradient, The safety threshold of the sound intensity gradient, D represents the duty cycle, τ represents the pulse width, T represents the feedback time of the feedback signal sequence, and α, β, γ represent the weight coefficients, satisfying α + β + γ = 1;

[0067] C2. Obtain the optimal values of the parameters of the ultrasonic stimulation based on the target optimization function.

[0068] It should be noted that the target optimization function is a mathematical expression used to find the optimal ultrasonic stimulation parameters. By optimizing this function, the optimal ultrasonic stimulation settings can be found; the safety threshold of the sound intensity gradient refers to that during the ultrasonic stimulation process, if the sound intensity gradient exceeds this threshold, it may cause adverse effects on the target object or organism; the feedback time refers to the time required from the start of the ultrasonic stimulation to the completion of the feedback signal; the weight coefficient is an adjustment parameter in the target optimization function, indicating the relative importance of each characteristic parameter in the optimization process.

[0069] In an optional embodiment, based on the PID closed-loop control algorithm and the optimal parameter values, real-time adaptation of the physiological state of the target nerve area is performed for ultrasonic stimulation, including:

[0070] D1. Adjust the parameters of the ultrasonic stimulation based on the PID closed-loop control algorithm to obtain the PID control function corresponding to the physiological state, where the PID control function is as follows:

[0071]

[0072] Among them, u represents the PID output value, K p represents the proportional coefficient, e(t) represents the physiological state error at the current time stamp, K i represents the integral coefficient, represents the historical integral of the physiological state error, K d represents the differential coefficient, represents the change rate of the physiological state error;

[0073] D2. Map the PID output value to the adjustment amount of the parameters of the ultrasonic stimulation to obtain the real-time adaptation of the optimal parameter values and the physiological state, so as to maximize the mechanical vibration effect of the ultrasonic stimulation to stimulate nerve excitability.

[0074] It should be noted that the PID control function refers to the mathematical expression based on the PID algorithm, which describes how to calculate the output value according to proportional, integral, and derivative operations; the proportional coefficient is the proportional relationship between the control signal and the error in PID control; the physiological state error refers to the gap between the current physiological state and the desired physiological state; the integral coefficient is the coefficient of the integral part in PID control, which reflects the accumulation of past errors. Its main role is to eliminate long-term small errors and ensure that the system can operate stably for a long time; the historical integral of the physiological state error is the cumulative value of all physiological state errors over the past time, which helps the PID controller calculate and compensate for historical errors, enabling long-term errors to be corrected; the derivative coefficient is the coefficient of the derivative part in PID control, which adjusts the controller output according to the rate of change of the error, thereby improving the system's response ability to future changes; the rate of change of the physiological state error refers to the rate of change of the error at the current moment. The derivative part of the PID controller makes adjustments in advance by considering the trend of error change, thereby reducing the overshoot and fluctuation of the system; the PID output value is the control signal output by the PID controller after calculation, which is usually used to adjust system parameters and is used to adjust the parameters of ultrasonic stimulation to ensure the real-time adaptation of the stimulation to the target physiological state; the parameter adjustment amount of ultrasonic stimulation is the specific value obtained by mapping the PID output value to the control parameters of ultrasonic stimulation; the real-time adaptation of the physiological state refers to monitoring the changes in the physiological state in real time and immediately adjusting the parameters of ultrasonic stimulation so that the ultrasonic stimulation and the physiological state maintain the best match to achieve the optimal stimulation effect; the mechanical vibration effect refers to the impact of the physical vibration generated by ultrasonic stimulation on the target area (such as the nerve area). Through appropriate vibration, nerve excitability can be stimulated, thereby affecting the physiological state of the target area.

[0075] In an optional embodiment, mapping the PID output value to the adjustment amount of the parameters of ultrasonic stimulation includes:

[0076] E1. The adjustment mapping formula for the pulse width is as follows:

[0077]

[0078] τ represents the optimal value of the pulse width, τ opt represents the reference value of the pulse width, sat(·) represents the saturation function, u(t) represents the PID output value, u mxa represents the PID output threshold;

[0079] E2. The adjustment mapping formula for the sound intensity gradient is as follows:

[0080] I = I opt +(1 + k·u(t));

[0081] I represents the optimal value of the sound intensity gradient, Iopt represents the reference value of the sound intensity gradient, k represents the sound intensity gradient adjustment coefficient, and u(t) represents the PID output value;

[0082] E3. The adjustment mapping formula for the duty cycle is as follows:

[0083]

[0084] D represents the optimal value of the duty cycle, D opt represents the reference value of the duty cycle, u(t) represents the PID output value, u ref represents the PID reference output value.

[0085] In an optional embodiment, the rehabilitation evaluation model includes a rehabilitation effect index feature vector input layer and a nerve function recovery degree quantization scoring output layer.

[0086] In an optional embodiment, based on the pre-trained rehabilitation evaluation model, a fusion analysis is performed on the real-time rehabilitation effect indicators to obtain the nerve function recovery degree quantization score, including:

[0087] F1. Standardize the real-time rehabilitation effect indicators to obtain the standardized indicators corresponding to the real-time rehabilitation effect indicators;

[0088] F2. Extract features from the standardized indicators to obtain the time-frequency feature vector corresponding to the standardized indicators;

[0089] F3. Input the time-frequency feature vector into the rehabilitation evaluation model to obtain the nerve function recovery degree quantization score.

[0090] It should be noted that the standardization process refers to converting the rehabilitation effect indicators with different dimensions and ranges into unified standardized values, which can eliminate the unit and scale differences between different measurement indicators and facilitate subsequent processing and analysis. Feature extraction refers to extracting representative information from the standardized data. The time-frequency feature vector refers to extracting the time-frequency features of the signals during the patient's rehabilitation process by converting the time-domain data into the frequency-domain data and combining the two dimensions of time and frequency, which helps to more accurately evaluate the nerve function recovery.

[0091] In an optional embodiment, when the nerve function recovery degree quantization score does not reach the preset rehabilitation threshold, an optimization compensation mechanism for ultrasonic stimulation is automatically triggered, including:

[0092] G1. When the nerve function recovery degree quantization score does not reach the preset rehabilitation threshold, the frequency, amplitude, and duration of the ultrasonic stimulation will be automatically adjusted;

[0093] G2. When the nerve function recovery degree quantization score reaches the preset rehabilitation threshold, automatically stop adjusting the frequency, amplitude, and duration of the ultrasonic stimulation.

[0094] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto, and various changes can be made without departing from the spirit of the present invention within the scope of knowledge possessed by those skilled in the art.

Claims

1. A method for nerve rehabilitation based on ultrasonic waves, comprising, characterized in that: Obtain a feedback signal sequence of ultrasonic stimulation of a target nerve area, and perform joint time-frequency domain analysis on the feedback signal sequence to obtain feedback signal characteristic parameters; Obtain the optimal parameter values of the ultrasonic stimulation based on the feedback signal characteristic parameters, wherein the ultrasonic stimulation parameters include pulse width, duty cycle and sound intensity gradient; Based on the PID closed-loop control algorithm and the optimal parameter values, perform real-time adaptation of the ultrasonic stimulation on the physiological state of the target nerve area, so that the mechanical vibration effect of the ultrasonic stimulation maximally stimulates nerve excitability; Obtain the real-time rehabilitation effect index when the nerve excitability of the target nerve area is maximally stimulated, and perform fusion analysis on the real-time rehabilitation effect index based on a pre-trained rehabilitation evaluation model to obtain a quantitative score of nerve function recovery; Dynamically compare the quantitative score of nerve function recovery with a preset rehabilitation threshold, and when the quantitative score of nerve function recovery does not reach the preset rehabilitation threshold, automatically trigger the optimization compensation mechanism of the ultrasonic stimulation.

2. The method for nerve rehabilitation based on ultrasonic waves according to claim 1, wherein Obtain a feedback signal sequence of ultrasonic stimulation of a target nerve area, wherein, including: Based on an ultrasonic probe, emit an ultrasonic signal, and apply the ultrasonic signal to the target nerve area; Based on the piezoelectric ceramic sensor of the ultrasonic probe, collect the ultrasonic feedback signal of the target nerve area in real time to obtain a feedback signal sequence of ultrasonic stimulation of the target nerve area.

3. The method for nerve rehabilitation based on ultrasonic waves according to claim 2, characterized in that Perform joint time-frequency domain analysis on the feedback signal sequence to obtain feedback signal characteristic parameters, including: Based on the feedback signal sequence, obtain a signal waveform diagram that changes with time; Perform frequency domain transformation on the signal waveform diagram based on Fourier transform to obtain a signal frequency domain diagram corresponding to the signal waveform diagram; Based on the signal waveform diagram and the signal frequency domain diagram, perform joint time-frequency domain analysis on the feedback signal sequence to obtain the feedback signal characteristic parameters.

4. A method for nerve rehabilitation based on ultrasonic waves according to claim 3, wherein Obtain the optimal parameter values of the ultrasonic stimulation based on the feedback signal characteristic parameters, including: Based on the feedback signal characteristic parameters, construct a target optimization function corresponding to the ultrasonic stimulation, wherein the target optimization function is as follows: wherein, L represents the measurement coefficient of the ultrasonic stimulation, F eff represents the effective feedback intensity of the feedback signal sequence, F max represents the maximum feedback intensity of the feedback signal sequence, represents the sound intensity gradient, the safety threshold of the sound intensity gradient, D represents the duty cycle, τ represents the pulse width, T represents the feedback time of the feedback signal sequence, and α, β, γ represent weight coefficients, satisfying α + β + γ = 1; Based on the target optimization function, obtain the optimal parameter values of the ultrasonic stimulation.

5. A method for nerve rehabilitation based on ultrasonic waves according to claim 4, characterized in that, Based on the PID closed-loop control algorithm and the optimal parameter values, perform real-time adaptation of the ultrasonic stimulation on the physiological state of the target nerve area, including: Based on the PID closed-loop control algorithm, adjust the parameters of the ultrasonic stimulation to obtain a PID control function corresponding to the physiological state, wherein the PID control function is as follows: Among them, u represents the PID output value, K p represents the proportionality coefficient, e(t) represents the physiological state error at the current time stamp, K i represents the integral coefficient, represents the historical integral of the physiological state error, K d represents the differential coefficient, represents the rate of change of the physiological state error; Map the PID output value to the adjustment amount of the parameters of the ultrasonic stimulation to obtain the real-time adaptation of the optimal parameter values and the physiological state, so that the mechanical vibration effect of the ultrasonic stimulation maximally stimulates nerve excitability.

6. A method for nerve rehabilitation based on ultrasonic waves according to claim 5, characterized in that, Map the PID output value to the adjustment amount of the parameters of the ultrasonic stimulation, including: The adjustment mapping formula of the pulse width is as follows: τ represents the optimal value of the pulse width, τ opt represents the reference value of the pulse width, sat(·) represents the saturation function, u(t) represents the PID output value, u max represents the PID output threshold; The adjustment mapping formula of the sound intensity gradient is as follows: I = I opt +(1 + k·u(t)); I represents the optimal value of the sound intensity gradient, I opt represents the reference value of the sound intensity gradient, k represents the sound intensity gradient adjustment coefficient, and u(t) represents the PID output value; The adjustment mapping formula of the duty cycle is as follows: D represents the optimal value of the duty cycle, D opt represents the reference value of the duty cycle, u(t) represents the PID output value, u ref represents the PID reference output value.

7. A method for nerve rehabilitation based on ultrasonic waves according to claim 6, characterized in that, The rehabilitation evaluation model includes an input layer of rehabilitation effect index feature vectors and an output layer of quantitative scores for nerve function recovery degree.

8. A method for nerve rehabilitation based on ultrasonic waves according to claim 7, characterized in that, Performing fusion analysis on the real-time rehabilitation effect indicators based on the pre-trained rehabilitation evaluation model to obtain a quantitative score for nerve function recovery degree, including: Performing standardization processing on the real-time rehabilitation effect indicators to obtain the standardized indicators corresponding to the real-time rehabilitation effect indicators; Performing feature extraction on the standardized indicators to obtain the time-frequency feature vectors corresponding to the standardized indicators; Inputting the time-frequency feature vectors into the rehabilitation evaluation model to obtain the quantitative score for nerve function recovery degree.

9. A method for nerve rehabilitation based on ultrasonic waves according to claim 8, characterized in that, When the quantitative score for nerve function recovery degree does not reach the preset rehabilitation threshold, automatically trigger the optimization compensation mechanism for the ultrasonic stimulation, including: When the quantitative score for nerve function recovery degree does not reach the preset rehabilitation threshold, automatically adjust the frequency, amplitude, and duration of the ultrasonic stimulation; When the quantitative score for nerve function recovery degree reaches the preset rehabilitation threshold, automatically stop adjusting the frequency, amplitude, and duration of the ultrasonic stimulation.

Images