An alternating cooperative stimulation control method suitable for asymmetric gait

Abstract

This invention relates to the fields of rehabilitation medicine, brain-computer interfaces, and functional electrical stimulation, and discloses an alternating synergistic stimulation control method suitable for asymmetric gait. The invention collects various baseline data from the patient's healthy and affected sides, calculates the initial asymmetry, and initializes training parameters based on the baseline data. After configuration, the patient undergoes rehabilitation training, and various training data are collected in real time. The training process is controlled in real time, and the phase synchronization rate and cumulative asymmetry are calculated based on the training data, thereby optimizing the training parameters in a closed-loop manner. This invention solves the specific problem of asymmetric gait in hemiplegic patients, significantly improves gait coordination and neural remodeling efficiency, and is applicable to lower limb rehabilitation devices such as brain-computer interface rehabilitation treadmills, possessing extremely high clinical application value.

Description

Technical Field

[0001] This invention relates to the fields of rehabilitation medicine, brain-computer interface and functional electrical stimulation, specifically to an alternating synergistic stimulation control method applicable to asymmetric gait. Background Technology

[0002] Stroke, spinal cord injury and other central nervous system diseases have become the leading cause of limb motor dysfunction worldwide. The core manifestation is a severe asymmetry in motor function between the healthy and affected limbs, resulting in asymmetrical gait, such as limping, circling gait, and significant differences in stride length.

[0003] The core rehabilitation need for these patients is to rebuild the coordinated movement relationship of both limbs and activate the neural and muscular pathways on the affected side. However, the stimulation control strategies of existing rehabilitation equipment have significant limitations: existing equipment mostly uses uniform stimulation parameters and modes, without distinguishing the functional differences between the healthy and affected sides, applying only a single intensity of FES stimulation to the affected side, ignoring the synergistic guidance role of the healthy side, resulting in a disconnect between the timing of movements on the affected side and those on the healthy side, making it difficult to correct asymmetrical gait; it does not establish a coordinated relationship between bilateral movements based on gait cycles, and the timing of FES stimulation does not match the phase of movement on the healthy side and gait on the affected side, which not only leads to poor rehabilitation effects but may also reinforce abnormal gait patterns; it does not take into account the patient's real-time movement intentions, and the stimulation parameters on the affected side are fixed, which cannot adapt to the changes in neuroplasticity at different rehabilitation stages, and is prone to ineffective stimulation or accidental triggering, affecting rehabilitation safety and compliance; traditional programs are mainly passive training, with insufficient active participation from the patient's brain, and cannot efficiently activate neuroplasticity to achieve functional reconstruction on the affected side.

[0004] With the development of brain-computer interface (BCI) and functional electrical stimulation (FES) technologies, the hybrid BCI paradigm has achieved high-precision motor intention recognition, providing technical support for personalized rehabilitation. Against this background, this invention proposes an alternating synergistic stimulation control method suitable for asymmetric gait. Through differentiated strategies and temporal coupling mechanisms, it addresses the aforementioned technical bottlenecks and improves the rehabilitation effect for hemiplegic patients. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an alternating synergistic stimulation control method suitable for asymmetric gait, which has the advantages of high efficiency, strong universality and high safety, and solves the above-mentioned technical problems.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an alternating cooperative stimulus control method suitable for asymmetric gait, comprising the following steps: S1: Collect various baseline data from the patient's healthy and affected sides; S2: The initial asymmetry is calculated based on the baseline data acquired in S1. ; S3: Initialize and configure training parameters based on the baseline data collected in S1; S4: The patient undergoes rehabilitation training, and various training data are collected in real time to control the training process in real time. S5: Calculate the phase synchronization rate based on the training data collected in S4. and cumulative asymmetry ; S6: Phase synchronization rate obtained based on S5 and cumulative asymmetry Closed-loop optimization of training parameters is performed.

[0007] As a preferred technical solution of the present invention, the initial asymmetry of S2 The expression is as follows: In the formula, Gait asymmetry, For the healthy side stride length, For the affected side's stride length, This indicates taking the maximum step size between the healthy and affected sides. It represents the absolute value.

[0008] As a preferred embodiment of the present invention, the initialization configuration of training parameters in S3 specifically includes: setting the functional electrical stimulation intensity on the healthy side to 3mA, locking the stimulation timing to mid-support phase; setting the initial intensity of functional electrical stimulation on the affected side according to muscle strength level; and setting the timing coupling coefficient. The initial value is set to 1.0.

[0009] As a preferred embodiment of the present invention, the real-time control of the training process in step S4 includes the following steps: Step A1: Calculate the overall confidence level using the weighted fusion formula. Its expression is as follows: In the formula, Weighting of EEG signals For electromyographic signal weights, Confidence level for EEG intent decoding The normalized confidence value of electromyographic activation is expressed as follows: In the formula, Real-time electromyographic activation intensity, This represents the baseline value for electromyographic activation intensity. When the overall confidence level When the confidence level is ≥85%, strong stimulation is directly triggered on the affected side; when the overall confidence level is ≤60%, strong stimulation is triggered on the affected side. When <85%, if the real-time electromyographic activation intensity ≥0.3 × baseline value of electromyographic activation intensity This triggers strong stimulation on the affected side, and when the overall confidence level is... When the effect rate is less than 60%, switch to 5mA mild auxiliary stimulation. Step A2: Real-time acquisition of the healthy side gait phase using a crank phase encoder. Calculate the target phase on the affected side based on the time-series coupling formula With the timing of stimulus triggering The relevant expressions are as follows: In the formula, This is the gait phase compensation value. These are the timing coupling coefficients. For real-time gait cycles; When the initial asymmetry When >30%, gait phase compensation value =30°, when 20% ≤ initial asymmetry When ≤30%, gait phase compensation value =20°, when the initial asymmetry is... When <20%, gait phase compensation value =10°; Step A3: Monitor the electromyographic signal on the affected side in real time, and determine the intensity of real-time electromyographic activation. ≥2.0 × baseline value of electromyographic activation intensity If muscle spasm is detected, functional electrical stimulation on the affected side is immediately cut off and a voice prompt is given. The signal is reacquired after a 5-second interval. If the real-time electromyographic activation intensity is... <1.5 × baseline value of electromyographic activation intensity Then, mild auxiliary stimulation will be restored.

[0010] As a preferred technical solution of the present invention, the phase synchronization rate is calculated in step S5. and cumulative asymmetry The relevant expressions are as follows: In the formula, For phase synchronization rate, The standard deviation of the bilateral phase difference over 5 gait cycles. To accumulate asymmetry, For the first Asymmetry of each gait cycle, This indicates a summation operation over 5 periods.

[0011] As a preferred technical solution of the present invention, the S6 method for closed-loop optimization of training parameters specifically involves: if the cumulative asymmetry... If the rate is >20%, the functional electrical stimulation intensity on the affected side increases by 1 mA, not exceeding 20 mA; if the phase synchronization rate is >20%, then the intensity increases by 1 mA. If the value is greater than 0.85, then the timing coupling coefficient is... The threshold should be reduced by 0.1, but not lower than 0.6; if the muscle strength on the affected side increases by 1 grade, the EEG intention confidence threshold should be increased by 5%, but not exceeding 95%; if the muscle synergistic contraction rate is >40%, the functional electrical stimulation intensity on the affected side should be reduced by 2 mA, but not lower than 8 mA.

[0012] Compared with the prior art, the present invention provides an alternating cooperative stimulus control method suitable for asymmetric gait, which has the following beneficial effects: This invention reconstructs bilateral limb motor associations through a temporal coupling mechanism, increasing phase synchronization rate to over 85%, effectively improving abnormal gait such as limping and circling. By combining EEG intention confidence judgment with multimodal feedback, the ineffective stimulation rate is reduced by 60%, and the false trigger rate is controlled within 5%. Furthermore, the differentiated configuration of strong / weak functional electrical stimulation adapts to bilateral functional states, balancing stimulation accuracy and safety. The training parameters are optimized in a closed loop through phase synchronization rate and cumulative asymmetry. Clinical pre-experiments show that after 4 weeks of training, patients experienced a reduction of gait asymmetry of over 30%, an average increase of 0.8 grade in muscle strength on the affected side, and a significant improvement in neural remodeling efficiency. It can dynamically adapt to the functional changes of patients from the acute phase to the recovery phase, enhancing the clinical universality of the product and providing a more efficient and safer technical solution for asymmetric gait rehabilitation in hemiplegic patients. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of the framework of the present invention; Figure 2 This is a schematic diagram illustrating the overall principle of the alternating synergistic stimulation control method of the present invention; Figure 3 A comparative graph showing the differences in FES stimulation parameters on the affected side; Figure 4 This is a schematic diagram of the temporal coupling of the gait cycle on the healthy and affected sides. Figure 5 This is a flowchart illustrating the control method of the present invention. Detailed Implementation

[0014] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0015] Please see Figures 1-5 An alternating cooperative stimulus control method applicable to asymmetric gait includes the following steps: S1: Collect various baseline data of the patient's healthy and affected sides, including healthy side step length, affected side step length, baseline values ​​of bilateral muscle activation intensity, and gait cycle parameters; S2: The initial asymmetry is calculated based on the baseline data acquired in S1. Its expression is as follows: In the formula, Gait asymmetry, For the healthy side stride length, For the affected side's stride length, This indicates taking the maximum step size between the healthy and affected sides. Represents absolute value; S3: Initialize and configure training parameters based on the baseline data acquired in S1, specifically: set the functional electrical stimulation intensity on the healthy side to 3mA, and lock the stimulation timing to mid-stance phase; set the initial intensity of functional electrical stimulation on the affected side according to muscle strength level; and adjust the timing coupling coefficient. The initial value is set to 1.0; S4: The patient undergoes rehabilitation training, and various training data are collected in real time to control the training process in real time, including the following steps: Step A1: Calculate the overall confidence level using the weighted fusion formula. Its expression is as follows: In the formula, Weighting of EEG signals For electromyographic signal weights, Confidence level for EEG intent decoding The normalized confidence value of electromyographic activation is expressed as follows: in, Real-time electromyographic activation intensity, This represents the baseline value for electromyographic activation intensity. When the overall confidence level When the confidence level is ≥85%, strong stimulation is directly triggered on the affected side; when the overall confidence level is ≤60%, strong stimulation is triggered on the affected side. When <85%, if the real-time electromyographic activation intensity ≥0.3 × baseline value of electromyographic activation intensity This triggers strong stimulation on the affected side, and when the overall confidence level is... When the effect rate is less than 60%, switch to 5mA mild auxiliary stimulation. Step A2: Real-time acquisition of the healthy side gait phase using a crank phase encoder. Calculate the target phase on the affected side based on the time-series coupling formula With the timing of stimulus triggering The relevant expressions are as follows: In the formula, This is the gait phase compensation value. These are the timing coupling coefficients. For real-time gait cycles; When the initial asymmetry When >30%, gait phase compensation value =30°, when 20% ≤ initial asymmetry When ≤30%, gait phase compensation value =20°, when the initial asymmetry is... When <20%, gait phase compensation value =10°; Step A3: Monitor the electromyographic signal on the affected side in real time, and determine the intensity of real-time electromyographic activation. ≥2.0 × baseline value of electromyographic activation intensity If muscle spasm is detected, functional electrical stimulation on the affected side is immediately cut off and a voice prompt is given. The signal is reacquired after a 5-second interval. If the real-time electromyographic activation intensity is... <1.5 × baseline value of electromyographic activation intensity Then, mild auxiliary stimulation is restored; S5: Calculate the phase synchronization rate based on the training data collected in S4. and cumulative asymmetry The relevant expressions are as follows: In the formula, For phase synchronization rate, The standard deviation of the bilateral phase difference over 5 gait cycles. To accumulate asymmetry, For the first Asymmetry of each gait cycle, This indicates a summation operation over 5 periods; S6: Phase synchronization rate obtained based on S5 and cumulative asymmetry Closed-loop optimization of the training parameters is performed, specifically: if the cumulative asymmetry... If the rate is >20%, the functional electrical stimulation intensity on the affected side increases by 1 mA, not exceeding 20 mA; if the phase synchronization rate is >20%, then the intensity increases by 1 mA. If the value is greater than 0.85, then the timing coupling coefficient is... The threshold should be reduced by 0.1, but not lower than 0.6; if the muscle strength on the affected side increases by 1 grade, the EEG intention confidence threshold should be increased by 5%, but not exceeding 95%; if the muscle synergistic contraction rate is >40%, the functional electrical stimulation intensity on the affected side should be reduced by 2 mA, but not lower than 8 mA.

[0016] The specific implementation method is as follows: The brain-computer interface rehabilitation treadmill integrates a crank phase encoder (0.1° resolution) and an IMU sensor (100Hz sampling rate); the EEG acquisition device uses a 32-channel wireless dry electrode EEG with a sampling rate of 1000Hz, supporting the MI+SSVEP hybrid paradigm; the FES stimulation module is 8-channel independently programmable, with adjustable current intensity from 0-20mA, frequency of 30Hz, and pulse width of 300μs; the sEMG sensor has a sampling rate of 2000Hz and is applied to the rectus femoris, gastrocnemius, and tibialis anterior muscles on the healthy and affected sides respectively; the supporting data processing system supports real-time fusion of multimodal signals and formula calculation.

[0017] Ten patients with hemiplegia after stroke (with a disease duration of 3-6 months, muscle strength grade 4 on the healthy side and grade 2-3 on the affected side, exhibiting obvious asymmetrical gait) were selected. All patients signed informed consent forms, and contraindications such as severe cognitive impairment, skin allergies, and severe cardiovascular diseases were excluded.

[0018] Step 1: Baseline Data Acquisition and Healthy / Affected Side Calibration The patient wore an EEG device and sEMG sensor and performed 5 minutes of unloaded treadmill training to collect the step length of the unaffected side. Step length on the affected side Bilateral muscle activation intensity baseline value gait cycle The initial asymmetry is calculated using the gait asymmetry formula, taking parameters such as gait and gait asymmetry. In the formula, Gait asymmetry (unit: %) Step length on the healthy side (unit: cm). Step length on the affected side (unit: cm). This indicates taking the maximum step size between the healthy and affected sides to establish an individual functional baseline for the patient and clearly identify the healthy and affected sides.

[0019] Step 2: Control Strategy Initialization Configuration Parameter initialization was performed based on baseline data: the FES intensity on the healthy side was set to 3 mA, and the stimulation timing was locked in the mid-stance phase (90°-120°); the initial FES intensity on the affected side was set according to muscle strength grade, with 14 mA corresponding to grade 2 muscle strength and 12 mA corresponding to grade 3 muscle strength; the initial value of the EEG intention confidence threshold was set to 85%; and the stimulation channels activated the rectus femoris (Ch1), gastrocnemius (Ch3), and tibialis anterior (Ch5) muscles; the temporal coupling coefficient was also determined. The initial value is set to 1.0, which is used to adjust the synchronous response speed between stimulation on the affected side and movement on the healthy side.

[0020] Step 3: Implementation and Real-time Control of Rehabilitation Training The patient undergoes 30-minute rehabilitation training sessions, with real-time control implemented according to the following logic: (1) Fusion decoding of EEG intention and EMG signals: The overall confidence level was calculated using a weighted fusion formula. Assess the intensity of the movement intention on the affected side: In the formula, The weight of the EEG signal is 0.7. The weight of the electromyographic signal is 0.3. Confidence level for EEG intent decoding This represents the normalized confidence value of electromyographic activation. , Real-time electromyographic activation intensity, This represents the baseline value of electromyographic activation intensity. ≥85% directly triggers strong FES stimulation on the affected side, 60%≤ If <85% ≥0.3× This triggers a strong stimulus. When the saturation level is less than 60%, switch to 5mA mild auxiliary stimulation.

[0021] (2) Gait timing coupling control: The healthy side gait phase is acquired in real time by a crank phase encoder. Calculate the target phase on the affected side based on the time-series coupling formula With the timing of stimulus triggering : In the formula, The gait phase compensation value (set according to the initial asymmetry) >30% =30°, 20%≤ ≤30% =20°, <20% =10°), T is the real-time gait cycle (unit: s), ensuring that the timing of the stance and swing phases on the affected side matches that on the healthy side. This represents the timing coupling coefficient.

[0022] (3) Safety protection monitoring: Real-time monitoring of electromyographic signals on the affected side, when ≥2.0× If muscle spasm is detected, FES stimulation on the affected side is immediately cut off and a voice prompt is given. The signal is reacquired after a 5-second interval. <1.5× Then, mild auxiliary stimulation is restored.

[0023] Step 4: Closed-loop parameter optimization and adjustment Closed-loop optimization is performed every 5 gait cycles to calculate the phase synchronization rate. With cumulative asymmetry : In the formula, Phase synchronization rate, representing the gait synchronization index (range 0-1, the closer to 1, the better the synchronization). The standard deviation of the bilateral phase difference over 5 gait cycles (unit: °). Let be the asymmetry of the i-th gait cycle. The cumulative asymmetry is represented by the average asymmetry over 5 periods. The optimization rule is: if If >20%, the FES intensity on the affected side increases by 1 mA (not exceeding 20 mA); if If the value is greater than 0.85, then the timing coupling coefficient is... Decrease by 0.1 (not less than 0.6); if the muscle strength on the affected side increases by 1 grade, the EEG intention confidence threshold increases by 5% (not exceeding 95%); if the muscle co-contraction rate is >40%, the FES intensity on the affected side decreases by 2mA (not less than 8mA).

[0024] After 4 weeks (3 times / week) of training, all 10 patients showed significant improvement in key indicators calculated by the formula: cumulative asymmetry. The phase synchronization rate decreased from 32.6% to 9.8%. The score increased from 0.62 to 0.89, with an average improvement of 0.9 grade in muscle strength on the affected side. The overall confidence level of EEG intention decoding... The rate remained stable at over 88%, with only two cases of mild muscle spasms occurring during training. These spasms were quickly relieved after stimulation was paused. No adverse reactions such as skin irritation or dizziness were observed, which fully verified the effectiveness and safety of the method of the present invention.

[0025] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for controlling alternating cooperative stimuli applicable to asymmetric gait, characterized in that: Includes the following steps: S1: Collect various baseline data from the patient's healthy and affected sides; S2: The initial asymmetry is calculated based on the baseline data acquired in S1. ; S3: Initialize and configure training parameters based on the baseline data collected in S1; S4: The patient undergoes rehabilitation training, and various training data are collected in real time to control the training process in real time. S5: Calculate the phase synchronization rate based on the training data collected in S4. and cumulative asymmetry ; S6: Phase synchronization rate obtained based on S5 and cumulative asymmetry Closed-loop optimization of training parameters is performed.

2. The alternating cooperative stimulus control method for asymmetric gait according to claim 1, characterized in that: The initial asymmetry of S2 The expression is as follows: In the formula, Gait asymmetry, For the healthy side stride length, For the affected side's stride length, This indicates taking the maximum step size between the healthy and affected sides. It represents the absolute value.

3. The alternating cooperative stimulus control method for asymmetric gait according to claim 2, characterized in that: The S3 parameter initialization configuration specifically involves: setting the functional electrical stimulation intensity on the healthy side to 3mA, locking the stimulation timing to mid-support phase; setting the initial intensity of functional electrical stimulation on the affected side according to muscle strength level; and setting the timing coupling coefficient. The initial value is set to 1.

0.

4. The alternating cooperative stimulus control method for asymmetric gait according to claim 3, characterized in that: The real-time control of the training process by S4 includes the following steps: Step A1: Calculate the overall confidence level using the weighted fusion formula. Its expression is as follows: In the formula, Weighting of EEG signals For electromyographic signal weights, Confidence level for EEG intent decoding The normalized confidence value of electromyographic activation is expressed as follows: In the formula, Real-time electromyographic activation intensity, This represents the baseline value for electromyographic activation intensity. When the overall confidence level When the confidence level is ≥85%, strong stimulation is directly triggered on the affected side; when the overall confidence level is ≤60%, strong stimulation is triggered on the affected side. When <85%, if the real-time electromyographic activation intensity ≥0.3 × baseline value of electromyographic activation intensity This triggers strong stimulation on the affected side, and when the overall confidence level is... When the effect rate is less than 60%, switch to 5mA mild auxiliary stimulation. Step A2: Real-time acquisition of the healthy side gait phase using a crank phase encoder. Calculate the target phase on the affected side based on the time-series coupling formula With the timing of stimulus triggering The relevant expressions are as follows: In the formula, This is the gait phase compensation value. These are the timing coupling coefficients. For real-time gait cycles; When the initial asymmetry When >30%, gait phase compensation value =30°, when 20% ≤ initial asymmetry When ≤30%, gait phase compensation value =20°, when the initial asymmetry is... When <20%, gait phase compensation value =10°; Step A3: Monitor the electromyographic signal on the affected side in real time, and determine the intensity of real-time electromyographic activation. ≥2.0 × baseline value of electromyographic activation intensity If muscle spasm is detected, functional electrical stimulation on the affected side is immediately cut off and a voice prompt is given. The signal is reacquired after a 5-second interval. If the real-time electromyographic activation intensity is... <1.5 × baseline value of electromyographic activation intensity Then, mild auxiliary stimulation will be restored.

5. The alternating cooperative stimulus control method for asymmetric gait according to claim 4, characterized in that: The phase synchronization rate is calculated by S5. and cumulative asymmetry The relevant expressions are as follows: In the formula, For phase synchronization rate, The standard deviation of the bilateral phase difference over 5 gait cycles. To accumulate asymmetry, For the first Asymmetry of each gait cycle, This indicates a summation operation over 5 periods.

6. The alternating cooperative stimulus control method for asymmetric gait according to claim 5, characterized in that: The S6 method for closed-loop optimization of training parameters specifically involves: if the cumulative asymmetry... If the rate is >20%, the functional electrical stimulation intensity on the affected side increases by 1 mA, not exceeding 20 mA; if the phase synchronization rate is >20%, then the intensity increases by 1 mA. If the value is greater than 0.85, then the timing coupling coefficient is... The threshold should be reduced by 0.1, but not lower than 0.6; if the muscle strength on the affected side increases by 1 grade, the EEG intention confidence threshold should be increased by 5%, but not exceeding 95%; if the muscle synergistic contraction rate is >40%, the functional electrical stimulation intensity on the affected side should be reduced by 2 mA, but not lower than 8 mA.

Images