Multi-module collaborative rehabilitation electrical stimulation acupoint self-adaptive positioning method

Abstract

The invention provides a multi-module collaborative rehabilitation electrical stimulation acupoint self-adaptive positioning method, and relates to the technical field of rehabilitation medicines.The multi-module collaborative rehabilitation electrical stimulation acupoint self-adaptive positioning method comprises an electrical stimulation module, an electrode module, a myoelectricity collection module and a back-end algorithm processing module, and through collaborative work of the electrical stimulation module, the electrode module, the myoelectricity collection module and the back-end algorithm processing module, the rehabilitation electrical stimulation acupoint self-adaptive positioning is achieved; time waste of traditional treatment is reduced, and the treatment efficiency is remarkably improved; the back-end algorithm processing module performs action classification processing on the electromyographic signals, accurately positions stimulation points and enhances treatment pertinence; internal association with the physical condition is mastered through deep learning of stimulation parameters, and personalized treatment scheme customization is achieved; a control module of the electrical stimulation module automatically adjusts electrical stimulation parameters, the treatment effect is guaranteed, and safety and comfort are improved; the electromyographic acquisition module acquires electromyographic signals in a stimulation gap in a time-sharing manner, interference is avoided, accurate data is provided for back-end processing, and the reliability and scientificity of the positioning method are guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of rehabilitation medical technology, and in particular to a multi-module coordinated rehabilitation electrical stimulation acupoint adaptive positioning method. Background Art

[0002] With the continuous development and in-depth research of science and technology, rehabilitation electrical stimulation technology has played a key role in the rehabilitation treatment of patients with limb disabilities. This technology stimulates nerves to discharge to varying degrees through external discharges at specific parts of the limbs of patients who have lost their functions, thereby stimulating the limbs to make corresponding movements and, to a certain extent, rebuilding the lost functions.

[0003] However, in the current use of electrical stimulation, the search for specific effective positions on the limbs has always relied on the experience of practitioners, who have been trying again and again to find the best stimulation points. However, under the dual uncertainties of stimulation voltage parameters and stimulation positions, it is often very difficult to find effective stimulation points. Often, the best time for rehabilitation is lost in multiple explorations, and the patient is brought physical pain and psychological harm. Obviously, the current practice of rehabilitation electrical stimulation has many disadvantages and hidden dangers, and cannot effectively meet the rehabilitation needs of patients. Summary of the invention

[0004] The purpose of the present invention is to provide a multi-module collaborative rehabilitation electrical stimulation acupoint adaptive positioning method, which solves the problem that it is very difficult to find effective stimulation points, and the best rehabilitation opportunity is often lost in multiple explorations, while causing physical pain and psychological harm to patients. Obviously, the current rehabilitation electrical stimulation practices have many drawbacks and hidden dangers, and cannot efficiently meet the technical effects of patients' rehabilitation needs.

[0005] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0006] A multi-module collaborative rehabilitation electrical stimulation acupoint adaptive positioning method includes an electrical stimulation module, an electrode module, an electromyography acquisition module and a back-end algorithm processing module.

[0007] As an improvement, the electrical stimulation module is composed of a control module and a stimulation current generating module, wherein the main control module is the main control center for adjusting the current output parameters and feedback data of the electrical stimulation system, and different types of processors can be selected through different algorithm tasks, and the output stimulation current can be automatically adjusted and parameterized.

[0008] As an improvement, the electrode module is composed of an electrical stimulation array and an electromyography acquisition array. The electrical stimulation array is mainly close to the stimulation area and releases the stimulation current after adjusting the parameters.

[0009] As an improvement, the electromyographic acquisition module collects electromyographic signals during the intervals between stimulation actions. Effective stimulation will produce electromyographic action features with more obvious characteristics. The sending of electrical stimulation signals and the collection of electromyographic signals are performed at time intervals.

[0010] As an improvement, the back-end algorithm processing module: performs motion classification processing on the collected electromyographic signals to find the best stimulation points, and obtains the intrinsic correlation between effective stimulation and physical conditions of different patients through deep learning of stimulation parameters, thereby guiding the control module to fine-tune parameters.

[0011] The beneficial effects of the present invention are as follows: through the coordinated work of electrical stimulation, electrodes, electromyography acquisition and back-end algorithm processing modules, the time waste of traditional treatment is reduced and the treatment efficiency is significantly improved; the back-end algorithm processing module performs action classification processing on the electromyography signal, accurately locates the stimulation point, and enhances the targeted treatment; through deep learning of stimulation parameters, the intrinsic relationship with the physical condition is grasped to achieve personalized treatment plan customization; the control module of the electrical stimulation module automatically adjusts the electrical stimulation parameters to ensure the treatment effect and improve safety and comfort; the electromyography acquisition module collects electromyography signals in time-sharing during the stimulation interval to avoid interference, provide accurate data for back-end processing, and ensure the reliability and scientificity of the positioning method. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A block diagram of a multi-module collaborative rehabilitation electrical stimulation acupoint adaptive positioning method of the present invention;

[0013] Figure 2 This is a system block diagram of a time migration algorithm for a multi-module collaborative rehabilitation electrical stimulation acupoint adaptive positioning method of the present invention;

[0014] Figure 3 This is a block diagram of the action classification and discrimination system of a multi-module collaborative rehabilitation electrical stimulation acupoint adaptive positioning method of the present invention;

[0015] Figure 4 The present invention provides a block diagram of a system for determining an adaptive electrical stimulation position in a multi-module collaborative rehabilitation electrical stimulation acupoint adaptive positioning method. DETAILED DESCRIPTION

[0016] like Figure 1As shown, a multi-module collaborative rehabilitation electrical stimulation acupoint adaptive positioning method includes an electrical stimulation module, an electrode module, an electromyography acquisition module and a back-end algorithm processing module. The electrical stimulation module is composed of a control module and a stimulation current generation module, wherein the main control module is the main control center for adjusting the current output parameters and feedback data of the electrical stimulation system. Different types of processors can be selected through different algorithm tasks, and the output stimulation current is automatically adjusted to set parameters; the electrode module is composed of an electrical stimulation stimulation array and an electromyography acquisition array. The electrical stimulation array is mainly close to the stimulation area and releases the stimulation current after adjusting the parameters; the electromyography acquisition module collects electromyography signals in the gaps between stimulation actions. Effective stimulation will produce electromyography action features with more obvious characteristics. The issuance of electrical stimulation signals and the collection of electromyography signals are performed at different time intervals; the back-end algorithm processing module is used to classify the collected electromyography signals to find the best stimulation point, and through deep learning of the stimulation parameters, the intrinsic correlation between effective stimulation and physical conditions of different patients is obtained, thereby guiding the control module to fine-tune the parameters.

[0017] When using, the first step is to mix the stimulation electrode array and the electromyography collection electrode array in a ratio of 1:4 to make a complete matrix electrode, each stimulation electrode is surrounded by 4 collection electrodes. Attach the matching electrode array to the approximate part of the stimulation.

[0018] Step 2: Start the control module and the algorithm processing module, set the working mode of the control module and the adaptive output current according to the database imported by the algorithm processing module, adjust the scanning mode of the electrode to line by line or interlaced scanning, and collect the electromyographic signals during the period when the output power of the stimulation current is zero.

[0019] Step 3: Start the electrical stimulation module to preheat the circuit.

[0020] Step 4: Start the stimulation output in the control module, and start the real-time processing system of the electromyographic acquisition signal to synchronously process the acquired electromyographic signals.

[0021] Step 5: Mark the stimulation positions and stimulation parameters with stronger limb signal characteristics.

[0022] Step 6: Start effective electrical stimulation.

[0023] Step 7: All collected stimulation positions and stimulation parameters are used through deep learning, classification and transfer learning to obtain a database of effective, specific actions and different patients at different times.

[0024] Positioning system design Figure 4 As shown:

[0025] In the main control circuit, the initial output current parameters are set as follows: frequency: 40Hz, pulse width: 400us, output voltage set to about 15v, output current is 14mA rectangular DC pulse; duty cycle set to 40%; after the voltage output is stable, the voltage is increased to 24v, and the output current is set to 22mA; then within 3s, the output voltage gradually stabilizes to frequency: 15Hz, voltage: 30v, and current is 28mA. (Under this condition, the electrode sheet has good contact with the human body)

[0026] The electrode sheet composed of the electrical stimulation electrode and the electromyographic signal collection electrode should be in a ratio of 1:4, with the electrical stimulation electrode in the middle and the electromyographic collection electrodes around. The center distance between the two stimulation electrodes is about 10mm; the number of electrodes is set according to the number of channels of the electrical stimulation module. In this example, the number of channels is set according to the 32-channel stimulation mode, so the electromyographic collection circuit is 128 channels.

[0027] like Figure 2 As shown:

[0028] Data migration algorithm: Two types of migration calculations are required, time migration and crowd migration.

[0029] In the time migration algorithm: in a detected time series In a sample of length l and dimension v Where K represents the dimension label of the data, so that a prediction model M can be obtained, in which the data Make a Δt-step prediction to predict the ball’s label

[0030] In Timing Drift, set the total length of the timing: When i=0, Lth0=0 is used to align the timing queue. k}middle When the dimensions of the previous and next data change, the time series drifts.

[0031] like Figure 3 As shown:

[0032] During the intervals between the discharges of the stimulation electrodes, the electromyographic signals are collected and the collected motion data are put into the classification algorithm for motion classification. In the classification algorithm: Assume that the LDA linear model is used, and the training set is the feature set D = {D1,...,D n}, the action is the training action of the patient during rehabilitation, and the electromyographic data obtained after electrical stimulation is P = {P1, ..., P m}, the sample labels are calibrated by time, the moving window of data capture uses superposition sliding, and the overlap rate is set to 30% for data clipping. Two-class classification is performed, and when the classification accuracy is less than 10%, it is considered that the electrical stimulation has achieved a good effect. (Because the normal action in the setting belongs to the action of the limbs without stimulation, after being stimulated, the gesture will change greatly, and the electromyographic action will be the same as before stimulation. Otherwise, the limb movement is the same as the normal movement, and the stimulation is ineffective).

[0033] In the electrical stimulator module, the average 1.5kΩ is calculated based on the human skin resistance. The voltage adjustment range of this experiment is set to: 5V-50V rectangular DC voltage, the frequency is set to an adjustable range of 10Hz-100Hz, and the output current is set to 5mA-20mA. The discharge channel of the electrical stimulator is designed with 32 channels, and at most one channel can be set as a discharge channel at the same time. The discharge sequence of the electrode sheet adopts interlaced scanning output to ensure the minimum interference when collecting electromyographic signals. In the acquisition module, in order to reduce residual electricity and power supply harmonic interference, when designing the stimulation voltage, a rectangular DC pulse with a duty cycle of more than 50% should be used.

[0034] After the whole system is running, it will continuously scan the part to be tested. When the action classification accuracy is less than 10%, the voltage parameters of the electrical stimulation will be saved in the database. At the same time, the stimulation part is the core data that the system needs to obtain and will be recorded. According to different rehabilitation processes, the stimulation voltage is continuously adjusted to obtain the best stimulation effect and stimulation part.

[0035] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A multi-module collaborative rehabilitation electrical stimulation acupoint adaptive positioning method, characterized in that: It includes an electrical stimulation module, an electrode module, an electromyography acquisition module and a back-end algorithm processing module.

2. A multi-module collaborative rehabilitation electrical stimulation acupoint adaptive positioning method according to claim 1, characterized in that: The electrical stimulation module is composed of a control module and a stimulation current generating module, wherein the main control module is the main control center for adjusting the current output parameters and feedback data of the electrical stimulation system. Different types of processors can be selected through different algorithm tasks, and the output stimulation current can be automatically adjusted and parameterized.

3. A multi-module collaborative rehabilitation electrical stimulation acupoint adaptive positioning method according to claim 1, characterized in that: The electrode module is composed of an electrical stimulation array and an electromyography collection array. The electrical stimulation array is mainly placed in close contact with the stimulation area to release stimulation current after adjusting parameters.

4. A multi-module collaborative rehabilitation electrical stimulation acupoint adaptive positioning method according to claim 1, characterized in that: The electromyographic acquisition module collects electromyographic signals in the intervals between stimulation actions. Effective stimulation will produce electromyographic action features with relatively obvious characteristics. The sending of electrical stimulation signals and the collection of electromyographic signals are performed at different time intervals.

5. The multi-module coordinated rehabilitation electrical stimulation acupoint adaptive positioning method according to claim 1, characterized in that: The back-end algorithm processing module: performs action classification processing on the collected electromyographic signals to find the best stimulation points, and obtains the intrinsic correlation between effective stimulation and physical conditions of different patients through deep learning of stimulation parameters, thereby guiding the control module to perform parameter fine-tuning.

Images