A channel multiplexing type electromyography-electric stimulation integrated active rehabilitation device
By designing a channel-reuse electromyography-electrical stimulation integrated active rehabilitation device, the existing electrical stimulation physiotherapy instruments are solved, and high real-time and low-cost electromyography feedback is achieved, and the equipment is improved. versatility and patient comfort.
Patent Information
- Application Number
- CN202311443655.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-11-02
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2043-11-02
AI Technical Summary
The existing electrical stimulation physiotherapy instruments have problems such as insufficient treatment effect, inconvenient use, low signal-to-noise ratio and poor expansion, especially the lack of integrated systems with low hardware noise, real-time electromyography feedback function, strong versatility, and the cost of rehabilitation algorithms. High, human-computer interaction is complex.
A channel multiplexed electromyography-electrical stimulation integrated active rehabilitation device is designed, adopting a architecture of multiple device ends and one receiving end. Each device end has four electrode channels, and the channel is expanded through the channel selection circuit and the CAN bus interface. The number and function of the contralateral closed-loop motion intention is realized through real-time acquisition of electromyography signals and closed-loop control of electrical stimulation, reducing noise, and fusing the patient's subjective motion intention through the contralateral closed-loop control system algorithm.
The channel multiplexed electrotherapy instrument hardware with real-time feedback reduces electromyography biofeedback noise, improves the real-time and versatility of the system, slows down patients' muscle fatigue, extends treatment time, and improves the wearable of the device. Sex and ease of use.
Smart Images

Figure CN117563131B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of electrical stimulation therapy apparatus, and in particular to a channel multiplexing type myoelectric-electric stimulation integrated active rehabilitation device. Background Art
[0002] Functional electrical stimulation (FES) is a technology that uses low-frequency pulse current to stimulate nerves and muscles to induce or simulate normal motor function. It can not only improve or restore the function of limbs and organs, but also promote the remodeling of the human motor neural network and improve the plasticity of the central nervous system. It can be used to treat human motor dysfunction caused by diseases such as stroke.
[0003] Surface electromyography (sEMG) can be used to evaluate muscle function. By analyzing the characteristics of sEMG signals, we can understand the contraction, fatigue level, coordination, etc. of muscles. It is very helpful for evaluating muscle function, formulating rehabilitation plans, and adjusting training intensity. By analyzing the changes in sEMG signals, we can directly understand the activation of the patient's muscle groups and reveal the intensity of the patient's exercise intention.
[0004] Simple open-loop FES motor rehabilitation functions all belong to continuous passive motion (CPM) rehabilitation, which does not include the feedback link of detecting the patient's subjective motor intention. In recent years, a large number of studies have shown that passive motor rehabilitation that integrates active motor intention will produce better neuromuscular rehabilitation effects than simple CPM. Related studies have also discussed in depth the insufficient rehabilitation effect of CPM and the fact that the motor rehabilitation controller that incorporates the patient's subjective intention can significantly improve the effect of patient rehabilitation. It can be seen that for future motor rehabilitation equipment, whether it is a medical rehabilitation robot or an electrical physiotherapy device, the combination of a controller that integrates the patient's subjective motor intention and passive collaborative motor rehabilitation hardware must be one of the mainstream research directions at this stage. However, as a non-invasive brain-computer interface technology that can directly extract motor intentions, it still has the following defects: the relevant technology is not yet mature, and has the disadvantages of high cost, difficulty in development, inconvenience for patients to use, and difficulty in clinical popularization.
[0005] Currently, most of the electrical stimulation therapy devices on the market are open-loop output devices and do not include electromyography acquisition feedback. The disadvantage is that the accuracy of the electrical stimulation settings is difficult to guarantee.
[0006] Some electrical physiotherapy devices that include electromyography assistance generally lead out the electromyography acquisition electrodes additionally. Because at least two pairs of electrodes must be placed in each muscle group, the electrode density is too high when performing collaborative rehabilitation of multiple muscle groups, and the patient will feel uncomfortable. Secondly, because electromyography acquisition is not isolated from electrical stimulation in hardware, due to problems such as amplifier saturation, the collected electromyography signals contain a lot of noise, and filtering algorithms are often required to filter the electromyography signals, which reduces the real-time performance of the system while also increasing costs.
[0007] At the same time, almost all electrotherapy devices on the market adopt a closed-source ecosystem, which is insufficient in terms of functional expandability and channel number expandability. The limited number of channels makes it impossible for the device to provide a coordinated exercise rehabilitation mode for multiple joints and multiple muscle groups; the lack of necessary open interfaces makes it impossible for the electrotherapy device to perform coordinated rehabilitation control with other rehabilitation equipment such as brain-computer interfaces and medical exoskeletons, which in turn leads to insufficient versatility of the device.
[0008] Simple open-loop FES exercise rehabilitation functions are all continuous passive motion (CPM) rehabilitation, which does not include a feedback link to detect the patient's subjective movement intention. Its disadvantage is that allowing the patient to receive passive electrical stimulation may cause the patient's muscles to produce antagonism, which in turn causes the muscles to quickly enter a state of fatigue, reducing the treatment effect. The principle of open-loop electrical stimulation treatment can be seen Figure 1 As shown, it is equivalent to passively activating most of the motor units and part of the sensory feedback nerves of the patient's muscles.
[0009] Most existing electrical physiotherapy devices require patients to input and adjust the electrical stimulation parameters by themselves. Only after repeated adjustments can the final appropriate electrical stimulation parameters be obtained. The human-computer interaction is complex and unfriendly. At the same time, due to the patient's misoperation, the agonist and antagonist muscles of a joint may be stimulated at the same time, which will cause the patient's muscle group to be treated to quickly enter a fatigue state, greatly reducing the rehabilitation effect.
[0010] At the same time, for the existing ipsilateral electromyographic-electrical stimulation equipment, because the electromyographic signal collection and electrical stimulation are performed simultaneously on the affected limb, a large amount of noise will be introduced into the electromyographic collection, making it difficult to truly implement a low-cost closed-loop control solution.
[0011] In summary, for an electrotherapy device used to treat movement dysfunction, there is currently a lack of hardware that has low noise, real-time electromyography feedback function, and strong versatility; there is also a lack of rehabilitation algorithm solutions that are low-cost, can fully integrate the patient's subjective movement intentions, and have simple human-computer interaction. Integrated system products. Summary of the invention
[0012] The purpose of the present invention is to solve the problems in the prior art and propose a channel multiplexing type myoelectric-electric stimulation integrated active rehabilitation device to solve the problems of insufficient treatment effect, inconvenient use, low signal-to-noise ratio and poor scalability of the existing electrical physiotherapy devices.
[0013] The present invention is realized by the following technical scheme. The present invention proposes a channel multiplexing type electromyography-electric stimulation integrated active rehabilitation device, which is composed of multiple device ends and a receiving end; the device end is a device unit that can independently provide electromyography acquisition and electrical stimulation, each device end has four electrode channels, and each channel can be configured as a sEMG acquisition function or a FES function through a channel selection circuit; at the same time, the device end is provided with a CAN bus interface for connecting multiple devices to expand the number of channels and the functional range; each device end is provided with an ESP32 for sending instructions and data to a host computer using wifi / Bluetooth; the receiving end serves as the host computer of the device, which has the functions of real-time monitoring of sEMG signals and electrical stimulation parameters of each port and issuing instructions to adjust the electrical stimulation parameters of the device end.
[0014] Furthermore, the channel selection circuit includes a universal electrode, a relay, a channel switching circuit, an electromyography acquisition circuit, a functional electrical stimulation circuit, a power management circuit and a main control circuit; the relay has two groups of independent switches, and each group of switches has a single-pole double-throw selection characteristic; the electrode seat is respectively connected to the electromyography acquisition circuit and the functional electrical stimulation circuit through the single-pole double-throw switch of the relay; the control end of the relay is connected to the IO pin of the main control through the main control circuit.
[0015] Furthermore, the channel switching circuit is composed of an electrode holder, a photorelay, a triode and a freewheeling diode for preventing sudden changes in voltage direction; the channel switching is controlled by the GPIO of the main control, the channel is connected to the electromyography acquisition circuit by default, and the channel function can be switched within 3ms after the control command is issued; through the channel switching circuit, the channel for the electromyography signal to be collected is isolated from the electrical stimulation channel, effectively reducing the noise introduced into the electromyography acquisition channel by the electrical stimulation.
[0016] Furthermore, the device end is an independent device that can provide sports rehabilitation function, which includes an electromyography-electric stimulation integrated function board and a main control core board;
[0017] The electromyography-electric stimulation integrated function board is used to collect surface electromyography signals sEMG and perform functional electrical stimulation FES; the surface electromyography signal collection function includes extraction, amplification and bandpass analog signal filtering of electromyography signals; the functional electrical stimulation function refers to the use of the designed voltage-controlled constant current source to generate bipolar stimulation waves, which directly act on the diseased muscle groups of the human body for sports rehabilitation treatment;
[0018] The main control core board is used to collect the electromyographic analog signal conditioned by the electromyographic-electric stimulation function board and control the waveform, frequency and stimulation intensity of the electric stimulation wave; the main control core board will run the contralateral closed-loop control system algorithm to guide the patient to perform sports rehabilitation training. The main control core board also includes a wifi / bluetooth module, and its main control is an ESP32C3 chip. The module is used to communicate with the receiving end device, and is used to send electromyographic signals and device status data to the receiving end device and receive instructions from the receiving end; and the main control core board has a CAN bus communication module, which can realize communication between multiple modules.
[0019] Furthermore, the contralateral closed-loop control system algorithm comprises the following steps:
[0020] Step 1: Bridge the healthy limbs of the human body with the limbs with motor dysfunction and ensure that the corresponding muscle groups are one-to-one corresponding;
[0021] Step 2: Contralateral closed-loop exercise rehabilitation begins;
[0022] Step 3: The patient applies force to both sides of his upper limbs and performs a certain movement of the upper limbs at the same time, maintaining the same movement pattern, and only observing the movement status of his healthy limb;
[0023] Step 4: Use the contralateral closed-loop motor rehabilitation algorithm to convert the electromyographic signal of the healthy limb into the electrical stimulation-assisted motion parameters of the affected limb, so that the affected limb of the patient can follow the motion pattern of the healthy limb;
[0024] Step 5: After the movement is restored, the rehabilitation equipment stops working for 10 seconds and the patient rests for 10 seconds;
[0025] Step 6: If the patient does not feel muscle fatigue, return to step 3 and start the cycle, otherwise end.
[0026] Furthermore, the specific steps of the contralateral closed-loop exercise rehabilitation algorithm in step 4 are:
[0027] I. A device on the healthy limb collects 20ms electromyographic signals on 4 channels;
[0028] II. Calculate the root mean square of the electromyographic signal in a 20 ms window on each channel;
[0029] III, classify healthy limb movements through a neural network action classifier;
[0030] IV. After classification, the corresponding electrical stimulation parameters are calculated using the biomechanical model as feedforward electrical stimulation parameters;
[0031] V. Call the time-division multiplexing electrical stimulation algorithm to stimulate for a period of time, disconnect the optocoupler after stimulation, and wait for 10ms to calm the electrical stimulation noise;
[0032] VI. The channels of the device end on the healthy limb and the device end on the affected limb are switched to the electromyographic acquisition mode to collect 10ms electromyographic signals;
[0033] VII. Calculate the root mean square of the electromyographic signal within 10 ms;
[0034] VIII. Using the PD controller, the difference between the root mean square values in step I and step VI is used as input to further adjust the electrical stimulation parameters;
[0035] IX. Repeat steps V-VIII before completing this rehabilitation exercise.
[0036] Furthermore, the time-division multiplexing electrical stimulation algorithm is specifically:
[0037] ①. According to the previous motion classification model, obtain the sequence number of the channels to be stimulated that need to stimulate the muscle groups simultaneously;
[0038] ②. Put the serial number into the queue to be stimulated according to the priority;
[0039] ③. Select the first corresponding channel in the queue and perform electrical stimulation for a certain period of time according to the electrical stimulation parameters calculated by the model;
[0040] ④. Pop up the first element of the queue. If the queue is empty, proceed to the next step, otherwise repeat steps ③-④;
[0041] ⑤. After a delay of 10ms, the device collects electromyographic signals from all channels, and the collection window size is 10ms;
[0042] ⑥. Before completing a rehabilitation exercise, repeat steps ②-⑤.
[0043] The present invention has the following beneficial effects:
[0044] (1) The present invention designs a channel multiplexing type electrical therapy instrument hardware architecture for real-time feedback, which has the advantages of low noise and high real-time performance of electromyographic biofeedback, and can effectively save the number of electrodes and layout space required when achieving the same electromyographic feedback function. At the same time, through the CAN bus interface of the device, it is relatively easy to achieve coordinated control between devices, and it can also achieve coordinated control of the device with other rehabilitation devices (such as exoskeleton robots, etc.), which has strong scalability and versatility;
[0045] (2) The myoelectric acquisition circuit and the electrical stimulation circuit of the present invention are simple and reliable, and have sufficient medical safety considerations; the circuit has high circuit integration and low cost, which is convenient for clinical mass popularization;
[0046] (3) The present invention also makes relevant designs to improve the wearability and usability of the device. While ensuring that the device is compact, a wearable fabric is also designed to indicate the electrode placement position. If the patient does not place the electrodes correctly, it will be difficult for the patient to wear the designed fabric. Therefore, it can effectively prevent the patient from causing muscle fatigue and damage due to incorrect electrode placement.
[0047] (4) The contralateral closed-loop control system of the present invention includes a feedforward link and a feedback link, which ensures that the system has excellent rapidity and accuracy, can effectively reduce the patient's muscle fatigue, and prolong the treatment time.
[0048] (5) The closed-loop contralateral control scheme used ensures the ability to extract the patient's subjective movement intention in real time; at the same time, the motion classifier in the feedforward link can detect the patient's limb movement, avoiding the conflict between the patient's subjective movement trend and the passive movement of the electrical therapy device, greatly improving comfort and treatment effects. BRIEF DESCRIPTION OF THE DRAWINGS
[0049] Figure 1 Schematic diagram of the principle of existing open-loop electrical stimulation technology.
[0050] Figure 2 Schematic diagram of the system hardware architecture of Example 1 of the present invention.
[0051] Figure 3 Schematic diagram of a channel switching circuit according to embodiment 2 of the present invention.
[0052] Figure 4 This is a schematic diagram of the use of the rehabilitation device of Example 3 of the present invention.
[0053] Figure 5 This is a control system block diagram of the opposite-side closed-loop control system of Example 4 of the present invention.
[0054] Figure 6 This is a schematic diagram of the improved Harmmerstein feedforward model according to Example 5 of the present invention.
[0055] Figure 7 Flow chart of contralateral active movement rehabilitation of the present invention.
[0056] Figure 8 This is a flow chart of the closed-loop control algorithm of the contralateral electromyographic stimulation of the present invention.
[0057] Fig. 9 This is a schematic diagram of the host computer interface. DETAILED DESCRIPTION
[0058] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0059] Example 1
[0060] See also Figure 2 The present invention proposes a channel multiplexing type electromyography-electric stimulation integrated active rehabilitation device, which is composed of multiple device ends and a receiving end; the device end is a device unit that can independently provide electromyography acquisition and electrical stimulation, each device end has four electrode channels, and each channel can be configured as sEMG acquisition function or FES function through a channel selection circuit; at the same time, the device end is provided with a CAN bus interface for connecting multiple devices to expand the number of channels and the range of functions; by using CAN bus technology, the pain points of traditional electrical stimulation equipment, such as the fixed number of channels, difficulty in coordinated control with other devices, and poor scalability, are solved, so that the device has excellent versatility; each device end is provided with an ESP32 for sending instructions and data to the upper computer using wifi / Bluetooth; the receiving end serves as the upper computer of the device, which has the functions of real-time monitoring of the sEMG signals and electrical stimulation parameters of each port and issuing instructions to adjust the electrical stimulation parameters of the device end.
[0061] The device end is an independent device that can provide sports rehabilitation function, which includes an electromyography-electric stimulation integrated function board and a main control core board; for scenarios and functional applications that require different computing power, the device cost and performance can be adjusted by replacing MCUs with different computing power.
[0062] The electromyography-electric stimulation integrated function board is used to collect surface electromyography signals sEMG and perform functional electrical stimulation FES; the surface electromyography signal collection function includes extraction, amplification and bandpass analog signal filtering of electromyography signals; the functional electrical stimulation function refers to the use of the designed voltage-controlled constant current source to generate bipolar stimulation waves, which directly act on the diseased muscle groups of the human body for sports rehabilitation treatment;
[0063] The main control core board is used to collect the electromyographic analog signal conditioned by the electromyographic-electric stimulation function board and control the waveform, frequency and stimulation intensity of the electric stimulation wave; the main control core board will run the contralateral closed-loop control system algorithm to guide the patient to perform sports rehabilitation training. The main control core board also includes a wifi / bluetooth module, and its main control is an ESP32C3 chip. The module is used to communicate with the receiving end device, and is used to send electromyographic signals and device status data to the receiving end device and receive instructions from the receiving end; and the main control core board has a CAN bus communication module, which can realize communication between multiple modules.
[0064] The receiving end actually refers to the host computer containing the wifi / bluetooth receiving device, which is generally a PC host or a Linux single board. The function of the receiving end is to monitor the status of each device end in real time and receive electromyographic signal data; it can also be used as a human-computer interaction interface to send relevant instructions such as adjusting electrical stimulation parameters to the device end.
[0065] Example 2
[0066] The channel selection circuit includes a universal electrode, a relay, a channel switching circuit, an electromyography acquisition circuit, a functional electrical stimulation circuit, a power management circuit and a main control circuit; the relay has two groups of independent switches, and each group of switches has a single-pole double-throw selection characteristic; the electrode seat is respectively connected to the electromyography acquisition circuit and the functional electrical stimulation circuit through the single-pole double-throw switch of the relay; the control end of the relay is connected to the IO pin of the main control through the main control circuit.
[0067] The channel switching circuit is composed of an electrode holder, a photorelay, a triode and a freewheeling diode for preventing sudden changes in voltage direction; the channel switching is controlled by the GPIO of the main control, the channel is connected to the electromyography acquisition circuit by default, and the channel function can be switched within 3ms after the control command is issued; through the channel switching circuit, the channel for the electromyography signal to be collected is isolated from the electrical stimulation channel, effectively reducing the noise introduced into the electromyography acquisition channel by the electrical stimulation.
[0068] See Figure 3, this embodiment provides a function switching circuit for electrode channels, which is the core of realizing the channel multiplexing function. The circuit includes a relay K1, a diode D1, resistors R1, R2 and R3, and a transistor Q1. The single-pole double-throw switch of the relay K1 is respectively connected to the electromyography acquisition and electrical stimulation circuits, and the other end of the switch is connected to the electrode interface. The diode D1 is a freewheeling diode, which is connected in series with the resistor R1 and then connected in parallel with the relay K1 to quickly reduce the back electromotive force generated when the relay switches, so that the relay can quickly switch channels. The transistor and the resistors R2 and R3 together constitute the control circuit of the relay. The pin of the single-chip microcomputer is directly connected in series with the resistor R2 and then connected to the base of the transistor Q1, so that the selected channel of the relay can be directly controlled. By default, when the microcontroller does not output a high-level control signal, the relay switch connects the electrode holder to the electromyography acquisition circuit, that is, CH1+ is connected to SEMG1+, and CH1- is connected to SEMG-; after the microcontroller outputs a high-level control signal, the relay switch connects the electrode holder to the electrical stimulation circuit in 3ms, that is, CH1+ is connected to FES+, and CH1- is connected to FES-, completing the channel switching.
[0069] Example 3
[0070] See Figure 4 , this embodiment provides a schematic diagram of the use of a contralateral closed-loop exercise rehabilitation device, which includes a wearable fabric 1 inlaid with an electrode interface, a rehabilitation device 2, and an electrode connecting wire 3. When in use, first stick the electrodes on the diseased muscle group to be rehabilitated and the corresponding healthy muscle group respectively and wear the wearable fabric 1; then connect the electrode wires of the corresponding muscle group to the rehabilitation device; then you can start the exercise rehabilitation training. The wearable fabric 1 inlaid with an electrode interface is actually to fix the position of the electrode connection port during weaving, and the channel to which the electrode belongs is marked on the fabric, which is convenient for patients to quickly wear the device and verify that there will be no large deviation in the treated site.
[0071] Example 4
[0072] The block diagram of the contralateral closed-loop electrical stimulation control system is shown in Figure 5 . The goal of this control system is to enable the muscle force on the affected limb to track the muscle force on the corresponding muscle group of the healthy limb, so that the healthy and affected limbs can achieve synchronous movements. On a macro level, the intensity of the patient's active movement intention is extracted through the electromyographic signal of the healthy limb, and the affected limb is electrically stimulated according to the intensity. This process enables the patient to complete a closed-loop neural circuit of generating intention movement-completing movement. Now some links of the control system are introduced: After extracting the electromyographic signal of the healthy limb, the multiplier is adjusted to K, where 0<K≤1. This link is mainly considered that the maximum muscle force that patients with different degrees of illness can generate is limited. Therefore, in order to improve the effect of training, the proportion of tracking muscle force is adjusted here.
[0073] The feedforward link in this system includes a neural network motion classifier and a biomechanical model feedforward. The neural network motion classifier is mainly used to identify the patient's healthy limb movements to avoid the auxiliary movement generated by subsequent electrical stimulation from antagonizing the patient's active movement intention. Biomechanics is used to quickly calculate the required electrical stimulation intensity parameters so that the affected limb can quickly track the healthy limb movement. The feedback link of this system mainly includes a PD controller, which is used to quickly correct the movement when there is still a certain error between the healthy and affected limbs after the feedforward link, thereby improving the consistency between the healthy and affected limb movements.
[0074] Example 5
[0075] In the feedforward channel, the biomechanical model used is as follows Figure 6 The improved Harmmerstein model has the advantages of high accuracy and convenient calculation. Because the amplitude of myoelectricity is directly positively correlated with the strength of muscle contraction, the relationship between the myoelectricity signal and the strength of the electrical stimulation parameters can be directly established to quickly configure the electrical stimulation parameters. The Hammerstein model is essentially a biomechanical model that establishes the strength of electrical stimulation to the force generated by the muscle. The improved Hammerstein model process of the present invention includes three links: hysteresis nonlinear link, second-order linear link and time lag link.
[0076] The Hammerstein nonlinear link is generally fitted using a polynomial. Previous experiments have found that the change in the EMG value corresponding to the increase and decrease of the electrical stimulation intensity will produce a certain offset. Therefore, the average EMG value corresponding to the increase and decrease of the electrical stimulation intensity can be taken as the final electrical stimulation intensity-EMG relationship. The expression is shown in formula (1).
[0077]
[0078] Where f1(u) and f2(u) respectively represent the changes in the nonlinear EMG value when the electrical stimulation intensity increases and decreases, and u represents the electrical stimulation intensity (generally the electrical stimulation voltage, current or pulse width, and in the present invention, u represents the electrical stimulation current). Preliminary experiments have shown that when q=3, that is, a cubic polynomial is used to model the effect is better.
[0079] The subsequent linear link and time lag link determine the subsequent movement of the entire joint system. The formula of the linear link is shown in formula (2). Generally, n=2, that is, a second-order system is used to model the muscle model.
[0080]
[0081] The time lag link reflects the link that the joint can only move after a period of electrical stimulation. The model of this link is shown in formula (3).
[0082] T Dealy =e -τs (3)
[0083] τ is the delay time constant. It has been confirmed experimentally that the reaction time delay of muscles after electrical stimulation is generally around 0.01 seconds. The parameter identification of the model is to obtain the key parameters using the least squares method after the experiment. After the key parameters have been obtained, the feedforward is used to compensate the model:
[0084] According to the principles of control engineering, the compensation of the nonlinear link in the feedforward link uses f -1 (sEMG), compensation of the linear link using Theoretically, it can achieve the minimum error and the fastest adjustment speed. The PD controller essentially contains a first-order differential term, and the second-order differential term feedforward will introduce system oscillation, so the calculation of the stimulus current amplitude in the final feedforward link is shown in formula (4). amp , U amp are the minimum current that enables people to move and the safe current limit that causes people to feel pain obtained in the preliminary experiment, K p It is the proportional gain factor, which is adjusted manually.
[0085]
[0086] Example 6
[0087] See Figure 7 This example introduces in detail the entire process of sports rehabilitation using the device provided by the present invention, which includes the following steps: bridging the healthy limbs of the human body with the limbs with motor dysfunction, and ensuring that the corresponding muscle groups are one-to-one corresponding as much as possible.
[0088] In terms of sports rehabilitation algorithms, the purpose of the present invention is to provide a method with strong versatility, which solves the problems of poor rehabilitation effect of traditional open-loop electrical stimulation and difficulty in use by patients, and also solves the problems of poor feedback performance and high electromyography acquisition noise of ordinary electromyography-assisted electrical stimulation on the market.
[0089] The contralateral closed-loop control system algorithm comprises the following steps:
[0090] Step 1: Bridge the healthy limbs of the human body with the limbs with motor dysfunction and ensure that the corresponding muscle groups are one-to-one corresponding;
[0091] Step 2: Contralateral closed-loop exercise rehabilitation begins;
[0092] Step 3: The patient applies force to both sides of his upper limbs and performs a certain movement of the upper limbs at the same time, maintaining the same movement pattern, and only observing the movement status of his healthy limb;
[0093] Step 4: Use the contralateral closed-loop motor rehabilitation algorithm to convert the electromyographic signal of the healthy limb into the electrical stimulation-assisted motion parameters of the affected limb, so that the affected limb of the patient can follow the motion pattern of the healthy limb;
[0094] Step 5: After the movement is restored, the rehabilitation equipment stops working for 10 seconds and the patient rests for 10 seconds;
[0095] Step 6: If the patient does not feel muscle fatigue, return to step 3 and start the cycle, otherwise end.
[0096] The bridging in step 1 refers to wearing integrated electrode fabrics on the affected muscle group on the affected limb of the patient and the muscle group on the corresponding healthy limb, and connecting the electrodes to the device of the present invention. If the number of channels is insufficient, the number of electrode channels of the device can be expanded using the CAN bus channel.
[0097] Example 7
[0098] See Figure 8 This embodiment systematically describes the contralateral electrical stimulation rehabilitation algorithm process, which is as follows:
[0099] The specific steps of the contralateral closed-loop exercise rehabilitation algorithm in step 4 are:
[0100] I. A device on the healthy limb collects 20ms electromyographic signals on 4 channels; any values with large deviations are directly deleted;
[0101] II. Calculate the root mean square of the electromyographic signal in a 20 ms window on each channel;
[0102] The calculation formula of the root mean square is shown in formula (5):
[0103]
[0104] where x i is the i-th myoelectric voltage value collected;
[0105] III, classify healthy limb movements through a neural network action classifier;
[0106] The neural network classifier used is introduced as follows:
[0107] The RMS in II is used as the input of the neural network, and the output is the one-hot encoding of the four actions. According to the characteristics of distributed computing, each independent master only processes four classifications at most;
[0108] The network has two hidden layers, and each hidden layer has 10 neurons;
[0109] The activation function of the neuron uses the ReLU function;
[0110] IV. After classification, the corresponding electrical stimulation parameters are calculated using the biomechanical model as feedforward electrical stimulation parameters;
[0111] V. Call the time-division multiplexing electrical stimulation algorithm to stimulate for a period of time, disconnect the optocoupler after stimulation, and wait for 10ms to calm the electrical stimulation noise;
[0112] VI. The channels of the device end on the healthy limb and the device end on the affected limb are switched to the electromyographic acquisition mode to collect 10ms electromyographic signals;
[0113] VII. Calculate the root mean square of the electromyographic signal within 10 ms;
[0114] VIII. Using the PD controller, the difference between the root mean square values in step I and step VI is used as input to further adjust the electrical stimulation parameters;
[0115] IX. Repeat steps V-VIII before completing this rehabilitation exercise.
[0116] The time period in step V of the algorithm is approximately in the range of 50-100 ms, the electric stimulation carrier used is between 20-50 Hz, and the duty cycle generally does not exceed 20%.
[0117] The time-division multiplexed channel multiplexing electromyographic acquisition-electric stimulation technology is related to the hardware architecture of the present invention, and its purpose is to further suppress noise and ensure that the electrical stimulation will not cause "cross-current" between multiple electrode pairs.
[0118] The time-division multiplexing electrical stimulation algorithm is specifically:
[0119] ①. According to the previous motion classification model, obtain the sequence number of the channels to be stimulated that need to stimulate the muscle groups simultaneously;
[0120] ②. Put the serial number into the queue to be stimulated according to the priority;
[0121] ③. Select the first corresponding channel in the queue and perform electrical stimulation for a certain period of time according to the electrical stimulation parameters calculated by the model;
[0122] ④. Pop up the first element of the queue. If the queue is empty, proceed to the next step, otherwise repeat steps ③-④;
[0123] ⑤. After a delay of 10ms, the device collects electromyographic signals from all channels, and the collection window size is 10ms;
[0124] ⑥. Before completing a rehabilitation exercise, repeat steps ②-⑤.
[0125] The electrical stimulation for a certain period of time in step ③ is the same as described in the previous V, which depends on the intensity of electrical stimulation required by the patient.
[0126] The priority in step ② refers to the biomechanical ranking of the contribution of each muscle involved in a specific action. Generally speaking, the muscle group that makes the main contribution is larger and requires higher-intensity electrical stimulation first.
[0127] Example 8
[0128] See Fig. 9 , this example is the corresponding host computer interface. It includes the following parts: 1 is the communication configuration area, where you can select one of the three communication methods: serial port, Bluetooth, and TCP / IP, and configure the parameters; 2 is the electrical stimulation parameter configuration area, which can be used to directly configure the electrical stimulation parameters and send them; 3 is the four-channel signal display coordinate system; 4 is the received data and information display window, and this part can also be used to record the received data in real time, which is convenient for the next step of data processing of electromyographic data. Through this host computer, the system can be debugged and used more conveniently.
Claims
1. A channel multiplexing type electromyography-electric stimulation integrated active rehabilitation device, characterized in that: The device is composed of multiple device ends and a receiving end; the device end is a device unit that can independently provide electromyographic acquisition and electrical stimulation, each device end has four electrode channels, and each channel can be configured as a surface electromyographic signal sEMG acquisition function or a functional electrical stimulation FES function through a channel selection circuit; at the same time, the device end is provided with a CAN bus interface for connecting multiple devices to expand the number of channels and the range of functions; each device end is provided with an ESP32 for sending instructions and data to a host computer using wifi / Bluetooth; the receiving end serves as the host computer of the device, and has the functions of real-time monitoring of the sEMG signals and electrical stimulation parameters of each device end and issuing instructions to adjust the electrical stimulation parameters of the device end; The device end is an independent device that can provide sports rehabilitation function, which includes an electromyography-electric stimulation integrated function board and a main control core board; The electromyography-electric stimulation integrated function board is used to collect surface electromyography signals sEMG and perform functional electrical stimulation FES; collecting surface electromyography signals includes extracting, amplifying and bandpass analog signal filtering of electromyography signals; performing functional electrical stimulation refers to using the designed voltage-controlled constant current source to generate bipolar stimulation waves, which directly act on the diseased muscle groups of the human body for sports rehabilitation treatment; The main control core board is used to collect the electromyographic analog signal conditioned by the electromyographic-electric stimulation function board and control the waveform, frequency and stimulation intensity of the electric stimulation wave; the main control core board will run the contralateral closed-loop control system algorithm to guide the patient to perform sports rehabilitation training. The main control core board also includes a wifi / bluetooth module, and its main control is an ESP32C3 chip. The module is used to communicate with the receiving device, and is used to send electromyographic signals and device status data to the receiving device and receive instructions from the receiving end; In addition, the main control core board has a CAN bus communication module, which can realize communication between multiple modules; The contralateral closed-loop control system algorithm comprises the following steps: Step 1: Bridge the healthy limbs of the human body with the limbs with motor dysfunction and ensure that the corresponding muscle groups are one-to-one corresponding; Step 2: Contralateral closed-loop exercise rehabilitation begins; Step 3: The patient applies force to both sides of his upper limbs and performs a certain movement of the upper limbs at the same time, maintaining the same movement pattern, and only observing the movement status of his healthy limb; Step 4: Use the contralateral closed-loop motor rehabilitation algorithm to convert the electromyographic signal of the healthy limb into the electrical stimulation-assisted motion parameters of the affected limb, so that the affected limb of the patient can follow the motion pattern of the healthy limb; Step 5: After the movement is restored, the rehabilitation equipment stops working for 10 seconds and the patient rests for 10 seconds; Step 6: If the patient does not feel muscle fatigue, return to step 3 to start the cycle, otherwise end; The specific steps of the contralateral closed-loop exercise rehabilitation algorithm in step 4 are: I. A device on the healthy limb collects 20ms electromyographic signals on 4 channels; II. Calculate the root mean square of the electromyographic signal in a 20 ms window on each channel; III, classify healthy limb movements through a neural network action classifier; IV. After classification, the corresponding electrical stimulation parameters are calculated using the biomechanical model as feedforward electrical stimulation parameters; V. Call the time-division multiplexing electrical stimulation algorithm to stimulate for a period of time, disconnect the optocoupler after stimulation, and wait for 10ms to calm the electrical stimulation noise; VI. The channels of the device end on the healthy limb and the device end on the affected limb are switched to the electromyographic acquisition mode to collect 10ms electromyographic signals; VII. Calculate the root mean square of the electromyographic signal within 10 ms; VIII. Using a proportional-derivative PD controller, the difference between the root mean square values in step II and step VII is used as input to further adjust the electrical stimulation parameters; IX. Repeat steps V-VIII before the end of this rehabilitation exercise. The time-division multiplexing electrical stimulation algorithm is specifically: ①. According to the previous motion classification model, obtain the sequence number of the channels to be stimulated that need to stimulate the muscle groups simultaneously; ②. Put the serial number into the queue to be stimulated according to the priority; ③. Select the first corresponding channel in the queue and perform electrical stimulation for a certain period of time according to the electrical stimulation parameters calculated by the model; ④. Pop up the first element of the queue. If the queue is empty, proceed to the next step, otherwise repeat steps ③-④; ⑤. After a delay of 10ms, the device collects electromyographic signals from all channels, and the collection window size is 10ms; ⑥. Before completing a rehabilitation exercise, repeat steps ②-⑤.
2. The device according to claim 1, characterized in that: The channel selection circuit includes a universal electrode, a relay, a channel switching circuit, an electromyography acquisition circuit, a functional electrical stimulation circuit, a power management circuit and a main control circuit; the relay has two groups of independent switches, and each group of switches has a single-pole double-throw selection characteristic; the electrode seat is respectively connected to the electromyography acquisition circuit and the functional electrical stimulation circuit through the single-pole double-throw switch of the relay; the control end of the relay is connected to the IO pin of the main control through the main control circuit.
3. The device according to claim 2, characterized in that: The channel switching circuit is composed of an electrode holder, a photorelay, a triode and a freewheeling diode for preventing sudden changes in voltage direction; the channel switching is controlled by the GPIO of the main control, the channel is connected to the electromyography acquisition circuit by default, and the channel function can be switched within 3ms after the control command is issued.
Citation Information
Patent Citations
Electrical stimulation rehabilitation device and method on basis of feedback control of angle information and electromyographic signals
CN103691059A
Multichannel electrical stimulation device based on myoelectricity feedback
CN107252525A
Functional electrical stimulation closed-loop regulation and control method based on muscle activation degree and LSTM
CN115177864A
Rehabilitation assessment system combining functional electrical stimulation device and CPM rehabilitation device
CN115779266A
