Rehabilitation glove training device based on surface electromyogram signals and active training system

Through the rehabilitation glove training device based on surface electromyography signals, the use of the Arduino Due board to control the airbag to drive finger movement, the problem of the inability to achieve cross-migration rehabilitation training in the existing technology is solved, the patient's active rehabilitation training effect is achieved, and the affected side hands are improved muscle strength and function.

CN119055487APending Publication Date: 2024-12-03SHANGHAI HIGH SCHOOL
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411325866.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2023-12-28
Filing Date
2024-09-23
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

Existing rehabilitation gloves cannot drive the corresponding finger movement of the affected hand through a single finger movement of a healthy hand, and cannot achieve cross-transfer rehabilitation training, which limits the patient's active rehabilitation training effect.

Method used

Design a rehabilitation glove training device based on surface electromyography signals. The electromyography signals of the healthy hand are collected through sensors, and finger movement is recognized, and the airbag drives the affected hand to perform corresponding finger movements. The Arduino Due board is used to control the gas circuit solenoid valve. Bend your fingers and stretch your fingers.

Benefits of technology

It realizes that patients exercise motor nerves while exercising the affected hand muscles, improve the muscle strength and function of the affected limbs, improve the initiative and accuracy of rehabilitation training, and has a fast recognition speed and high accuracy rate. It is suitable for a variety of rehabilitation products.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119055487A_ABST
    Figure CN119055487A_ABST
Patent Text Reader

Abstract

The invention relates to a rehabilitation glove training device based on a surface electromyogram signal and an active training system, and designs the rehabilitation glove training device based on the surface electromyogram signal in allusion to passive rehabilitation training of an affected hand by a preset instruction of an existing rehabilitation glove product. The training device comprises an Arduino Due plate, a myoelectricity sensor, a sensor signal processing plate, a sensor dry electrode bracelet, a relay, two electromagnetic valves and an air pump. The rehabilitation glove active training system based on the surface electromyogram signals is good in mobility, other rehabilitation products can be embedded conveniently, and the rehabilitation glove active training system can be embedded into VR and remote control products and exoskeleton devices or artificial limb products. The device can help a patient to realize active rehabilitation training, so that the patient exercises the muscle of the affected side hand and exercises the motor nerve by utilizing the cross migration effect, thereby improving the muscle strength and function of the affected side hand and promoting the rehabilitation of the patient.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a rehabilitation glove training device and an active training system based on surface electromyography signals, belonging to the technical field of medical rehabilitation. Background Art

[0002] In recent years, there are more and more patients with hand movement disorders caused by diseases such as "stroke" and musculoskeletal degenerative diseases. Since the affected hand of the patient is in a state of flaccid paralysis and loses all movement ability, the patient's ability to take care of himself and the quality of life are seriously affected. In order to help patients with the rehabilitation of the affected hand, with the development of mechanical technology and electronic technology, more and more hand robots are on the market to help patients with hand disorders recover. These products promote the recovery of neuromuscular reaction ability and prevent muscle rigidity and atrophy by driving the affected hand to move. Among them, airbag-driven household flexible rehabilitation gloves are particularly popular because of their ease of use and low price.

[0003] However, at present, rehabilitation gloves generally focus on the rehabilitation exercises of the finger muscles of the affected hand. The preset program in the microcontroller is activated by pressing a button to drive the affected hand to do rehabilitation exercises. Even if some products have a mirroring function, they simply drive the affected hand to move by pressing the palm sensor with a fist of the healthy hand. Although some products can select which fingers of the affected hand to move through the program interface settings, it is impossible to drive the corresponding finger of the affected hand to move by the movement of a single finger of the healthy hand, and the cross-transfer rehabilitation training effect cannot be achieved.

[0004] Therefore, an active training system for rehabilitation gloves based on surface electromyography signals is invented and designed, which is embedded in existing rehabilitation glove products. It automatically detects the electromyography signals of the healthy hand and identifies the movements of each finger, and then drives the rehabilitation glove of the affected hand to drive the affected hand to do the same finger rehabilitation movements. Based on the cross-migration effect, the patient can do active rehabilitation training, and exercise the corresponding motor nerves while exercising the muscles of the affected hand, thereby improving the muscle strength and function of the affected limb and promoting the early recovery of the patient, which has become a technical problem that urgently needs to be solved in this field. Summary of the invention

[0005] The present invention aims to provide a rehabilitation glove training device based on surface electromyography signals, which can help patients achieve active rehabilitation training, so that patients can exercise the motor nerves while exercising the muscles of the affected hand using the cross-migration effect, thereby improving the muscle strength and function of the affected hand and promoting the patient's rehabilitation.

[0006] In order to achieve the above technical objectives, the technical solution adopted by the present invention is as follows:

[0007] A rehabilitation glove training device based on surface electromyography signal, comprising:

[0008] The sensor dry electrode bracelet is composed of N dry electrode plates of electromyographic sensors connected by elastic bands.

[0009] N sensor signal processing boards, each of which is connected to the dry electrode boards of each electromyographic sensor of the sensor dry electrode bracelet through a sensor dry electrode connecting line.

[0010] Arduino Due board, the positive and negative electrodes of each sensor signal processing board are connected to the positive and negative electrodes of the Arduino Due board through the 3P lead wires of the sensor signal processing board, and the signal ends of each sensor signal processing board are connected to the A0, A2, A4, and A6 analog signal ports of the Arduino Due board respectively.

[0011] Each finger airbag of the rehabilitation glove has an air circuit that includes a relay, two solenoid valves and an air pump. The two solenoid valves are connected to the relay. The air intake and the air charging port of the air pump are each connected to a solenoid valve. The two solenoid valves are respectively connected to the finger airbags to control the movement of the finger airway. The air pump is connected to a power supply.

[0012] D3 to D7 of the Arduino Due board are respectively connected to the signal terminals of the relays on the gas paths corresponding to the thumb, index finger, middle finger, ring finger and little finger.

[0013] Furthermore, the number of the electromyographic sensors and the number of the sensor signal processing boards are both 4.

[0014] The present invention also provides a rehabilitation glove active training system based on surface electromyography signals, comprising the following steps:

[0015] S1. Build a signal acquisition module consisting of a surface electromyography sensor and an Arduino board to collect surface electromyography signals from four locations on the healthy forearm;

[0016] S2. Build a glove finger drive module, and control the air circuit solenoid valve to inflate or exhaust the airbags of each finger, so as to drive the glove to drive the affected hand to perform corresponding finger movements;

[0017] S3, analyzing the collected signal data by drawing graphs, and selecting three signal data features of “total value, number of non-zero values, and proportion” of signal data with good action differentiation;

[0018] S4, designing an action recognition algorithm, comparing the selected three data features of the action signal to be recognized with the three data features of the finger action signal collected when the device is turned on, and determining the signal action according to the nearest neighbor algorithm;

[0019] S5. Write the integrated system Arduino program according to the system control program and upload it to the Arduino Due board as the prototype control system.

[0020] Further, in step S1, the signal acquisition module is a sensor dry electrode bracelet.

[0021] Further, in step S3, the calculation formulas of the three signal data features "total value, number of non-zero values, and proportion" are as follows:

[0022] Total value: i = 1 to 4, representing the signal values ​​of the four sensors respectively;

[0023] j is the jth data collected by the sensor, and a total of 1500 times are collected here.

[0024] Number of non-zero values:

[0025] i = 1 to 4, representing the signal values ​​of the four sensors respectively;

[0026] j is the jth data collected by the sensor, and a total of 1500 times are collected here.

[0027] Specific gravity: 1) i = 1 to 4, representing the signal values ​​of the four sensors respectively;

[0028] 2)Weight (i) =Sum (i) / Sum i=1~4, representing the signal values ​​of the four sensors respectively.

[0029] Furthermore, in step S5, the prototype control system can complete the sampling of 6 types of finger bending and stretching movements and establish a sample database.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] 1) The active training system of the rehabilitation gloves based on surface electromyography signals of the present invention has good portability and is convenient for embedding into other rehabilitation products, such as VR and remote control products, exoskeleton devices or prosthetic products. The components selected by the present invention are small in size, light in weight and cheap in price; the circuit and program of the Arduino development board selected are open source; the designed signal analysis and motion recognition algorithm has very little computational complexity and can be quickly completed in the small Arduino development board; these features make the system easier to integrate into rehabilitation gloves and other products.

[0032] 2) The present invention can identify finer movements and has a fast sampling speed. The present invention can identify fine movements, and can identify the bending and stretching movements of each of the five fingers and the bending and stretching movements of the five fingers at the same time; a small amount of sampling can establish a sample library that meets the recognition requirements (each movement only needs to be sampled 5 times, and the sampling of 6 movements and the establishment of a sample database can be completed within 2-3 minutes of starting up).

[0033] 3) The invention has fast recognition and high accuracy. The motion recognition is fast (recognition is completed within 0.5 seconds), the result is accurate, and the accuracy rate can be as high as 100%. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 FIG. 1 is a connection diagram of an embodiment of the present invention.

[0035] Figure 2 The figure is a schematic diagram of the control electrical circuit of each finger airbag involved in the present invention.

[0036] In the figure, a is the inflation air path, and b is the suction air path.

[0037] Figure 3 These are the three signal data feature calculation formulas of step S3 involved in the present invention.

[0038] Attached Figure 4 This is a flow chart of the system control program of step S5 involved in the present invention. DETAILED DESCRIPTION

[0039] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. According to the following description, the purpose, technical solutions and advantages of the present invention will become clearer. It should be noted that the described embodiments are preferred embodiments of the present invention, rather than all embodiments.

[0040] Combination Figure 1 and Figure 2 As shown, a rehabilitation glove training device based on surface electromyography signal comprises:

[0041] The sensor dry electrode bracelet 5 is formed by connecting four dry electrode plates of electromyographic sensors through elastic bands;

[0042] 4 sensor signal processing boards 3, each sensor signal processing board 3 is connected to the dry electrode boards of each electromyographic sensor of the sensor dry electrode bracelet 5 in a one-to-one correspondence through a sensor dry electrode connecting line 4;

[0043] Arduino Due board 1, the positive and negative electrodes of each sensor signal processing board 3 are connected to the positive and negative electrodes of the Arduino Due board 1 through the sensor signal processing board 3P lead wire 2, and the signal end of each sensor signal processing board 3 is connected to the A0, A2, A4, A6 analog signal port of the Arduino Due board 1 respectively;

[0044] Each finger airbag 12a of the rehabilitation glove 12, the air circuit of each finger airbag 12a includes a relay 8, two solenoid valves 9 and an air pump 10, the two solenoid valves 9 are connected to the relay 8, the air inlet 10a and the inflation port 10b of the air pump 10 are each connected to a solenoid valve 9, the two solenoid valves 9 are respectively connected to the finger airbag 12a to control the movement of the finger airway 12, the air pump 10 is connected to the power supply 13, and the power supply adopts a power bank;

[0045] D3 to D7 of the Arduino Due board 1 are respectively connected to the signal terminals of the relay 8 on the gas path corresponding to the thumb, index finger, middle finger, ring finger and little finger.

[0046] The present invention also provides a rehabilitation glove active training system based on surface electromyography signals, comprising the following steps:

[0047] S1, the dry electrode plates of four electromyographic sensors are connected by elastic bands to form a sensor dry electrode bracelet 5, which collects surface electromyographic signals at four places of the healthy forearm;

[0048] S2. Build a glove finger drive module, and control the air circuit solenoid valve to inflate or exhaust the airbags of each finger, so as to drive the glove to drive the affected hand to perform corresponding finger movements;

[0049] S3, analyzing the collected signal data by drawing graphs, and selecting three signal data features of “total value, number of non-zero values, and proportion” of signal data with good action differentiation;

[0050] S4, designing an action recognition algorithm, comparing the selected three data features of the action signal to be recognized with the three data features of the finger action signal collected when the device is turned on, and determining the signal action according to the nearest neighbor algorithm;

[0051] S5. Write the integrated system Arduino program according to the system control program and upload it to the Arduino Due board as the prototype control system.

[0052] In step S3, the calculation formulas for the three signal data features "total value, number of non-zero values, and proportion" are as follows:

[0053] Total value: i = 1 to 4, representing the signal values ​​of the four sensors respectively;

[0054] j is the jth data collected by the sensor, and a total of 1500 times are collected here.

[0055] Number of non-zero values:

[0056] i = 1 to 4, representing the signal values ​​of the four sensors respectively;

[0057] j is the jth data collected by the sensor, and a total of 1500 times are collected here.

[0058] Specific gravity: 1) i = 1 to 4, representing the signal values ​​of the four sensors respectively;

[0059] 2)Weight (i) =Sum (i) / Sum i=1~4, representing the signal values ​​of the four sensors respectively.

[0060] The system control procedure in step S5 is as shown in the attached Figure 4 shown.

[0061] Prototype experiments have proved that the prototype can collect surface electromyographic signals of the healthy hand forearm on the Arduino Due board, quickly identify six types of finger bending and extension movements, and drive the rehabilitation gloves to drive the affected hand to perform corresponding finger bending and extension movements, thereby realizing active rehabilitation training; the invented active training system of rehabilitation gloves based on surface electromyographic signals is reasonable and successful.

[0062] The above is only a description of the preferred embodiment of the present invention, and is not intended to limit the scope of protection of the present invention. Obviously, any technician familiar with the field can easily think of replacements or changes to obtain other embodiments based on the above embodiments, which should all be included in the scope of protection of the present invention.

Claims

1. A rehabilitation glove training device based on surface electromyography signal, characterized in that: It includes: The sensor dry electrode bracelet (5) is composed of N dry electrode plates (5a) of myoelectric sensors connected by elastic bands (5b). N sensor signal processing boards (3), each sensor signal processing board (3) is connected to the dry electrode boards of each electromyographic sensor of the sensor dry electrode bracelet (5) in a one-to-one correspondence via a sensor dry electrode connecting line (4), Arduino Due board (1), the positive and negative electrodes of each sensor signal processing board (3) are connected to the positive and negative electrodes of the Arduino Due board (1) through the sensor signal processing board 3P lead wire (2), and the signal end of each sensor signal processing board (3) is connected to the A0, A2, A4, A6 analog signal ports of the Arduino Due board (1), respectively. Each finger airbag (12) of the rehabilitation glove, the air circuit of each finger airbag (12) comprises a relay (8), two solenoid valves (9) and an air pump (10), the two solenoid valves (9) are connected to the relay (8), the air inlet (10a) and the air charging port (10b) of the air pump (10) are each connected to a solenoid valve (9), the two solenoid valves (9) are respectively connected to the finger airbag (12) to control the movement of the finger air tube (12a), and the air pump (10) is connected to a power source (13); D3 to D7 of the Arduino Due board (1) are respectively connected to the signal ends of the relay (8) on the air path corresponding to the thumb, index finger, middle finger, ring finger and little finger.

2. The rehabilitation glove training device based on surface electromyography signal according to claim 1, characterized in that: The number of the electromyographic sensors and sensor signal processing boards (3) is 4.

3. The active training system for rehabilitation gloves based on surface electromyography signals according to claim 1 or 2, characterized in that: The following steps are involved: S1. Build a signal acquisition module consisting of a surface electromyography sensor and an Arduino board to collect surface electromyography signals from four locations on the healthy forearm; S2. Build a glove finger drive module, and control the air circuit solenoid valve to inflate or exhaust the airbags of each finger, so as to drive the glove to drive the affected hand to perform corresponding finger movements; S3, analyze the collected signal data by drawing graphs, and select the three signal data features of "total value, number of non-zero values, and proportion" of signal data with good action differentiation; S4, designing an action recognition algorithm, comparing the selected three data features of the action signal to be recognized with the three data features of the finger action signal collected when the device is turned on, and determining the signal action according to the nearest neighbor algorithm; S5. Write the integrated system Arduino program according to the system control program and upload it to the Arduino Due board as the prototype control system.

4. The active training system for rehabilitation gloves based on surface electromyography signals according to claim 3 is characterized in that: In step S1, the signal acquisition module is a sensor dry electrode bracelet (5).

5. The active training system for rehabilitation gloves based on surface electromyography signals according to claim 3 is characterized in that: In step S3, the calculation formulas for the three signal data features "total value, number of non-zero values, and proportion" are as follows: Total value: i = 1 to 4, representing the signal values ​​of the four sensors respectively; j is the jth data collected by the sensor, and a total of 1500 times are collected here. Number of non-zero values: i = 1 to 4, representing the signal values ​​of the four sensors respectively; j is the jth data collected by the sensor, and a total of 1500 times are collected here. Specific gravity: 1) i = 1 to 4, representing the signal values ​​of the four sensors respectively; 2)Weight (i) =Sum (i) / Sum i=1~4, representing the signal values ​​of the four sensors respectively.

6. The active training system for rehabilitation gloves based on surface electromyography signals according to claim 3 is characterized in that: In step S5, the prototype control system can complete the sampling of 6 types of finger bending and stretching movements and establish a sample database.

Citation Information

Cited By

  • Hand rehabilitation training glove and system based on electromyographic signal guidance and pressure feedback

    CN121101610A