CN115778760A - Hand function self-adaptive rehabilitation training system and working method - Google Patents
Hand function self-adaptive rehabilitation training system and working method Download PDFInfo
- Publication number
- CN115778760A CN115778760A CN202211707012.9A CN202211707012A CN115778760A CN 115778760 A CN115778760 A CN 115778760A CN 202211707012 A CN202211707012 A CN 202211707012A CN 115778760 A CN115778760 A CN 115778760A
- Authority
- CN
- China
- Prior art keywords
- patient
- signal
- intention
- classification
- signals
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Rehabilitation Tools (AREA)
Abstract
The invention provides a hand function self-adaptive rehabilitation training system and a working method, wherein a data acquisition device acquires a finger pressing force signal, a finger electromyographic signal and an electroencephalographic signal; the data acquisition device transmits the signals to the signal processing device, the signals are preprocessed to form classification features, the classification features are evaluated to obtain quantitative result values, and whether the intention of the patient is effective or not is judged according to the quantitative values; the signal processing device drives the force feedback assisted exercise device to assist the patient in performing the action when the intention is valid, otherwise the force feedback assisted exercise device does not perform the action. The invention comprehensively judges the movement intention of the patient based on the multi-signal acquisition of finger pressing force, myoelectricity and electroencephalogram, improves the judgment accuracy, can assist patients with damaged nervous systems to carry out rehabilitation treatment, improves the recovery speed and the recovery level of hand functions and increases the applicability of rehabilitation equipment. By adjusting the training difficulty strategy in real time, the patient is ensured to be always in a proper challenge interval, and the participation degree of the patient is increased.
Description
Technical Field
The invention relates to the technical field of medical equipment, in particular to rehabilitation equipment.
Background
The stroke is a disease causing disability of adults, in recent years, the number of Chinese stroke patients is in the front of the world, and the number of people with limb movement function damage caused by stroke sequelae is increased year by year. The hand function impairment in stroke patients accounts for about eight tenths of a turn, which manifests as difficulty in stretching and bending the hand joints and inability to perform a simple grasping action. The hand movement function occupies a great proportion in daily life, and the loss of the function seriously affects the life quality of patients. Relevant researches show that after stroke patients suffer from diseases, timely motor rehabilitation therapy is carried out, relevant motor nerve circuits of the patients can be remodeled, and the hand function recovery level of the patients is improved to a certain degree. Among various treatment methods, compared with passive hand pressing, if the patient can actively participate in completing hand movement, the rehabilitation effect is better.
At present, new technologies aiming at the treatment are continuously developed, partial rehabilitation equipment identifies the intention of controlling hands of a patient by detecting the brain electricity, the muscle electricity or the hand action of the patient, and an external device is operated to help the hands of the patient to complete the related actions. Such devices can interfere with the process of patient cranial nerve remodeling, speeding hand function recovery. Due to the fact that partial patients are deeply injured, hand movement cannot be performed and myoelectricity is weak, and therefore treatment equipment based on myoelectricity and hand movement cannot be used. Another type of treatment equipment which only depends on electroencephalogram often cannot accurately identify the movement intention of a patient due to large signal interference and other reasons. In addition, most rehabilitation products do not have the capability of automatically adjusting the training difficulty according to the recovery degree of the patient, and are not beneficial to the active participation of the patient.
Disclosure of Invention
The purpose of the invention is as follows: the hand function self-adaptive rehabilitation training system and the working method thereof are provided, the movement intention of a patient is judged by detecting the finger pressing force, myoelectricity and electroencephalogram signals of the patient, and if the effort degree of the patient exceeds a certain threshold value, feedback is implemented to assist movement.
The technical scheme is as follows: the invention provides a hand function self-adaptive rehabilitation training system, which comprises:
the data acquisition device is used for acquiring finger pressing force signals, finger myoelectric signals and electroencephalogram signals;
the signal processing device is connected with and acquires the signals acquired by the data acquisition device, performs characteristic classification on the signals, and judges the intention strength of the hand activity of the patient after outputting a quantitative result value;
the force feedback auxiliary motion device is connected with the signal processing device, and the signal processing device drives the force feedback auxiliary motion device to assist the patient in limb motion when the intention of the hand motion of the patient is strong.
Further, the signal processing device comprises a signal preprocessing module, a classification module and a control module, the signal preprocessing module preprocesses signals to obtain classification features, the classification features are input into the classification module to perform second-class judgment, a quantization result value is output, the quantization result value exceeds a threshold value to indicate that the movement intention of the movable hand of the patient is strong, the control module drives the force to feed back the auxiliary movement device to act, otherwise, the movement intention is weak, and the control module does not send signals.
Still further, visual and auditory devices are also included; when the exercise intention is continuously weak, the control module drives the visual and auditory device to send out encouragement information.
The invention also provides a working method of the hand function self-adaptive rehabilitation training system, which comprises the following steps:
(1) The data acquisition device acquires a finger pressing force signal, a finger myoelectric signal and an electroencephalogram signal;
(2) The data acquisition device transmits the signals to the signal processing device, the signals are preprocessed to form classification features, the classification features are evaluated to obtain quantitative result values, and whether the intention of the patient is effective or not is judged according to the quantitative values;
(3) If the intention is valid, the signal processing device drives the force feedback auxiliary motion device to assist the patient to perform limb motion, otherwise, the force feedback auxiliary motion device does not act.
Further, the signal processing device in step (2) first performs preprocessing through a signal preprocessing module to obtain classification characteristics:
(1) the finger electromyographic signals pass through a Kalman filtering filter, a high-pass filter and a low-pass filter, interference information is removed, and a first electromyographic feature is formed; calculating the variance of the finger electromyographic signals to form a second electromyographic characteristic;
(2) intercepting a section of fragment data at intervals of a period of time by the first electromyographic feature and the second electromyographic feature, and splicing the first electromyographic feature and the second electromyographic feature within corresponding time to form an electromyographic feature;
(3) after filtering the electroencephalogram signal, intercepting segment data at intervals to serve as electroencephalogram characteristics;
(4) taking the obtained finger pressing force signal as a finger pressing force characteristic;
(5) carrying out standardization processing on finger pressing pressure data, myoelectricity data and electroencephalogram data;
(6) and summarizing the finger pressing force characteristics, the myoelectric characteristics and the electroencephalogram characteristics of the corresponding time periods to form classification characteristics.
Further, a classification module of the signal processing device in the step (2) evaluates classification characteristics through a classification module based on a convolutional neural network, and the classification characteristics sequentially pass through a time convolution layer, a space convolution layer, a normalization layer, an activation layer, a pooling layer, a full-link layer and an output layer to finally obtain a quantitative result value; and comparing the quantitative result value with a given output threshold, if the quantitative result value is higher than the output threshold, judging that the exercise intention of the patient is valid, and if the quantitative result value is higher than the output threshold, the control module drives the force feedback auxiliary exercise device to act, otherwise, the intention is invalid, and the control module does not send a signal.
Further, initializing an output threshold value and the maximum value and the minimum value of the output threshold value, wherein the smaller the output threshold value is, the smaller the control difficulty of the patient is; according to the patient performance, the classification module adaptively adjusts the output threshold value between the maximum value and the minimum value, so that the patient is always in a proper difficulty interval.
In addition, when the classification module compares the quantitative result value with the output threshold value, the quantitative result value is continuously lower than the output threshold value, the control module drives the visual-auditory device to give out encouragement information, and the control module drives the force feedback auxiliary device to act no matter whether the intention of the patient is strong or weak in the next action.
Has the advantages that:
the invention comprehensively judges the movement intention of the patient based on the multi-signal acquisition of finger pressing force, myoelectricity and electroencephalogram, improves the judgment accuracy, can assist patients with damaged nervous systems to carry out rehabilitation treatment, improves the recovery speed and the recovery level of hand functions, and increases the applicability of rehabilitation equipment; in addition, the strategy of the training difficulty is adjusted in real time, so that the patient is always in a proper challenge interval, and the participation degree of the patient is increased.
Drawings
FIG. 1 is a schematic diagram of an embodiment of a training system.
Detailed Description
A hand function self-adaptive rehabilitation training system is shown in figure 1 and comprises a data acquisition device, a signal processing device, a force feedback auxiliary exercise device and a visual and auditory device.
The data acquisition device comprises a pressure acquisition device, an electromyography acquisition device and a multi-lead electroencephalogram acquisition device which respectively and correspondingly acquire five finger pressing force signals, finger electromyography signals and electroencephalogram signals of a patient. The pressure acquisition device is a film piezoelectric sensor and is placed in the hand exoskeleton equipment to detect the finger force condition of the patient. The electromyographic acquisition device is wound near the long forearm palmaris to detect the electromyographic signals of the forearm in real time. The multi-lead electroencephalogram acquisition device is a multi-lead electroencephalogram cap, such as Pz, C3 and C4 which comprise 3 electrodes and are respectively arranged in a 10-20 system, wherein Pz is a reference electrode.
The signal processing device is connected with the data acquisition device, the force feedback auxiliary movement device and the visual and auditory device and comprises a signal preprocessing module, a classification module and a control module. The data acquisition device transmits acquired signals to a signal preprocessing module of the signal processing device, the signal preprocessing module processes the signals to obtain classification characteristics, the classification characteristics are input into the classification module to perform class II judgment, and a quantization result numerical value is output. If the quantitative result value is greater than the threshold value, the intention of the patient is strong, and the control module controls the force feedback auxiliary motion device to act to assist the patient in limb motion; if the quantitative result value is smaller than the threshold value, the intention of the patient is weak, the control module does not send out a signal, and the force feedback auxiliary exercise device does not act. The larger the quantitative result value is, the stronger the movement intention of the patient to move the hand is. The three signals are integrated, the system judgment accuracy is improved, the application range of patients is expanded, and the system can be used by the patients with disabilities of different degrees.
And the force feedback auxiliary movement device receives the instruction of the control module and assists the patient to carry out the movement of the related limb. In this implementation, the force feedback assisted exercise device is a hand exoskeleton device that retracts or extends the patient's hand.
And the visual and auditory device is connected with the control module of the signal processing device and is used for providing compensation feedback to relieve anxiety of the patient. Specifically, the classification module of the signal processing device continuously judges that the intention of the patient is weak for a plurality of times, the control module controls the visual and auditory device to give out encouraging information, and the control module drives the force feedback device to act no matter how the patient is represented next time, so that the anxiety of the patient is reduced.
The work of the hand function self-adaptive rehabilitation training system is as follows:
(1) The finger pressing force signal, the finger electromyographic signal and the electroencephalographic signal are collected through a pressure collecting device, a myoelectricity collecting device and a multi-lead electroencephalographic collecting device respectively.
(2) The data acquisition device transmits signals to the signal processing device, and the signal processing device firstly carries out preprocessing through the signal preprocessing module to obtain classification characteristics:
1) The sampling rate of the electromyographic signals is 2000Hz, multiple channels are supported, and a plurality of the channels can be selected as signal sources as required;
2) The electromyographic signals pass through Kalman filtering, sequentially pass through a high-pass filter and a low-pass filter, interference information is removed, and a first electromyographic feature is formed;
3) Calculating the variance of each electromyographic signal to form a second electromyographic characteristic;
4) Every 100 milliseconds of the first electromyographic characteristic and the second electromyographic characteristic, intercepting 200 milliseconds (400 data points) of data for fragmentation, and splicing a first electromyographic characteristic signal and a second electromyographic characteristic within corresponding time to form an electromyographic characteristic;
5) The sampling rate of the electroencephalogram signal is 250Hz, the electroencephalogram signal is processed through 15 channels, and 4-40Hz signals are reserved through 3-order Butterworth band-pass filtering;
6) Every 100 milliseconds of the electroencephalogram signal interval, 800 milliseconds (200 data points) are intercepted and sliced to serve as electroencephalogram characteristics;
7) Carrying out standardization processing on the finger pressing pressure signal, myoelectric data and electroencephalogram data;
8) And summarizing the pressure characteristic, the first myoelectric characteristic, the second myoelectric characteristic and the electroencephalogram characteristic of the corresponding time periods to form a classification characteristic.
(3) The classification characteristics are subjected to second-class discrimination through a classification module of the signal processing device again, and a quantization result numerical value is output; optionally, the classification module evaluates the classification features, and the classification features sequentially pass through a time, a space convolution layer, a normalization layer, an activation layer, a pooling layer, a full-link layer and an output layer by using a classification model based on a convolutional neural network to finally obtain a quantitative result value, wherein the higher the quantitative result value is, the more obvious the intention of the patient to move the hand is.
Before the classification model based on the convolutional neural network is used, the data of the patient is used for training so as to achieve a higher classification effect:
a. firstly, collecting a certain amount of finger pressing force signal data, myoelectric signal data and electroencephalogram signal data of a patient;
b. data is divided into two categories, where the first category of data is collected when the patient attempts to move his or her hand; when the second type of data is acquired, the patient does not move any part and is in a relaxed state; the quantity of the two types of data is consistent;
c. processing data according to the method in the step (2) to obtain classification characteristics as training samples;
d. and (3) using the training sample for training the classification model based on the convolutional neural network to obtain a model capable of identifying the hand movement intention of the patient according to the finger pressing force, myoelectricity and electroencephalogram data.
The characteristics are converted into quantitative result values by the model, the patient movement intention is judged to be effective if the quantitative result values are higher than a given output threshold value, and the control module drives the force feedback auxiliary movement device to assist the patient to complete actions. Conversely, the force feedback assisted exercise device is not driven. The smaller the output threshold, the less difficult the patient to control. The signal processing device adjusts the output threshold value and dynamically controls the operation difficulty of the patient.
For the output threshold, the classification module first initializes the output threshold to a reasonable interval:
firstly, collecting data according to the steps a and b, inputting test sample data into a signal processing device, obtaining a quantitative result numerical value of each test sample data, and if the numerical value is greater than an output threshold value, classifying related samples into a first type:
the first type of accuracy calculation formula:
wherein N is 1 Is the total number of samples of the first type, N t1 The number of the first type samples which are correctly classified;
setting the output threshold value as the maximum value of the sample quantization result value, and continuously reducing the output threshold value until ACC 1 Start to increase, record the first to ACC 1 Generating a variation to an output threshold value, which is the maximum value of the output threshold value;
continuing with less output threshold, record the first let ACC 1 A value of 80% is reached, which is the initial threshold of the output threshold;
continue to reduce the output threshold until ACC 1 Not further added, record the first to ACC 1 A value that reaches a maximum value, which is the minimum value of the output threshold;
a reasonable interval of the output threshold is obtained.
In order to ensure that the patient is always in a proper difficulty interval, increase the participation of the patient and improve the recovery level of the patient, an output threshold value needs to be adaptively adjusted:
i, setting a current output threshold value of a classification module as an initial output threshold value;
II. The patient starts to try to exercise the hand, the classification module judges whether the patient exercises the hand, and the control module drives the power feedback auxiliary exercise device to turn over according to the judgment result; when the patient repeats n 1 After moving the hand the second time, the classification module calculates the nearest n 1 Secondary patient successful drive force feedback assisted exercise device proportional ACC n1 ；
III if ACC n1 ACC less than the lower limit 2 Then decrease the output threshold to let ACC n1 Just not less than ACC 2 (ii) a If the output threshold value reaches the minimum value of the output threshold value, the output threshold value is not reduced;
IV, F.ACC n1 Greater than upper limit ACC 3 Then raise the output threshold to let ACC n1 Just no greater than ACC 3 (ii) a If the output threshold value reaches the maximum value of the output threshold value, the output threshold value is not increased;
v, every time the patient finishes n 2 After the second hand movement, repeating the steps (II) to (IV), dynamically adjusting the output threshold value, and ensuring that the system is always in a proper difficulty;
wherein ACC 2 、ACC 3 、n 1 、n 2 Not fixed, set according to the patient's condition. In the present embodiment, n 1 Is 10 times, n 2 Is 5,ACC 2 70% of ACC 3 The content was 90%.
(4) When the weak motor intention is continuously judged, the patients do not have good performance, and the anxiety of the patients is relieved, so that the rehabilitation treatment effect is ensured:
when the classification module detects n consecutive patients 3 If the user fails, the control module controls the visual and audio device to give out encouragement information, such as' fuel filling! ", and regardless of the patient's next performance, the control module actuates the force feedback device,to reduce anxiety in the patient. After providing compensation feedback, the number of consecutive failures is recalculated. Compensating feedback does not affect ACC n1 And (4) calculating. Example n 3 Is set to 2.
The use method for carrying out rehabilitation treatment on the patient by adopting the training system comprises the following steps:
s1, selecting a quiet and comfortable environment, and enabling a patient to be located 1 m in front of a display. The hand exoskeleton equipment is worn by a patient, fingers are arranged on the pressure sensor, and the myoelectric sensor and the electroencephalogram sensor are worn by the patient. And connecting the sensor and the display into a signal processing device.
And S2, opening the training system and starting off-line treatment.
And S3, trying to finish hand-gripping action by the patient according to the visual and auditory device prompt.
And S4, after 5 seconds, the training system controls the hand exoskeleton equipment to complete the holding action, lasts for 1 second after holding, and then releases.
And repeating the steps S3 and S4 for 20 times to finish a group of treatment training.
And S5, analyzing the data acquired by the data acquisition device by the signal processing device, extracting signals, training a classifier, and initializing an output threshold.
And S6, after 5 seconds, the training system controls the hand exoskeleton equipment to complete the gripping action, lasts for 1 second after the hand exoskeleton equipment is gripped, and then releases the hand exoskeleton equipment.
And S7, repeating the steps S3 to S6 for 20 times to finish a group of treatment training.
And S8, training a classification model in the classification module by using the acquired data and initializing parameters such as an output threshold value. The patient may repeat S3 to S7 several times as needed. The more the repetition times are, the more accurate the classification model is, and the more beneficial the on-line rehabilitation treatment is.
And S9, after the model training is finished, starting on-line rehabilitation treatment.
And S10, trying to complete one hand-holding action by the patient according to the prompt of the display.
And S11, judging the intention of the patient to grip the hand in real time every 0.1 second by the system, and obtaining a quantitative result numerical value. If the quantitative result value is larger than the output threshold value, the system controls the hand exoskeleton equipment to complete the gripping action, the duration lasts for 1 second after the gripping action, and then the hand exoskeleton equipment is loosened. If the quantitative result value calculated by the system is not greater than the output threshold value within 5 seconds, the control of the patient fails.
S12, when the patient finishes the hand-holding action for 5 times, the process control module judges whether the patient reaches the self-adaptive intervention condition, and if so, the system automatically adjusts an output threshold value and controls the difficulty.
And S13, judging whether the patient meets the compensation condition or not by the flow control module every time the patient finishes the hand-gripping action for 1 time, and if so, providing compensation feedback.
And (5) repeating the steps S10 to S13 according to the needs of the patient, and completing a set of treatment training.
Claims (8)
1. A hand function adaptive rehabilitation training system, comprising:
the data acquisition device is used for acquiring a finger pressing force signal, a finger myoelectric signal and an electroencephalogram signal;
the signal processing device is connected with and acquires the signals acquired by the data acquisition device, performs characteristic classification on the signals, and judges the intention strength of the hand activity of the patient after outputting a quantitative result value;
the force feedback auxiliary motion device is connected with the signal processing device, and the signal processing device drives the force feedback auxiliary motion device to assist the patient in limb motion when the intention of the hand motion of the patient is strong.
2. The hand function adaptive rehabilitation training system according to claim 1, wherein the signal processing device comprises a signal preprocessing module, a classification module and a control module, the signal preprocessing module preprocesses signals to obtain classification features, the classification features are input into the classification module to perform class II judgment, a quantization result value is output, the quantization result value exceeds a threshold value, the motion intention of the movable hand of the patient is strong, the control module drives the force feedback auxiliary motion device to act, otherwise, the motion intention is weak, and the control module does not send out signals.
3. The hand function adaptive rehabilitation training system of claim 2, further comprising a visual-auditory device; when the exercise intention is continuously weak, the control module drives the visual and auditory devices to send out encouragement information.
4. A working method of a hand function adaptive rehabilitation training system is characterized by comprising the following steps:
(1) The data acquisition device acquires a finger pressing force signal, a finger myoelectric signal and an electroencephalogram signal;
(2) The data acquisition device transmits the signals to the signal processing device, the signals are preprocessed to form classification features, the classification features are evaluated to obtain quantitative result values, and whether the intention of the patient is effective or not is judged according to the quantitative values;
(3) If the intention is valid, the signal processing device drives the force feedback auxiliary motion device to assist the patient to perform limb motion, otherwise, the force feedback auxiliary motion device does not act.
5. The operating method of the hand function adaptive rehabilitation training system according to claim 4, wherein the signal processing device in step (2) first performs preprocessing through the signal preprocessing module to obtain classification features:
(1) the finger electromyographic signals pass through a Kalman filtering filter, a high-pass filter and a low-pass filter to remove interference information to form first electromyographic characteristics; calculating the variance of the finger electromyographic signals to form a second electromyographic characteristic;
(2) intercepting a section of fragment data at intervals of a period of time by the first electromyographic feature and the second electromyographic feature, and splicing the first electromyographic feature and the second electromyographic feature within corresponding time to form an electromyographic feature;
(3) after filtering the electroencephalogram signal, intercepting segment data at intervals to serve as electroencephalogram characteristics;
(4) taking the obtained finger pressing force signal as a finger pressing force characteristic;
(5) carrying out standardization processing on finger pressing pressure data, myoelectricity data and electroencephalogram data;
(6) and summarizing the finger pressing force characteristics, the myoelectric characteristics and the electroencephalogram characteristics of the corresponding time periods to form classification characteristics.
6. The working method of the hand function adaptive rehabilitation training system according to claim 4 or 5, wherein the classification module of the signal processing device in the step (2) evaluates the classification features through the classification module based on the convolutional neural network, and passes the classification features through a time, a space convolution layer, a normalization layer, an activation layer, a pooling layer, a full connection layer and an output layer in sequence to finally obtain a quantitative result value; and comparing the quantitative result value with a given output threshold, if the quantitative result value is higher than the output threshold, judging that the exercise intention of the patient is valid, and if the quantitative result value is higher than the output threshold, the control module drives the force feedback auxiliary exercise device to act, otherwise, the intention is invalid, and the control module does not send a signal.
7. The method of claim 6, wherein the output threshold and the maximum and minimum values of the output threshold are initialized, and the smaller the output threshold, the less difficult the patient is to control; according to the patient performance, the classification module adaptively adjusts the output threshold value between the maximum value and the minimum value, so that the patient is always in a proper difficulty interval.
8. The method as claimed in claim 6, wherein when the classification module compares the quantized result value with the output threshold value, and the quantized result value is continuously lower than the output threshold value, the control module drives the audiovisual device to provide motivation information, and the control module drives the feedback assistance device to act no matter how strong the intention of the patient is in the next action.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211707012.9A CN115778760A (en) | 2022-12-29 | 2022-12-29 | Hand function self-adaptive rehabilitation training system and working method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211707012.9A CN115778760A (en) | 2022-12-29 | 2022-12-29 | Hand function self-adaptive rehabilitation training system and working method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115778760A true CN115778760A (en) | 2023-03-14 |
Family
ID=85428124
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211707012.9A Pending CN115778760A (en) | 2022-12-29 | 2022-12-29 | Hand function self-adaptive rehabilitation training system and working method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115778760A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117898710A (en) * | 2024-03-18 | 2024-04-19 | 华中科技大学 | Device for determining abnormal hand movement based on finger movement signals |
-
2022
- 2022-12-29 CN CN202211707012.9A patent/CN115778760A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117898710A (en) * | 2024-03-18 | 2024-04-19 | 华中科技大学 | Device for determining abnormal hand movement based on finger movement signals |
| CN117898710B (en) * | 2024-03-18 | 2024-05-17 | 华中科技大学 | Device for determining abnormal hand movement based on finger movement signals |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110238863B (en) | Lower limb rehabilitation robot control method and system based on electroencephalogram-electromyogram signals | |
| CN100594867C (en) | Apparel type robot for healing hand function and control system thereof | |
| CN101874758B (en) | Supplementary motion system and control method thereof | |
| CN110179643A (en) | A kind of neck rehabilitation training system and training method based on annulus sensor | |
| JP2017532093A (en) | Method and system for therapy and / or user device control based on brain activity signals | |
| AU2008212131A1 (en) | A system and method for processing brain signals in a BCI system | |
| CN103440498A (en) | Surface electromyogram signal identification method based on LDA algorithm | |
| CN111584030A (en) | Idea control intelligent rehabilitation system based on deep learning and complex network and application | |
| CN113058157B (en) | Feedback type functional electrical stimulation system with multi-signal fusion | |
| CN115778760A (en) | Hand function self-adaptive rehabilitation training system and working method | |
| CN115206484B (en) | Cerebral apoplexy rehabilitation training system | |
| CN105892676A (en) | Human-machine interaction device, system and method of vascular intervention operation wire feeder | |
| CN110931104A (en) | Upper limb rehabilitation robot intelligent training system and method based on machine learning | |
| CN114021604A (en) | Motion imagery training system based on real-time feedback of 3D virtual reality technology | |
| CN113143676A (en) | Control method of external limb finger based on brain-muscle-electricity cooperation | |
| CN113713333B (en) | Dynamic virtual induction method and system for lower limb rehabilitation full training process | |
| CN114587934A (en) | Finger back extension training system with myoelectricity self-adaptive pressure threshold changing function | |
| Bahri et al. | Sub-band-power-based efficient brain computer interface for wheelchair control | |
| CN113647938A (en) | Method and system for advanced detection of motion state change based on physiological signals | |
| CN116831874A (en) | Lower limb rehabilitation device control method based on electromyographic signals | |
| CN113476799B (en) | Hand training and evaluation method based on myoelectricity and inertia information | |
| WO2022116784A1 (en) | Brain plasticity-based action assisting apparatus, and control method and circuit therefor | |
| Du et al. | The effect of data reduction by independent component analysis and principal component analysis in hand motion identification | |
| CN112394818A (en) | Movement intention identification method of movement auxiliary device based on brain plasticity | |
| Faller et al. | Online co-adaptive brain-computer interfacing: Preliminary results in individuals with spinal cord injury |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |