CN114587391A - Brain-computer interface-based rehabilitation training device and training method - Google Patents
Brain-computer interface-based rehabilitation training device and training method Download PDFInfo
- Publication number
- CN114587391A CN114587391A CN202210229903.1A CN202210229903A CN114587391A CN 114587391 A CN114587391 A CN 114587391A CN 202210229903 A CN202210229903 A CN 202210229903A CN 114587391 A CN114587391 A CN 114587391A
- Authority
- CN
- China
- Prior art keywords
- rehabilitation training
- target user
- intention
- motor
- imagery
- 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
- 238000012549 training Methods 0.000 title claims abstract description 224
- 238000000034 method Methods 0.000 title claims abstract description 38
- 230000033001 locomotion Effects 0.000 claims abstract description 66
- 238000013145 classification model Methods 0.000 claims abstract description 33
- 230000009471 action Effects 0.000 claims abstract description 17
- 238000003062 neural network model Methods 0.000 claims description 31
- 238000011156 evaluation Methods 0.000 claims description 14
- 238000000605 extraction Methods 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 8
- 238000005516 engineering process Methods 0.000 abstract description 5
- 238000012545 processing Methods 0.000 abstract description 4
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 210000004556 brain Anatomy 0.000 description 11
- 230000000284 resting effect Effects 0.000 description 8
- 230000006870 function Effects 0.000 description 6
- 230000004913 activation Effects 0.000 description 5
- 210000002569 neuron Anatomy 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000001228 spectrum Methods 0.000 description 5
- 230000003925 brain function Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 210000003710 cerebral cortex Anatomy 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000010219 correlation analysis Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 210000004092 somatosensory cortex Anatomy 0.000 description 2
- 238000010183 spectrum analysis Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 206010061296 Motor dysfunction Diseases 0.000 description 1
- 208000028389 Nerve injury Diseases 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000001364 causal effect Effects 0.000 description 1
- 210000003169 central nervous system Anatomy 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 210000003141 lower extremity Anatomy 0.000 description 1
- 230000004630 mental health Effects 0.000 description 1
- 230000008764 nerve damage Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 208000020431 spinal cord injury Diseases 0.000 description 1
- 210000001364 upper extremity Anatomy 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/389—Electromyography [EMG]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
- A61B5/316—Modalities, i.e. specific diagnostic methods
- A61B5/369—Electroencephalography [EEG]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7225—Details of analog processing, e.g. isolation amplifier, gain or sensitivity adjustment, filtering, baseline or drift compensation
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
- A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Abstract
The invention discloses a rehabilitation training device and a training method based on a brain-computer interface technology, and relates to the technical field of physiological signal processing. This rehabilitation training device includes: the rehabilitation training device comprises signal acquisition equipment and a rehabilitation training device connected with the signal acquisition equipment; the rehabilitation training device is used for: identifying a classification model according to the physiological electric signals and the movement intention of the target user acquired by the signal acquisition equipment to obtain the movement intention imagination type and the expected rehabilitation training level of the target user; the motor intention imagery types include left hand motor imagery, right hand motor imagery, foot motor imagery, tongue motor imagery; the expected rehabilitation training level comprises the intensity strength of the action, the moving distance and the completion speed; and generating the rehabilitation training scheme of the target user according to the type of the motor intention imagination of the target user and the expected rehabilitation training level. The invention increases the recognition precision of the action during the rehabilitation training by using the brain-computer interface for assistance, thereby being beneficial to improving the rehabilitation effect of the rehabilitation training.
Description
Technical Field
The invention relates to the technical field of physiological signal processing, in particular to a rehabilitation training device and a training method based on a brain-computer interface.
Background
The brain is the high-level central nervous system of human control vegetative, and for some serious nerve damages, such as stroke, spinal cord injury and the like, the motor dysfunction is accompanied, the physical and mental health of patients is seriously threatened, and the life quality is reduced. The increasing aging trend and the large base of physical disabilities make great gaps in clinical rehabilitation treatment means and rehabilitation auxiliary equipment.
In recent years, a Brain-computer interface (BCI) technology emerging in recent years can establish communication between a Brain and an external environment to achieve the purpose of controlling equipment, but still is in a starting stage, only a simple exercise intention can be recognized, a given rehabilitation training action is assisted, deep communication and interaction with a patient are lacked, so that the use feeling of the patient is poor, the participation degree is low, the recognition precision of the exercise intention is low, effective control between the Brain-computer interface and the external equipment is hindered, and an expected exercise rehabilitation effect is difficult to achieve.
The existing brain-computer interface rehabilitation training system can support the recognition and rough classification of a few movement intentions of a user, such as the movement of the left hand, the right hand, the foot and the tongue, and the movement intention recognition of the action types is called as transverse research. However, there is insufficient attention and research on the depth to which a certain exercise needs or wants to be performed, such as the intensity and strength of the exercise, the moving distance, the speed of completing the exercise, etc., but these longitudinal factors are just barriers to the development of intelligent, humanized and precise BCI rehabilitation exercise systems, so that the clinical and rehabilitation training effects are greatly reduced.
Disclosure of Invention
The invention aims to provide a rehabilitation training device and a training method based on a brain-computer interface, which can assist in improving high-precision recognition of a brain during rehabilitation training by using a brain-computer interface technology, and further improve the effect of the rehabilitation training.
In order to achieve the purpose, the invention provides the following scheme:
a rehabilitation training device based on a brain-computer interface, comprising: the rehabilitation training device comprises signal acquisition equipment and a rehabilitation training device connected with the signal acquisition equipment;
the signal acquisition equipment is used for acquiring physiological electric signals of a target user;
the rehabilitation training device is used for:
identifying a classification model according to the physiological electric signal and the movement intention of the target user to obtain the movement intention imagination type and the expected rehabilitation training level of the target user; the motor intention imagery types include left hand motor imagery, right hand motor imagery, foot motor imagery and tongue motor imagery; the expected rehabilitation training level comprises the intensity and strength of the action, the moving distance and the speed of finishing the action;
and generating a rehabilitation training scheme of the target user according to the motor intention imagination type and the expected rehabilitation training level of the target user.
Optionally, the rehabilitation training device specifically includes:
the prediction module is used for inputting the physiological electric signals of the target user into the movement intention recognition and classification model, and the movement intention recognition and classification model outputs the movement intention imagination type and the expected rehabilitation training level of the target user;
and the rehabilitation scheme generation module is used for determining a rehabilitation training instruction according to the motor intention imagination type and the expected rehabilitation training level of the target user, and determining the rehabilitation training scheme of the target user according to the rehabilitation training instruction when the rehabilitation training instruction belongs to a preset evaluation range.
Optionally, the rehabilitation training device further includes:
the model training module is used for determining the motion intention recognition classification model;
the model training module specifically comprises:
a training data determination unit for acquiring training data; the training data comprises physiological electrical signals of the user, corresponding motor intention imagery types and corresponding expected rehabilitation training levels;
the network construction unit is used for constructing a neural network model;
and the network training unit is used for inputting the training data into the neural network model, training the neural network model by adopting a gradient descent strategy and determining the trained neural network model as the movement intention recognition classification model.
Optionally, the method further comprises: a signal processor;
the signal acquisition equipment is connected with the rehabilitation training device through the signal processor; the signal processor is used for carrying out feature extraction on the physiological electric signal of the target user to obtain the signal feature of the target user.
Optionally, the method further comprises: a display;
the display is connected with the rehabilitation training device;
the display is used for displaying the rehabilitation training scheme of the target user.
A brain-computer interface based rehabilitation training method comprises the following steps:
acquiring a physiological electric signal of a target user;
identifying a classification model according to the physiological electric signal and the movement intention of the target user to obtain the movement intention imagination type and the expected rehabilitation training level of the target user; the type of motor intention imagery includes left hand motor imagery, right hand motor imagery, foot motor imagery, tongue motor imagery; the expected rehabilitation training level comprises the intensity strength of the action, the moving distance and the completion speed;
and generating a rehabilitation training scheme of the target user according to the motor intention imagination type and the expected rehabilitation training level of the target user.
Optionally, the identifying and classifying a model according to the physiological electrical signal and the motor intention of the target user to obtain the motor intention imagery type and the expected rehabilitation training level of the target user specifically includes:
and inputting the physiological electric signal of the target user into the movement intention recognition and classification model, and outputting the movement intention imagination type and the expected rehabilitation training level of the target user by the movement intention recognition and classification model.
Optionally, the generating a rehabilitation training plan of the target user according to the type of motor intention imagery of the target user and the expected rehabilitation training level specifically includes:
and determining a rehabilitation training instruction according to the type of the motor intention imagination of the target user and the expected rehabilitation training level, and determining a rehabilitation training scheme of the target user according to the rehabilitation training instruction when the rehabilitation training instruction belongs to a preset evaluation range.
Optionally, the determination method of the motion intention recognition classification model is as follows:
acquiring training data; the training data comprises physiological electrical signals of the user, corresponding motor intention imagery types and corresponding expected rehabilitation training levels;
constructing a neural network model;
inputting the training data into the neural network model, training the neural network model by adopting a gradient descent strategy, and determining the trained neural network model as the movement intention recognition classification model.
Optionally, after the generating of the rehabilitation training plan of the target user according to the type of motor intention imagery of the target user and the expected rehabilitation training level, the method further includes:
and displaying the rehabilitation training scheme of the target user.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention discloses a rehabilitation training device and a training method based on a brain-computer interface, wherein the rehabilitation training device identifies a classification model according to a physiological electric signal and a movement intention of a target user to obtain a movement intention imagination type and an expected rehabilitation training level of the target user so as to generate a rehabilitation training scheme of the target user, wherein the movement intention imagination type comprises a left hand movement imagination, a right hand movement imagination, a foot movement imagination and a tongue movement imagination, the expected rehabilitation training level comprises the strength of movement, the movement distance and the movement completing speed, and the rehabilitation training device solves the problem that the high-precision identification cannot be carried out on the aspects of longitudinal movement strength, the movement distance, the movement completing speed and the like in the prior art.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a structural diagram of a rehabilitation training device based on a brain-computer interface according to an embodiment of the present invention;
fig. 2 is a schematic diagram of an application logic of a rehabilitation training device based on a brain-computer interface according to an embodiment of the present invention;
fig. 3 is a schematic flow chart of a rehabilitation training method based on a brain-computer interface according to an embodiment of the present invention;
fig. 4 is a signal acquisition flowchart of a rehabilitation training method based on a brain-computer interface according to an embodiment of the present invention;
fig. 5 is a schematic diagram of a neural network model of a rehabilitation training method based on a brain-computer interface according to an embodiment of the present invention;
fig. 6 is a schematic diagram of an activation function form of a neural network model of a rehabilitation training method based on a brain-computer interface according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention aims to provide a rehabilitation training device and a training method based on a brain-computer interface, which are beneficial to improving the rehabilitation effect of rehabilitation training by using the brain-computer interface to assist in increasing the recognition precision of actions during the rehabilitation training.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
As shown in fig. 1-2, a rehabilitation training device based on a brain-computer interface according to an embodiment of the present invention includes: signal acquisition equipment and with the rehabilitation training device that signal acquisition equipment is connected.
Specifically, the signal acquisition equipment is used for acquiring physiological electric signals of a target user; the rehabilitation training device is used for: identifying a classification model according to the physiological electric signal and the movement intention of the target user to obtain the movement intention imagination type and the expected rehabilitation training level of the target user; the motor intention imagery types include left hand motor imagery, right hand motor imagery, foot motor imagery and tongue motor imagery; the expected rehabilitation training level comprises the intensity and strength of the action, the moving distance and the speed of finishing the action; and generating a rehabilitation training scheme of the target user according to the motor intention imagination type and the expected rehabilitation training level of the target user.
Further, the rehabilitation training device specifically comprises a prediction module and a rehabilitation scheme generation module.
Specifically, the prediction module is used for inputting the physiological electrical signal of the target user into the motor intention recognition classification model, and the motor intention recognition classification model outputs a motor intention imagination type and an expected rehabilitation training level of the target user; the rehabilitation scheme generation module is used for determining a rehabilitation training instruction according to the motor intention imagination type and the expected rehabilitation training level of the target user, and determining the rehabilitation training scheme of the target user according to the rehabilitation training instruction when the rehabilitation training instruction belongs to a preset evaluation range. The preset evaluation range is an evaluation range of training depth judged by a doctor according to experience; and when the rehabilitation training instruction exceeds the evaluation range, performing rehabilitation training according to the boundary value of the evaluation range.
Further, the rehabilitation training device further comprises a model training module, and the model training module is used for determining the movement intention recognition classification model.
Further, the model training module specifically comprises a training data determining unit, a network constructing unit and a network training unit.
Specifically, the training data determination unit is configured to obtain training data; the training data comprises physiological electrical signals of the user, corresponding motor intention imagery types and corresponding expected rehabilitation training levels; the network construction unit is used for constructing a neural network model; and the network training unit is used for inputting the training data into the neural network model, training the neural network model by adopting a gradient descent strategy and determining the trained neural network model as the movement intention recognition classification model.
Furthermore, the rehabilitation training device based on the brain-computer interface technology further comprises a signal processor, and the signal acquisition equipment is connected with the rehabilitation training device through the signal processor.
Specifically, the signal processor is configured to perform feature extraction on the physiological electrical signal of the target user to obtain a signal feature of the target user. The feature extraction adopts a signal processing and analyzing method, and the signal processing and analyzing method comprises denoising anti-interference, envelope analysis, correlation analysis, time-frequency analysis, power spectrum analysis, brain function connectivity analysis, event-related desynchronization (ERD)/synchronization phenomenon (ERS) analysis, energy statistical analysis and the like.
The denoising and anti-interference are realized by utilizing a Butterworth band-pass filter and reserving signal components related to electroencephalogram characteristics in a 4-30 Hz pass band; the amplitude, area, variance, lag time, etc. domain features of the signal can then be directly computed. The envelope analysis is to utilize an Empirical Mode Decomposition (EMD) method or directly calculate an extremum in a region, and connect to output a signal envelope. The correlation analysis is an output pearson correlation coefficient. The time frequency analysis is based on short-time Fourier transform or wavelet transform, and outputs time frequency spectrum. The power spectrum analysis is based on an AR model and outputs a power spectrum. The brain functional connectivity analysis is of many kinds, and in this example, causal analysis (Granger calvariaty) is used; partial directed coherence analysis (PDC). ERD/ERS analysis: namely event-related desynchronization/synchronization analysis, output ERD/ERS waveform, and further solve the amplitude, extreme point moment and the like of the waveform under different tasks. The energy statistical analysis is to calculate the energy of the brain electrical signal in a time period. The data are used for constructing a neural network model.
The signal features include time domain features, frequency domain features, time-frequency domain features, and spatial domain features of the signal, wherein the time domain features of the signal include, but are not limited to, amplitude, area, energy, variance, envelope, lag time, correlation. The frequency domain features of the signal include spectral components, AR model-based power spectra, and combinations thereof. The time-frequency domain characteristics of the signal include a time-frequency spectrum based on a short-time Fourier transform, and a time-frequency spectrum based on a wavelet transform. The spatial domain characteristics of the signal include different channel electroencephalogram signal activation levels, brain function connection levels, characteristics extracted based on a Common Spatial Pattern (CSP) filter, and the like.
Further, the rehabilitation training device based on the brain-computer interface technology further comprises a display, and the display is connected with the rehabilitation training device. The display is used for displaying the rehabilitation training scheme of the target user.
Further, the signal acquisition device further comprises a physiological electrical signal acquisition device and an evaluation input device.
Specifically, the physiological electric signal acquisition equipment is used for acquiring electroencephalogram signals and/or electromyogram signals in a resting state and a motor imagery state. The evaluation input device is used for inputting the preset evaluation range into the rehabilitation scheme generation module.
Further, the signal acquisition device further comprises an external movement assistance device. The external movement assisting device is used for helping the target user to complete the rehabilitation training scheme or suggestion together, and the purpose of assisting the user in rehabilitation training is achieved.
In the present embodiment, the brain movement-related cortex includes a primary movement region (main movement region), an auxiliary movement region, and an anterior movement region. Studies have shown that the brain activates different levels of motor-related and somatosensory cortex under producing different motor imagery intensities. Therefore, the expectation of the current user for the execution of a specific action image can be evaluated according to the difference of the electroencephalogram signal characteristics of the cortical areas in the motor image process.
As shown in fig. 3 to fig. 6, a rehabilitation training method based on a brain-computer interface according to an embodiment of the present invention includes:
step 100: acquiring physiological electrical signals of a target user.
Step 200: identifying a classification model according to the physiological electric signal and the movement intention of the target user to obtain the movement intention imagination type and the expected rehabilitation training level of the target user, and specifically comprising the following steps of:
and inputting the physiological electric signal of the target user into the movement intention recognition and classification model, and outputting the movement intention imagination type and the expected rehabilitation training level of the target user by the movement intention recognition and classification model. The type of motor intention imagery includes left hand motor imagery, right hand motor imagery, foot motor imagery, tongue motor imagery; the expected rehabilitation training level includes intensity of the action, distance moved and speed of completion.
Step 300: generating a rehabilitation training scheme of the target user according to the type of the motor intention imagination of the target user and the expected rehabilitation training level, which specifically comprises the following steps:
and determining a rehabilitation training instruction according to the motor intention imagination type and the expected rehabilitation training level of the target user, and determining a rehabilitation training scheme of the target user according to the rehabilitation training instruction when the rehabilitation training instruction belongs to a preset evaluation range.
In step 100, the process of acquiring the physiological electrical signal specifically includes: the method comprises the steps of requiring a target user to be in a resting state to carry out signal acquisition in a resting period, and requiring the target user to be in a motor imagery state to carry out signal acquisition in an imagery period. The resting state refers to a state that the brain does not perform imaginary activities when the body of the user is in a resting state; the motor imagery state is a state in which the body of the user is at rest and the brain performs an imagery activity indicating a specific action. As shown in fig. 4.
The acquired physiological electric signals comprise electroencephalogram signals and/or electromyogram signals in a resting state and a motor imagery state. The electroencephalogram signals are electroencephalogram signals of brain movement related cortex and somatosensory cortex, at least comprise electroencephalogram signals of C3 and C4 channels, and the electromyogram signals are electromyogram signals of upper limbs and peripheral ends thereof or lower limbs and peripheral ends thereof.
Wherein in step 200, the determination method of the motion intention recognition classification model comprises the following three steps.
Firstly, acquiring training data; the training data includes physiological electrical signals of the user, corresponding motor intent imagery types, and corresponding expected rehabilitation training levels.
Second, a neural network model is constructed, as shown in fig. 4.
1) An input layer: the signal characteristic parameter X ═ { xi (i ═ 1, 2, 3 … m) }.
2) Hiding the layer: two neuron nodes are included, V ═ vik (i ═ 1, 2, 3 … m, k ═ 1, 2) } denotes the connection weight between the ith neuron of the input layer and the kth neuron of the hidden layer, and W ═ wki (k ═ 1, 2, i ═ 1, 2, 3 … n) } denotes the connection weight between the kth neuron of the hidden layer and the ith neuron of the output layer.
3) An output layer: the expected rehabilitation training level Y ═ { yi (i ═ 1, 2, 3 … n) }, such as: intensity of the action, distance of movement, speed of completion, and the like.
4) Activation function: the activation function adopted by the neural network model is a sigmoid function, the function is expressed as follows, and the form of the function is shown in FIG. 5.
And thirdly, inputting the training data into the neural network model, training the neural network model by adopting a gradient descent strategy, and determining the trained neural network model as the movement intention recognition classification model. The gradient descent strategy is: when the error between the model output and the actual output is small enough (e.g. less than 0.01) or the training times reach a certain number (e.g. more than 1 x 10^4), the training is finished, and the model coefficients learned at this time represent the coefficients of the final motion intention recognition classification model.
In this embodiment, the logic for establishing the neural network model is: when a target user starts to use the system, different training depths are used as target labels, electroencephalogram signals in a resting state and an imagination state are obtained, characteristics are extracted, the electroencephalogram signals are input into a neural network model for training, model coefficients are obtained and stored, and the model coefficients are used for predicting a subsequent formal training process. The process can be combined with a learning model for identifying the type of the motor intention for data acquisition and training without generating additional time consumption or workload.
And fitting and analyzing corresponding training indexes (the type of motor intention imagination and the expected rehabilitation training level of the target user) according to the characteristic difference of the electroencephalogram signals under the resting state and the motor imagination state of the user, and providing a training scheme. The fitting analysis may use neural network models, non-linear regression models, decision tree models, and the like.
After the rehabilitation training scheme of the target user is determined in step 300, the predicted expected training rehabilitation level is supplemented into the original training scheme to form a training scheme customized for the user; the individually customized training regimen needs to conform to the assessment range of the training depth that the doctor is appropriate for the user; if the evaluation range is exceeded, performing rehabilitation training by using the range boundary; and visual feedback is carried out in an animation mode, a user is guided to carry out motor imagery, and the result of the motor imagery is presented.
Further, after the generating of the rehabilitation training scheme for the target user according to the type of the motor intention imagery of the target user and the expected rehabilitation training level, the method further comprises: and displaying the rehabilitation training scheme of the target user.
In this embodiment, when performing the exercise rehabilitation training process, the customized training scheme may be adaptively adjusted according to the user's change to the exercise expectation; the rehabilitation training method and the system can also be used for brain function trainers to exercise the activation of the cerebral cortex and the brain function network and improve the regulation of the brain on attention, emotion and self-consciousness.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (10)
1. A rehabilitation training device based on a brain-computer interface, comprising: the rehabilitation training device comprises signal acquisition equipment and a rehabilitation training device connected with the signal acquisition equipment;
the signal acquisition equipment is used for acquiring physiological electric signals of a target user;
the rehabilitation training device is used for:
identifying a classification model according to the physiological electric signal and the movement intention of the target user to obtain the movement intention imagination type and the expected rehabilitation training level of the target user; the motor intention imagery types include left hand motor imagery, right hand motor imagery, foot motor imagery and tongue motor imagery; the expected rehabilitation training level comprises the intensity and strength of the action, the moving distance and the speed of finishing the action;
and generating a rehabilitation training scheme of the target user according to the motor intention imagination type and the expected rehabilitation training level of the target user.
2. The brain-computer interface based rehabilitation training device according to claim 1, specifically comprising:
the prediction module is used for inputting the physiological electric signals of the target user into the movement intention recognition and classification model, and the movement intention recognition and classification model outputs the movement intention imagination type and the expected rehabilitation training level of the target user;
and the rehabilitation scheme generation module is used for determining a rehabilitation training instruction according to the motor intention imagination type and the expected rehabilitation training level of the target user, and determining the rehabilitation training scheme of the target user according to the rehabilitation training instruction when the rehabilitation training instruction belongs to a preset evaluation range.
3. The brain-computer interface based rehabilitation training device of claim 2, further comprising:
the model training module is used for determining the motion intention recognition classification model;
the model training module specifically comprises:
a training data determination unit for acquiring training data; the training data comprises physiological electrical signals of the user, corresponding motor intention imagery types and corresponding expected rehabilitation training levels;
the network construction unit is used for constructing a neural network model;
and the network training unit is used for inputting the training data into the neural network model, training the neural network model by adopting a gradient descent strategy and determining the trained neural network model as the movement intention recognition classification model.
4. The brain-computer interface based rehabilitation training device of claim 1, further comprising: a signal processor;
the signal acquisition equipment is connected with the rehabilitation training device through the signal processor; the signal processor is used for carrying out feature extraction on the physiological electric signal of the target user to obtain the signal feature of the target user.
5. The brain-computer interface based rehabilitation training device of claim 1, further comprising: a display;
the display is connected with the rehabilitation training device;
the display is used for displaying the rehabilitation training scheme of the target user.
6. A rehabilitation training method based on a brain-computer interface is characterized by comprising the following steps:
acquiring a physiological electrical signal of a target user;
identifying a classification model according to the physiological electric signal and the movement intention of the target user to obtain the movement intention imagination type and the expected rehabilitation training level of the target user; the type of motor intention imagery includes left hand motor imagery, right hand motor imagery, foot motor imagery, tongue motor imagery; the expected rehabilitation training level comprises the intensity strength of the action, the moving distance and the completion speed;
and generating a rehabilitation training scheme of the target user according to the motor intention imagination type and the expected rehabilitation training level of the target user.
7. The brain-computer interface-based rehabilitation training method according to claim 6, wherein the identifying and classifying model according to the physiological electrical signal and the motor intention of the target user to obtain the motor intention imagery type and the expected rehabilitation training level of the target user specifically comprises:
and inputting the physiological electric signal of the target user into the movement intention recognition and classification model, and outputting the movement intention imagination type and the expected rehabilitation training level of the target user by the movement intention recognition and classification model.
8. The brain-computer interface based rehabilitation training method according to claim 6, wherein the generating of the rehabilitation training regimen for the target user according to the type of motor intention imagery and the expected rehabilitation training level of the target user specifically comprises:
and determining a rehabilitation training instruction according to the motor intention imagination type and the expected rehabilitation training level of the target user, and determining a rehabilitation training scheme of the target user according to the rehabilitation training instruction when the rehabilitation training instruction belongs to a preset evaluation range.
9. The brain-computer interface-based rehabilitation training method according to claim 7, wherein the determination method of the motor intention recognition classification model is as follows:
acquiring training data; the training data comprises physiological electrical signals of the user, corresponding motor intention imagery types and corresponding expected rehabilitation training levels;
constructing a neural network model;
inputting the training data into the neural network model, training the neural network model by adopting a gradient descent strategy, and determining the trained neural network model as the movement intention recognition classification model.
10. The brain-computer interface based rehabilitation training method according to claim 6, further comprising, after the generating the rehabilitation training regimen for the target user according to the type of motor intention imagery and the expected rehabilitation training level of the target user:
and displaying the rehabilitation training scheme of the target user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210229903.1A CN114587391A (en) | 2022-03-10 | 2022-03-10 | Brain-computer interface-based rehabilitation training device and training method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210229903.1A CN114587391A (en) | 2022-03-10 | 2022-03-10 | Brain-computer interface-based rehabilitation training device and training method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114587391A true CN114587391A (en) | 2022-06-07 |
Family
ID=81809711
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210229903.1A Pending CN114587391A (en) | 2022-03-10 | 2022-03-10 | Brain-computer interface-based rehabilitation training device and training method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114587391A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115381465A (en) * | 2022-07-28 | 2022-11-25 | 山东海天智能工程有限公司 | Rehabilitation training system based on BCI/VR and AR technologies |
-
2022
- 2022-03-10 CN CN202210229903.1A patent/CN114587391A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115381465A (en) * | 2022-07-28 | 2022-11-25 | 山东海天智能工程有限公司 | Rehabilitation training system based on BCI/VR and AR technologies |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102573619B (en) | Device and method for generating a representation of a subject's attention level | |
| Ebrahimi et al. | Brain-computer interface in multimedia communication | |
| Kus et al. | Asynchronous BCI based on motor imagery with automated calibration and neurofeedback training | |
| CN107981997B (en) | A kind of method for controlling intelligent wheelchair and system based on human brain motion intention | |
| Lu et al. | Tractor assistant driving control method based on EEG combined with RNN-TL deep learning algorithm | |
| CN111110230A (en) | Motor imagery electroencephalogram feature enhancement method and system | |
| CN110993056A (en) | Hybrid active rehabilitation method and device based on mirror image neurons and brain-computer interface | |
| Chhabra et al. | Investigation of deep convolutional neural network for classification of motor imagery fNIRS signals for BCI applications | |
| CN112244774A (en) | Brain-computer interface rehabilitation training system and method | |
| CN111976733B (en) | Method and system for continuously predicting steering intention of driver | |
| Fatima et al. | Towards a low cost Brain-computer Interface for real time control of a 2 DOF robotic arm | |
| CN114587391A (en) | Brain-computer interface-based rehabilitation training device and training method | |
| Wang et al. | A method from offline analysis to online training for the brain-computer interface based on motor imagery and speech imagery | |
| Ramadoss et al. | Complexity and information-based analysis of the electroencephalogram (EEG) signals in standing, walking, and walking with a brain–computer interface | |
| CN115067970A (en) | Rehabilitation effect evaluation method and system based on electroencephalogram and electromyogram signals | |
| Cososchi et al. | EEG features extraction for motor imagery | |
| Bueno et al. | Human-robot cognitive interaction | |
| CN112987917A (en) | Motion imagery enhancement method, device, electronic equipment and storage medium | |
| Coyle et al. | Extracting features for a brain-computer interface by self-organising fuzzy neural network-based time series prediction | |
| Fu et al. | Calculation and analysis of microstate related to variation in executed and imagined movement of force of hand clenching | |
| Xu et al. | Algorithm of imagined left-right hand movement classification based on wavelet transform and AR parameter model | |
| Yang et al. | An sEMG-Based Human-Exoskeleton Interface Fusing Convolutional Neural Networks With Hand-Crafted Features | |
| CN108875580A (en) | A kind of multiclass Mental imagery EEG signal identification method based on because imitating network | |
| Jiralerspong et al. | Identification of three mental states using a motor imagery based brain machine interface | |
| Abuzaher et al. | Mind-wave wheelchair system |
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 | ||
| CB02 | Change of applicant information |
Country or region after: China Address after: Room 306, building 4, future venture Plaza, hi tech Zone, Jinan City, Shandong Province Applicant after: Shandong Zhongke advanced technology Co.,Ltd. Address before: Room 306, building 4, future venture Plaza, hi tech Zone, Jinan City, Shandong Province Applicant before: Shandong Zhongke Advanced Technology Research Institute Co.,Ltd. Country or region before: China |