An attention training system based on steady-state visual evoked potential

Abstract

The present invention relates to an attention training system based on steady-state visual evoked potential, and belongs to the field of information technology. The system includes: presetting a patient identification type, constructing an intention expresser based on attention training according to the patient identification type; obtaining the EEG signal generated by the patient to the intention expresser through a data collector, and sending the EEG signal to a loop processor, the loop processor performs pattern recognition on the EEG signal to obtain a steady-state visual evoked potential signal; decoding and analyzing the steady-state visual evoked potential signal to obtain an attention feature based on a target stimulus block in the intention expresser, and tracking the attention feature according to the position information of the target stimulus block to obtain a processing feedback instruction; the intention expresser receives the processing feedback instruction, and drives the intention expresser to display the attention expression result according to the processing feedback instruction. An attention training application method with strong environmental adaptability, high recognition rate and response speed is realized.

Description

Technical Field

[0001] The invention belongs to the field of information technology, and in particular relates to an attention training system based on steady-state visual evoked potential. Background Art

[0002] The brain is composed of many types of neurons. The signals generated between these neurons form brain waves, and the connections between neurons prompt human thinking activities. When the signals received by brain neurons from other neurons reach a certain threshold, brain waves will be generated; in this case, human thinking activities can be carried out. For patients with aphasia due to brain injury, patients with neuromuscular diseases, patients with acute stroke aphasia, etc., who face problems such as impaired speech expression, comprehension, communication ability and cognitive dysfunction, they need to undergo rehabilitation training through cognitive training. However, traditional rehabilitation training is too single and cannot fully meet the personalized needs of patients. It is impossible to flexibly adjust the training content and methods, resulting in limited rehabilitation effects. The rehabilitation process requires long-term and continuous rehabilitation training. Long-term training is subject to the influence of time, economy, resources and other factors, which makes it difficult for patients to obtain sufficient continuous training.

[0003] With the development of BCI technology, brain-computer interfaces are widely used in multi-scenario application exploration in conjunction with peripherals such as virtual reality and enhanced display. By allowing the subject to watch a flickering light source with a relatively stable frequency (such as an LED light or a computer screen), the cerebral cortex will produce a steady-state visual evoked potential corresponding to its frequency. By recording and analyzing these signals, information related to the subject's visual perception process can be obtained. At present, the above method is applied to attention-based training and there are still the following problems to be solved: Steady-state visual evoked potentials are prone to cause visual fatigue, and each stimulus block of the steady-state visual evoked potential stimulation paradigm needs to flicker at a fixed frequency and have a certain area, which cannot be close to natural application scenarios. The recognition accuracy and response speed of the system are also affected by the number of encoding targets of the steady-state visual evoked potential. Summary of the invention

[0004] In order to solve the above problems existing in the prior art, the present invention provides an attention training system based on steady-state visual evoked potential.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] S1: Preset the patient identification type, and construct an intention expresser based on attention training according to the patient identification type;

[0007] S2: acquiring the EEG signal generated by the patient in response to the intention expressor through a data collector, and sending the EEG signal to a cyclic processor, wherein the cyclic processor performs pattern recognition on the EEG signal to obtain a steady-state visual evoked potential signal;

[0008] S3: decoding and analyzing the steady-state visual evoked potential signal through the loop processor to obtain an attention feature based on the target stimulus block in the intention expressor, and tracking the attention feature according to the position information of the target stimulus block to obtain a processing feedback instruction;

[0009] S4: The intention expresser receives the processing feedback instruction, and drives the intention expresser to display the attention expression result according to the processing feedback instruction.

[0010] Specifically, the cyclic processor performs signal equalization processing after receiving the EEG signal, and the signal equalization processing method is:

[0011] Extracting an evoked response signal, removing the evoked response from the received EEG signal to obtain a background noise estimate;

[0012] The background noise estimation is analyzed by a multivariate autoregressive model, and the expression is:

[0013]

[0014] Wherein, w(n) is the background noise estimate at the nth moment, p is the order of the multivariate autoregressive model, P is the total order, A(p) is the pth order coefficient matrix of the model, w(np) is the background noise estimate at the npth moment, and ε(n) is the time-domain uncorrelated noise at the nth moment;

[0015] A space-time irrelevant constraint is constructed, and the background noise estimation is equalized according to the space-time irrelevant constraint to obtain the equalized noise. The equalized noise calculation formula is:

[0016]

[0017] Where, h(n) is the equalized noise at the nth moment, M is the covariance matrix of the time-domain uncorrelated noise at the nth moment, I is the whitening matrix;

[0018] The corresponding EEG signal is reconstructed according to the equalized noise and the evoked response signal.

[0019] Specifically, the evoked response signal is a linear aliasing signal of a multi-frequency sinusoidal signal, and the extraction method is: according to the time mark of the stimulation signal sent by the intention expresser, the signal segment corresponding to the time mark in the EEG signal is extracted and used as an event-related potential signal, and the event-related potential signals of the same type are superimposed and averaged to obtain the evoked response signal.

[0020] Specifically, the loop processor and the intention expresser set up a message queue, and synchronize the EEG signals processed by the loop processor and the intention expresser through the message queue. A training memory is also set in the loop processor, and the training memory is used to store and retain the EEG signals generated under a single attention training.

[0021] Specifically, the pattern recognition method is: sampling and collecting the electroencephalogram signal according to the stimulation start and stop time period and stimulation frequency of the target stimulation block in the intention expresser to obtain the steady-state visual evoked potential signal.

[0022] Specifically, the decoding analysis method is:

[0023] The recognition hypothesis result of the stimulus target is established according to the total number of target stimulus blocks in the intention expresser, and the recognition statistic is obtained by a posteriori estimation of the recognition hypothesis result through the observation result of the steady-state visual evoked potential signal. The calculation formula is:

[0024]

[0025] Among them, J is the identification statistic, argmax q is the parameter function, q is the target stimulus block count, Q is the total number of target stimulus blocks, j is the sum count, H q is the recognition hypothesis result of the qth target, P(H q ) is the prior probability that the recognition hypothesis result of the qth target is established, p(x′(n)|H q ) is the probability density function of the steady-state visual evoked potential signal under the condition that the recognition hypothesis result of the qth target is established; P(H j )p(x′(n)|H j ) is the recognition statistical probability measurement item of each target stimulus block;

[0026] The position information of the target stimulus block corresponding to the recognition statistic is used as the attention feature.

[0027] An attention training system based on steady-state visual evoked potential, comprising a preprocessing module, a data acquisition module, a data analysis module, and a feedback control module;

[0028] The preprocessing module is used to preset the patient identification type and construct an intention expresser based on attention training according to the patient identification type;

[0029] The data acquisition module is used to acquire the EEG signal generated by the patient in response to the intention expresser through a data acquisition device, and send the EEG signal to a circulation processor, and the circulation processor performs pattern recognition on the EEG signal to obtain a steady-state visual evoked potential signal;

[0030] The data analysis module is used to decode and analyze the steady-state visual evoked potential signal through the loop processor to obtain the attention feature based on the target stimulus block in the intention expressor, and track the attention feature according to the position information of the target stimulus block to obtain a processing feedback instruction;

[0031] The feedback control module is used for the intention expresser to receive the processing feedback instruction, and drive the intention expresser to display the attention expression result according to the processing feedback instruction.

[0032] The beneficial effects of the present invention are:

[0033] By setting up corresponding visual stimulation trainers and target stimulation paradigms according to different symptoms, the intensity and duration of target stimulation are controlled to reduce the fatigue effect on the visual system and improve the targetedness and effectiveness of treatment; the background noise is analyzed through a multivariate autoregressive model, and the stationary and non-stationary features in the background noise are balanced to reduce the impact of noise. When extracting evoked response signals, the superposition averaging method is used to reduce the random noise in the EEG signal, so that the evoked response can be observed and analyzed more clearly, the computational complexity is low, and the recognition efficiency is improved; by setting up a message queue and training memory, the information synchronization problem and data congestion problem in the acquisition and processing of EEG signals are solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0035] Figure 1 The present invention is a flowchart of an attention training system based on steady-state visual evoked potential. DETAILED DESCRIPTION

[0036] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation method, structure, characteristics and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0037] See also Figure 1 , an attention training system based on steady-state visual evoked potentials:

[0038] S1: Preset the patient identification type, and construct an intention expresser based on attention training according to the patient identification type;

[0039] S2: acquiring the EEG signal generated by the patient in response to the intention expressor through a data collector, and sending the EEG signal to a cyclic processor, wherein the cyclic processor performs pattern recognition on the EEG signal to obtain a steady-state visual evoked potential signal;

[0040] S3: decoding and analyzing the steady-state visual evoked potential signal through the loop processor to obtain an attention feature based on the target stimulus block in the intention expressor, and tracking the attention feature according to the position information of the target stimulus block to obtain a processing feedback instruction;

[0041] S4: The intention expresser receives the processing feedback instruction, and drives the intention expresser to display the attention expression result according to the processing feedback instruction.

[0042] Specifically, the cyclic processor performs signal equalization processing after receiving the EEG signal, and the signal equalization processing method is:

[0043] Extracting an evoked response signal, removing the evoked response from the received EEG signal to obtain a background noise estimate;

[0044] The background noise estimation is analyzed by a multivariate autoregressive model, and the expression is:

[0045]

[0046] Wherein, w(n) is the background noise estimate at the nth moment, p is the order of the multivariate autoregressive model, P is the total order, A(p) is the pth order coefficient matrix of the model, w(np) is the background noise estimate at the npth moment, and ε(n) is the time-domain uncorrelated noise at the nth moment;

[0047] A space-time irrelevant constraint is constructed, and the background noise estimation is equalized according to the space-time irrelevant constraint to obtain the equalized noise. The equalized noise calculation formula is:

[0048]

[0049] Where, h(n) is the equalized noise at the nth moment, M is the covariance matrix of the time-domain uncorrelated noise at the nth moment, I is the whitening matrix;

[0050] The corresponding EEG signal is reconstructed according to the equalized noise and the evoked response signal.

[0051] In this embodiment, it is assumed that the evoked response and the background noise are independent of each other. During the acquisition process, the EEG signal is received in a differential manner. In the equalization processing application, a multi-channel FIR filter is used as an equalizer for the steady-state structure of the EEG signal. The system automatically updates once at a fixed interval using a timed update method. During the steady-state structure update process, the system stops recognition and detection.

[0052] Specifically, the evoked response signal is a linear aliasing signal of a multi-frequency sinusoidal signal, and the extraction method is: according to the time mark of the stimulation signal sent by the intention expresser, the signal segment corresponding to the time mark in the EEG signal is extracted and used as an event-related potential signal, and the event-related potential signals of the same type are superimposed and averaged to obtain the evoked response signal.

[0053] Specifically, the loop processor and the intention expresser set up a message queue, and synchronize the EEG signals processed by the loop processor and the intention expresser through the message queue. A training memory is also set in the loop processor, and the training memory is used to store and retain the EEG signals generated under a single attention training.

[0054] Specifically, the pattern recognition method is: sampling and collecting the electroencephalogram signal according to the stimulation start and stop time period and stimulation frequency of the target stimulation block in the intention expresser to obtain the steady-state visual evoked potential signal.

[0055] In this embodiment, the intention expresser, data collector and loop processor adopt an asynchronous parallel structure, perform real-time analysis during the data collection process, and provide real-time feedback based on the detection results; the intention expresser is responsible for stimulus presentation, connected to the loop processor through the TCP protocol, and receives feedback and other control instructions from the loop processor in real time. The data collector is used to receive and forward the collected EEG signals through the TCP protocol. The data collector sends real-time data to the loop processor in the form of data messages, and the loop processor receives the data packets sent by the data collector and calls the algorithm for real-time processing.

[0056] Specifically, the decoding analysis method is:

[0057] The recognition hypothesis result of the stimulus target is established according to the total number of target stimulus blocks in the intention expresser, and the recognition statistic is obtained by a posteriori estimation of the recognition hypothesis result through the observation result of the steady-state visual evoked potential signal. The calculation formula is:

[0058]

[0059] Among them, J is the identification statistic, argmax q is the parameter function, q is the target stimulus block count, Q is the total number of target stimulus blocks, j is the sum count, H q is the recognition hypothesis result of the qth target, P(H q ) is the prior probability that the recognition hypothesis result of the qth target is established, p(x′(n)|H q ) is the probability density function of the steady-state visual evoked potential signal under the condition that the recognition hypothesis result of the qth target is established; P(H j )p(x′(n)|Hj ) is the recognition statistical probability measurement item of each target stimulus block;

[0060] The position information of the target stimulus block corresponding to the recognition statistic is used as the attention feature.

[0061] An attention training system based on steady-state visual evoked potential includes a preprocessing module, a data acquisition module, a data analysis module, and a feedback control module;

[0062] The preprocessing module is used to preset the patient identification type and construct an intention expresser based on attention training according to the patient identification type;

[0063] The data acquisition module is used to acquire the EEG signal generated by the patient in response to the intention expresser through a data acquisition device, and send the EEG signal to a circulation processor, and the circulation processor performs pattern recognition on the EEG signal to obtain a steady-state visual evoked potential signal;

[0064] The data analysis module is used to decode and analyze the steady-state visual evoked potential signal through the loop processor to obtain the attention feature based on the target stimulus block in the intention expressor, and track the attention feature according to the position information of the target stimulus block to obtain a processing feedback instruction;

[0065] The feedback control module is used for the intention expresser to receive the processing feedback instruction, and drive the intention expresser to display the attention expression result according to the processing feedback instruction.

[0066] The computer storage medium of the embodiment of the present invention may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device.

[0067] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0068] The program code included on the computer readable medium can be transmitted with any appropriate medium, including but not limited to wireless, electric wire, optical cable, RF, etc., or any suitable combination of the above. The computer program code for performing the operation of the present invention can be written in one or more programming languages ​​or their combinations, and the programming language includes object-oriented programming languages-such as Java, Smalltalk, C++, and also includes conventional procedural programming languages-such as "C" language or similar programming languages. The program code can be executed completely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on the remote computer, or completely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0069] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. An attention training system based on steady-state visual evoked potential, characterized in that: It includes preprocessing module, data acquisition module, data analysis module and feedback control module; The preprocessing module is used to preset the patient identification type and construct an intention expresser based on attention training according to the patient identification type; The data acquisition module is used to acquire the EEG signal generated by the patient in response to the intention expresser through a data acquisition device, and send the EEG signal to a circulation processor, and the circulation processor performs pattern recognition on the EEG signal to obtain a steady-state visual evoked potential signal; The cyclic processor performs signal equalization processing after receiving the EEG signal, and the signal equalization processing method is: Extracting an evoked response signal, removing the evoked response from the received EEG signal to obtain a background noise estimate; The background noise estimation is analyzed by a multivariate autoregressive model, and the expression is: Wherein, w(n) is the background noise estimate at the nth moment, p is the order of the multivariate autoregressive model, P is the total order, A(p) is the pth order coefficient matrix of the model, w(np) is the background noise estimate at the npth moment, and ε(n) is the time-domain uncorrelated noise at the nth moment; A space-time irrelevant constraint is constructed, and the background noise estimation is equalized according to the space-time irrelevant constraint to obtain the equalized noise. The equalized noise calculation formula is: Where, h(n) is the equalized noise at the nth moment, M is the covariance matrix of the time-domain uncorrelated noise at the nth moment, I is the whitening matrix; reconstructing the corresponding EEG signal according to the equalized noise and the evoked response signal; The evoked response signal is a linear aliasing signal of a multi-frequency sinusoidal signal, and the extraction method is: according to the time mark of the stimulus signal sent by the intention expresser, the signal segment corresponding to the time mark in the electroencephalogram signal is extracted and used as the event-related potential signal, and the event-related potential signals of the same type are superimposed and averaged to obtain the evoked response signal; The data analysis module is used to decode and analyze the steady-state visual evoked potential signal through the loop processor to obtain the attention feature based on the target stimulus block in the intention expressor, and track the attention feature according to the position information of the target stimulus block to obtain a processing feedback instruction; The decoding analysis method is: The recognition hypothesis result of the stimulus target is established according to the total number of target stimulus blocks in the intention expresser, and the recognition statistic is obtained by a posteriori estimation of the recognition hypothesis result through the observation result of the steady-state visual evoked potential signal. The calculation formula is: Among them, J is the identification statistic, argmax q To find the parameter function, q is the target stimulus block count, Q is the total number of target stimulus blocks, j is the sum count, H q is the recognition hypothesis result of the qth target, P(H q ) is the prior probability that the recognition hypothesis result of the qth target is established, p(x′(n)|H q ) is the probability density function of the steady-state visual evoked potential signal under the condition that the recognition hypothesis result of the qth target is established; P(H j )p(x′(n)|H j ) is the recognition statistical probability measurement item of each target stimulus block; using the position information of the target stimulus block corresponding to the recognition statistic as an attention feature; The feedback control module is used for the intention expresser to receive the processing feedback instruction, and drive the intention expresser to display the attention expression result according to the processing feedback instruction.

2. The system according to claim 1, characterized in that The loop processor and the intention expresser set up a message queue, and synchronize the EEG signals processed by the loop processor and the intention expresser through the message queue. A training memory is also set in the loop processor, and the training memory is used to store and retain the EEG signals generated under a single attention training.

3. The system according to claim 1, characterized in that The pattern recognition method is: sampling and collecting the electroencephalogram signal according to the stimulation start and stop time period and stimulation frequency of the target stimulation block in the intention expresser to obtain the steady-state visual evoked potential signal.

Images