阅读说明：本技术 一种脑电生物反馈得康复方法及系统 (Electroencephalogram biofeedback rehabilitation method and system ) 是由 李龙 康威 周志勇 茆顺明 江亚娟 康文 王荣荣 许红霞 于 2021-09-03 设计创作，主要内容包括：本发明涉及一种脑电生物反馈得康复方法及系统,具体包括以下步骤：获得参考信号REF(n),参考信号长度大概为诱发电位信号长度的50-100倍；采集目标诱发电位信号X(n)；通过IIR干扰相消算法,得到最接近诱发电位信号；消除的最接近诱发电位信号中的高频和低频噪声,得到诱发电位信号EP(n)。将诱发电位信号序列EP(n)按下述频率分段：得到分段诱发电位信号；将分段诱发电位信号进行小波变换、加窗处理,得到分段诱发电位数据,根据分段诱发电位数据进行动作分类识别,根据识别到的动作实施FES电刺激,并进行视频反馈。本发明的康复系统中大脑参与自主康复过程,能够缩短康复时间。(The invention relates to a brain electrical biofeedback rehabilitation method and a brain electrical biofeedback rehabilitation system, which specifically comprise the following steps: obtaining a reference signal REF (n), wherein the length of the reference signal is about 50-100 times of the length of the evoked potential signal; collecting target evoked potential signals X (n); obtaining a closest evoked potential signal through an IIR interference cancellation algorithm; eliminating the high frequency and low frequency noise in the nearest evoked potential signal to obtain an evoked potential signal EP (n). The evoked potential signal sequence EP (n) is segmented according to the following frequency: obtaining a segmented evoked potential signal; and performing wavelet transformation and windowing on the segmented evoked potential signals to obtain segmented evoked potential data, performing action classification and identification according to the segmented evoked potential data, performing FES (feed forward) electrical stimulation according to the identified action, and performing video feedback. The brain in the rehabilitation system of the invention participates in the autonomous rehabilitation process, and the rehabilitation time can be shortened.)

1. A brain electrical biofeedback rehabilitation method is characterized by comprising the following steps:

obtaining a reference signal REF (n), wherein the length of the reference signal is about 50-100 times of the length of the evoked potential signal; collecting target evoked potential signals X (n);

obtaining a closest evoked potential signal through an IIR interference cancellation algorithm;

eliminating the high frequency and low frequency noise in the nearest evoked potential signal to obtain an evoked potential signal EP (n). The evoked potential signal sequence EP (n) is segmented according to the following frequency:

the delta frequency band is 0.5 Hz-4 Hz, the theta frequency band is 4 Hz-8 Hz, the alpha frequency band is 8 Hz-13 Hz, and the beta frequency band is 13 Hz-30 Hz;

obtaining a segmented evoked potential signal;

performing wavelet transformation and windowing on the segmented evoked potential signals to obtain segmented evoked potential data, performing action classification and identification according to the segmented evoked potential data,

FES electrical stimulation is performed according to the recognized action, and video feedback is performed.

2. The brain electrical biofeedback rehabilitation method according to claim 1, wherein the action classification recognition comprises the following steps:

acquiring segmented evoked potential sample data, and adding the segmented evoked potential sample data into a database;

performing classification training according to the segmented evoked potential sample data in the database to obtain a plurality of local features;

assembling the multidimensional local features into complete one-dimensional features through a matrix by adopting a full connection layer according to the acquired segmented evoked potential data;

and mapping the one-dimensional features into the probability of rehabilitation actions according to the soft-max layer, so as to realize the identification of the actions.

3. The brain electrical biofeedback recovery method according to claim 2, characterized in that a CART decision tree is used for classification training.

4. The electroencephalogram biofeedback rehabilitation method according to claim 1, characterized in that high-frequency and low-frequency noise in the closest evoked potential signal is eliminated by adding a band-pass IIR filter.

5. The electroencephalogram biofeedback recovery method according to claim 1, wherein the calculation formula of the nearest evoked potential signal EP (n) is as follows:

EP(n)＝S(n)-W*REF_M(n)

s (n) is an acquired stimulation signal data sequence;

REF_M(n) is the maximum correlation reference sequence, ep (n) can be optimized by adjusting the W matrix, and the value of W is calculated according to the following formula:

W＝∑S(n)*REF_M(n)/∑[REF_M(n)]²。

6. an electroencephalogram biofeedback-derived rehabilitation system, the rehabilitation system comprising:

a reference signal acquisition unit for acquiring a reference signal REF (n), the length of the reference signal being approximately 50-100 times the length of the evoked potential signal;

an evoked potential signal acquisition unit for acquiring a target evoked potential signal X (n);

the interference cancellation processing unit obtains a signal closest to the evoked potential through an IIR interference cancellation algorithm;

a high-low frequency noise elimination unit which eliminates high-frequency and low-frequency noise in the evoked potential signal closest to the evoked potential signal to obtain an evoked potential signal EP (n);

a segmenting unit for segmenting the evoked potential signal sequence EP (n) according to the following frequency:

the delta frequency band is 0.5 Hz-4 Hz, the theta frequency band is 4 Hz-8 Hz, the alpha frequency band is 8 Hz-13 Hz, and the beta frequency band is 13 Hz-30 Hz; obtaining a segmented evoked potential signal;

a processing unit for performing wavelet transformation and windowing on the segmented evoked potential signals to obtain segmented evoked potential data,

an action recognition unit for classifying and recognizing actions according to the segmented evoked potential data,

and the feedback unit is used for implementing FES electrical stimulation according to the recognized action and carrying out video feedback.

7. The brain electrical biofeedback rehabilitation system according to claim 6, wherein said action classification recognition unit comprises:

acquiring segmented evoked potential sample data, and adding the segmented evoked potential sample data into a database;

performing classification training according to the segmented evoked potential sample data in the database to obtain a plurality of local features;

assembling the multidimensional local features into complete one-dimensional features through a matrix by adopting a full connection layer according to the acquired segmented evoked potential data;

and mapping the one-dimensional features into the probability of rehabilitation actions according to the soft-max layer, so as to realize the identification of the actions.

8. The brain electrical biofeedback recovery system according to claim 7, wherein a CART decision tree is used for classification training.

9. The brain wave biofeedback recovery system according to claim 6, wherein high and low frequency noise in the most proximal evoked potential signal is removed by adding a band pass IIR filter.

10. The brain electrical biofeedback recovery system according to claim 6, wherein the calculation formula of the nearest evoked potential signal EP (n) is:

EP(n)＝S(n)-W*REF_M(n)

s (n) is an acquired stimulation signal data sequence;

REF_M(n) is maximumWith respect to the reference sequence, the best value for ep (n) can be obtained by adjusting the W matrix, and the value of W is calculated according to the following formula:

W＝∑S(n)*REF_M(n)/∑[REF_M(n)]²。

Technical Field

The invention relates to a brain electrical biofeedback rehabilitation method and system.

Background

At present, the neural rehabilitation is mostly in an open loop state, and the brain does not participate in the autonomous rehabilitation process. Most utilize electrical stimulation and mechanical motion for rehabilitation exercises, relying too much on external passive rehabilitation. The recovery period is long.

Disclosure of Invention

The invention aims to solve the defects of the prior art and provides a brain electrical biofeedback rehabilitation method, which comprises the following steps:

obtaining a reference signal REF (n), wherein the length of the reference signal is about 50-100 times of the length of the evoked potential signal;

collecting target evoked potential signals X (n);

obtaining a closest evoked potential signal through an IIR interference cancellation algorithm;

the high and low frequency noise in the eliminated closest evoked potential signal results in an evoked potential signal E P (n).

The evoked potential signal sequence EP (n) is segmented according to the following frequency: the delta frequency band is 0.5 Hz-4 Hz, the theta frequency band is 4 Hz-8 Hz, the alpha frequency band is 8 Hz-13 Hz, and the beta frequency band is 13 Hz-30 Hz; obtaining a segmented evoked potential signal;

the segmented evoked potential signals are subjected to wavelet transformation and windowing processing to obtain segmented evoked potential data,

the action classification and identification are carried out according to the segmented evoked potential data,

FES electrical stimulation is performed according to the recognized action, and video feedback is performed.

An electroencephalogram biofeedback-derived rehabilitation system, the rehabilitation system comprising:

a reference signal acquisition unit for acquiring a reference signal REF (n), the length of the reference signal being approximately 50-100 times the length of the evoked potential signal;

an evoked potential signal acquisition unit for acquiring a target evoked potential signal X (n);

the interference cancellation processing unit obtains a signal closest to the evoked potential through an IIR interference cancellation algorithm;

and a high-low frequency noise elimination unit for eliminating high-frequency and low-frequency noise in the most approximate evoked potential signal to obtain an evoked potential signal EP (n).

A segmenting unit for segmenting the evoked potential signal sequence EP (n) according to the following frequency: delta frequency band is 0.5 Hz-4 Hz, theta frequency band is 4 Hz-8 Hz, alpha frequency band is 8 Hz-13 Hz, beta frequency band is 13 Hz-30H z; obtaining a segmented evoked potential signal;

a processing unit for performing wavelet transformation and windowing on the segmented evoked potential signals to obtain segmented evoked potential data,

an action recognition unit for classifying and recognizing actions according to the segmented evoked potential data,

and the feedback unit is used for implementing FES electrical stimulation according to the recognized action and carrying out video feedback.

The nearest evoked potential signal EP (n) is calculated by the formula:

EP(n)＝S(n)-W*REF_M(n)

s (n) is an acquired stimulation signal data sequence; REF_M(n) is the maximum correlation reference sequence, ep (n) can be optimized by adjusting the W matrix, and the value of W is calculated according to the following formula:

W＝∑S(n)*REF_M(n)/∑[REF_M(n)]²。

the brain in the rehabilitation system of the invention participates in the autonomous rehabilitation process, and the rehabilitation time can be shortened.

The above-described and other features, aspects, and advantages of the present application will become more apparent with reference to the following detailed description.

Drawings

Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:

FIG. 1 is a schematic diagram of a brain electrical biofeedback rehabilitation system.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the described embodiments of the invention without any inventive step, are within the scope of protection of the invention.

Unless defined otherwise, technical or scientific terms used herein shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention belongs. The use of "first," "second," and similar terms in the description and claims of the present application do not denote any order, quantity, or importance, but rather the terms are used to distinguish one element from another. Also, the use of the terms "a" or "an" and the like do not denote a limitation of quantity, but rather denote the presence of at least one.

A brain electrical biofeedback rehabilitation method comprises the following steps:

obtaining a reference signal REF (n), wherein the length of the reference signal is about 50-100 times of the length of the evoked potential signal; collecting target evoked potential signals X (n); obtaining a closest evoked potential signal through an IIR interference cancellation algorithm;

the calculation of the nearest evoked potential signal EP (n) is disclosed below:

EP(n)＝X(n)-REF(n)

where x (n) is the integrated signal, REF (n) is the reference signal,

x (n) and REF (n) are autocorrelation functions, so:

R(m)＝X(n)·REF(n)

where m is the number of activity points (left to right) of the sequence of reference signals, the sequence of maximum auto-correlation is REF_M(n), "·" is the associative operator, R (m) is the phaseA relationship array;

the value of the maximum correlation coefficient is found as:

R＝MAX[R(m)]

the adaptive relationship matrix is disclosed as follows:

EP(n)＝S(n)-W*REF_M(n)

where W is the adjustment matrix. REF_M(n) is a maximum correlation reference sequence, and EP (n) can obtain an optimal value by adjusting the W matrix, and S (n) is an acquired stimulation signal data sequence;

according to the formula:

W＝∑S(n)*REFM(n)/∑[REF_M(n)]²

after the value of w is found out,

according to the following formula

EP(n)＝S(n)-W*REF_M(n)

The interference is mutually eliminated, and the nearest evoked potential signal EP (n) can be obtained.

The high and low frequency noise in the eliminated closest evoked potential signal results in an evoked potential signal E P (n). High and low frequency noise in the nearest evoked potential signal is eliminated by adding a band pass IIR filter.

The evoked potential signal sequence EP (n) is segmented according to the following frequency: the delta frequency band is 0.5 Hz-4 Hz, the theta frequency band is 4 Hz-8 Hz, the alpha frequency band is 8 Hz-13 Hz, and the beta frequency band is 13 Hz-30 Hz; obtaining a segmented evoked potential signal; performing wavelet transformation and windowing on the segmented evoked potential signals to obtain segmented evoked potential data;

wavelet Transform (WT) is a new transform analysis method, which inherits and develops the idea of short-time Fourier transform localization, and overcomes the disadvantage that the window size does not change with frequency, etc., and can provide a time-frequency window changing with frequency, and is an ideal tool for signal time-frequency analysis and processing. The method is mainly characterized in that the characteristics of certain aspects of the problem can be fully highlighted through transformation, the time (space) frequency can be locally analyzed, signals (functions) are gradually subjected to multi-scale refinement through telescopic translation operation, finally, high-frequency time refinement and low-frequency refinement are achieved, the requirements of time-frequency signal analysis can be automatically adapted, therefore, any details of the signals can be focused, the problem of difficulty of Fourier transformation is solved, and the method becomes a major breakthrough on a scientific method following the Fourier transformation.

In signal processing, it can be said that windowing is a necessary process, because our computer can only process signals of limited length, so the original signal x (T) is truncated by T (sampling time), i.e. limited, and further processed after xt (T), this process sequence is windowing, and for the case of non-whole period sampling, it must be considered how to reduce the leakage error caused by windowing, and the main measure is to use a reasonable windowing function to passivate the sharp angle of signal truncation, thereby minimizing the spread of spectrum.

And performing action classification and identification according to the segmented evoked potential data, performing FE S (electro-stimulation) according to the identified action, and performing video feedback. The action classification recognition comprises the following steps: acquiring segmented evoked potential sample data, and adding the segmented evoked potential sample data into a database; performing classification training according to the segmented evoked potential sample data in the database to obtain a plurality of local features; assembling the multidimensional local features into complete one-dimensional features through a matrix by adopting a full connection layer according to the acquired segmented evoked potential data; and mapping the one-dimensional features into the probability of rehabilitation actions according to the soft-max layer, so as to realize the identification of the actions. And carrying out classification training by adopting a CART decision tree.

As shown in fig. 1, a brain electrical biofeedback rehabilitation system, the rehabilitation system comprises: a reference signal acquisition unit for acquiring a reference signal REF (n), the length of the reference signal being approximately 50-100 times the length of the evoked potential signal; an evoked potential signal acquisition unit for acquiring a target evoked potential signal X (n); the interference cancellation processing unit obtains a signal closest to the evoked potential through an IIR interference cancellation algorithm; and a high-low frequency noise elimination unit for eliminating high-frequency and low-frequency noise in the most approximate evoked potential signal to obtain an evoked potential signal EP (n). A segmenting unit for segmenting the evoked potential signal sequence EP (n) according to the following frequency: the delta frequency band is 0.5 Hz-4 Hz, the theta frequency band is 4 Hz-8 Hz, the alpha frequency band is 8 Hz-13 Hz, and the beta frequency band is 13 Hz-30 Hz; obtaining a segmented evoked potential signal; the processing unit is used for performing wavelet transformation and windowing on the segmented evoked potential signals to obtain segmented evoked potential data, the action recognition unit is used for performing action classification recognition according to the segmented evoked potential data, and the feedback unit is used for performing FES (feed forward) electrical stimulation according to recognized actions and performing video feedback.

As a preferred embodiment, the motion classification recognition unit includes: acquiring segmented evoked potential sample data, and adding the segmented evoked potential sample data into a database; performing classification training according to the segmented evoked potential sample data in the database to obtain a plurality of local features; assembling the multidimensional local features into complete one-dimensional features through a matrix by adopting a full connection layer according to the acquired segmented evoked potential data; and mapping the one-dimensional features into the probability of rehabilitation actions according to the soft-max layer, so as to realize the identification of the actions. And carrying out classification training by adopting a CART decision tree.

The brain in the rehabilitation system of the invention participates in the autonomous rehabilitation process, and the rehabilitation time can be shortened.

While there have been shown and described what are at present considered the fundamental principles and essential features of the invention and its advantages, it will be apparent to those skilled in the art that the invention is not limited to the details of the foregoing exemplary embodiments, but is capable of other specific forms without departing from the spirit or essential characteristics thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.

Furthermore, it should be understood that although the present description refers to embodiments, not every embodiment may contain only a single embodiment, and the description is given here only for clarity, and those skilled in the art should integrate the description, and the embodiments may be combined appropriately to form other embodiments understood by those skilled in the art.