CN113855023B

CN113855023B - Iterative tracing-based lower limb movement BCI electrode selection method and system

Info

Publication number: CN113855023B
Application number: CN202111249220.4A
Authority: CN
Inventors: 彭小波; 宋霖; 刘俊宏
Original assignee: Shenzhen University
Current assignee: Shenzhen University
Priority date: 2021-10-26
Filing date: 2021-10-26
Publication date: 2023-07-04
Anticipated expiration: 2041-10-26
Also published as: CN113855023A

Abstract

The invention discloses a lower limb movement BCI electrode selection method and system based on iterative tracing, wherein the method comprises the following steps: s1: selecting a tracing initial electrode; s2: setting a plurality of tested, calculating and selecting a tracing time point, and tracing the source of each tested initial electrode to obtain a tracing characteristic distribution diagram; s3: calculating an overlapped average map; s4: setting a pixel threshold value and processing the superimposed average graph to obtain a tracing result graph; s5: superposing the tracing result diagram and the electrode-brain diagram to obtain an electrode selected by the characteristic distribution of the tracing; s6: and (3) tracing the electrode obtained in the step (S5) as the initial electrode of the next tracing to obtain the electrode selected by the tracing result, ending tracing until the electrode is the same as the electrode selected by the last tracing result, and taking the electrode selected by the tracing result as the optimal electrode of the lower limb movement BCI. The optimal electrode selected by the invention can extract the characteristic information with the most distinguishing degree, and further improves the identification accuracy and the running speed of the BCI system.

Description

Iterative tracing-based lower limb movement BCI electrode selection method and system

Technical Field

The invention relates to the technical field of brain-computer interfaces, in particular to a lower limb movement BCI electrode selection method and system based on iterative tracing.

Background

Along with the aggravation of the aging problem, various cerebral diseases are increased, such as cerebral apoplexy and the like, which cause great inconvenience to the life of people. In addition, some accidents are difficult to avoid, such as brain injury caused by traffic accidents, resulting in partial or complete disability. The above patients all have in common: brain function is sound but neuromuscular channels are impaired. Scientists have been eager in recent years to study in the area of brain-computer interfaces (BCI) in an attempt to solve this problem by BCI technology. Meanwhile, the development of the BCI technology is greatly promoted by the rapid development of the computer technology, the continuous maturation of the artificial intelligence technology and the demands in the fields of traffic, medical treatment and the like. The BCI technology refers to a system which is not dependent on human muscle tissue and conventional nerve pathways, and a direct pathway is created between the human brain and external equipment through technical means, so that the brain and external environment are controlled in an exchange mode. From a macroscopic view, the research of the BCI technology is helpful for the vigorous development of the rehabilitation medical field and the public transportation industry, and simultaneously helps the high-speed development of artificial intelligence in China. In the application level, the BCI technology is widely applied to the fields of medical treatment, traffic, military, exoskeleton robots and the like, and is a development assistance for various industries. However, BCI itself still has many technical problems to be solved, such as insufficient recognition accuracy, poor real-time performance, low response speed, etc., and is still in laboratory research stage at present, and is not popularized in people's daily life. The problem of identification accuracy is a key problem of the BCI technology, and determines whether the BCI technology can be further put into practical use to a certain extent. In the process of processing signals by the BCI system, the links of signal preprocessing, electrode selection, feature extraction and classification recognition all have influence on the recognition accuracy of the BCI system.

The electrode selection includes upper limb motor imagery electrode selection and lower limb motor imagery electrode selection. Because the movement cortex area associated with the left hand and the right hand in the movement of the upper limb has obvious distinction in anatomy, the selected electrode is definite in electrode selection; in the lower limb movement, the movement cortex areas associated with the left foot and the right foot are close to each other and partially overlapped anatomically, so that the classification difficulty of the lower limb movement imagination is higher, and a relatively perfect lower limb movement imagination electroencephalogram signal electrode selection method is not available at present.

Most of the current researches on the optimal electrode of the lower limb motor imagery still adopt the same method as the upper limb motor imagery, namely the electrode of the motion area is directly selected, and the electrodes are usually C3, C4 and Cz electrodes. In the prior art, the publication number is: CN101433460a chinese patent invention discloses a lower limb imagination action potential spatial filtering method in 5 months and 20 days 2009, the scheme establishes a wavelet packet independent component analysis method for imagination action potential spatial filtering by using C3, C4 and Cz electrodes for imagination action potentials induced by 3 kinds of key compound lower limb imagination actions (imagination standing up, imagination left hand and same side single leg cooperative action, imagination right hand and same side single leg cooperative action), and meets the spatial resolution requirement of compound lower limb imagination action potential research. Although effective electroencephalogram characteristics can be extracted to a certain extent by extracting lower limb motor imagery electroencephalogram signals of the C3, C4 and Cz electrodes, the characteristics are not optimal.

Disclosure of Invention

The invention provides a lower limb movement BCI electrode selection method and system based on iterative tracing, which are used for overcoming the defect that the existing lower limb movement imagination brain electrical signal electrode selection method cannot obtain an optimal electrode.

The primary purpose of the invention is to solve the technical problems, and the technical scheme of the invention is as follows:

a lower limb movement BCI electrode selection method based on iterative tracing comprises the following steps:

s1: selecting a tracing initial electrode;

s2: setting a plurality of tested, calculating the time point with the maximum amplitude on each tested time domain, taking the time point as the tracing time point, tracing each tested initial electrode to obtain a tracing characteristic distribution diagram of each tested;

s3: calculating a superposition average map by using the tested tracing characteristic distribution map;

s4: setting a pixel threshold value, and processing the superimposed average graph by using the pixel threshold value to obtain a tracing result graph;

s5: superposing the tracing result graph and the electrode-brain graph to obtain an electrode-brain graph containing tracing distribution, wherein an electrode surrounded by a tracing result area is an electrode selected by the tracing characteristic distribution;

s6: and (3) taking the electrode obtained by tracing in the step (S5) as an initial electrode of the next tracing, carrying out the next tracing to obtain an electrode selected by a tracing result until the electrode selected by the tracing result is the same as the electrode selected by the last tracing result, ending tracing, and taking the electrode selected by the tracing result as the optimal electrode of the lower limb movement BCI.

Further, the specific process of step S1 is as follows: and respectively extracting brain electrical signals of the middle 17 leads and the whole brain leads from all the tested, performing short-time Fourier transform, and then performing baseline correction on video domain signals after the short-time Fourier transform to respectively obtain time-frequency diagrams of the whole brain electrodes and the middle 17 leads, wherein an electrode with obvious ERD phenomenon in a motor imagery time period in the time-frequency diagrams is an initial electrode.

Further, the time for baseline correction was 1s before the task.

Further, by setting a uniform amplitude threshold in the traceability feature distribution map in step S2, only the region with the amplitude higher than the maximum amplitude multiplied by the amplitude threshold is displayed in each tested traceability feature distribution map.

Further, in step S2, by drawing time domain contour diagrams of different leads, a time point with the maximum amplitude is determined as a time point of electroencephalogram tracing.

Further, in step S3, the specific process of calculating the overlay average map by using the tested traceability feature distribution map is as follows:

setting a blank template of the whole brain region, performing difference processing on each tested tracing characteristic distribution map and the blank template, and superposing and averaging all images subjected to the difference processing to obtain a superposition average map.

Further, setting a pixel threshold value, and processing the superimposed average graph by using the pixel threshold value to obtain a tracing result graph, wherein the specific process is as follows:

setting a threshold pixel value, extracting the superposition average graph by using the pixel threshold value, and subtracting the extracted image by using a white template to obtain a tracing result graph.

Further, the white template is an image template with a pixel value of 0.

Further, the pixel threshold value is the same every time tracing.

The second aspect of the invention provides a lower limb movement BCI electrode selection system based on iterative tracing, which comprises: the device comprises a memory and a processor, wherein the memory comprises a lower limb movement BCI electrode selection method program based on iterative tracing, and the lower limb movement BCI electrode selection method program based on iterative tracing realizes the following steps when being executed by the processor:

s1: selecting a tracing initial electrode;

Compared with the prior art, the technical scheme of the invention has the beneficial effects that:

according to the invention, by means of iterative tracing, the electrode selected by the previous iterative tracing result is used as the initial electrode of the next tracing until the electrodes selected by the two times of tracing results are the same, tracing is ended, and the electrode selected by the current tracing result is used as the optimal electrode of the lower limb movement BCI.

Drawings

Fig. 1 is a flowchart of a lower limb movement BCI electrode selection method based on iterative tracing.

Fig. 2 is an intermediate 17-lead diagram of an embodiment of the invention.

Fig. 3 is a time-frequency diagram of an overall brain electrode according to an embodiment of the present invention.

Fig. 4 is a time-frequency diagram of an intermediate 17-electrode according to an embodiment of the present invention.

Fig. 5 is a diagram of a motor imagery tracing result according to an embodiment of the present invention.

Fig. 6 is a full brain lead time domain contour diagram of an embodiment of the present invention.

Fig. 7 is a diagram showing the first trace-out feature distribution of 20 subjects according to the embodiment of the present invention.

FIG. 8 is a blank template diagram according to an embodiment of the invention.

FIG. 9 is a difference plot of 20 tested first-time tracing results and blank templates according to an embodiment of the present invention.

Fig. 10 is a pixel distribution diagram of 20 post-test image overlay average images according to an embodiment of the present invention.

Fig. 11 is a superimposed average of 20 images after trial differences according to the embodiment of the present invention.

Fig. 12 is a superimposed average diagram after setting a pixel threshold according to an embodiment of the present invention.

Fig. 13 is a first tracing result diagram according to an embodiment of the present invention.

Fig. 14 is an electrode-brain diagram of an embodiment of the present invention.

FIG. 15 is an electrode-brain graph with traceable distribution according to an embodiment of the present invention.

FIG. 16 is a schematic diagram of electrodes selected for a first trace-out distribution in accordance with an embodiment of the present invention.

Fig. 17 is a second tracing result diagram according to an embodiment of the present invention.

FIG. 18 is a diagram of electrodes selected for a second trace-out distribution according to an embodiment of the present invention.

Fig. 19 is a third tracing result diagram according to an embodiment of the present invention.

FIG. 20 is a schematic diagram of electrodes selected for a third trace-out distribution according to an embodiment of the present invention.

Detailed Description

In order that the above-recited objects, features and advantages of the present invention will be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description. It should be noted that, in the case of no conflict, the embodiments of the present application and the features in the embodiments may be combined with each other.

In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those described herein, and therefore the scope of the present invention is not limited to the specific embodiments disclosed below.

Noun interpretation:

tested: referring to the subject of the psychological test or examination, the observed psychological phenomenon or behavioral trait may be produced or displayed.

Example 1

As shown in fig. 1, the lower limb movement BCI electrode selection method based on iterative tracing includes the following steps:

s1: selecting a tracing initial electrode;

it should be noted that, the iterative tracing is to select a plurality of electrodes above the average result of the previous tracing of the tested to trace again, and verify and lock the result area as the final tracing result area after multiple iterations.

When iterative tracing is carried out, firstly, tracing initial electrodes are selected, the electroencephalogram signal quality of electrodes in the middle area of the brain is always higher than that of electrodes in the peripheral area, noise interference such as head movement is small, and time-frequency analysis can reveal the vibration activities related to events by measuring the power changes of different frequency bands on each electrode. In the embodiment of the invention, all the tested electroencephalogram signals of the same task are averaged, and time-frequency analysis is carried out to obtain all the tested time-frequency spectrograms. The brain electrical signals of the intermediate 17 leads (shown in figure 2) and the whole brain leads are respectively extracted and subjected to short-time Fourier transform (STFT), so that the brain electrical signals can be converted into time-frequency domain display from time sequence. Meanwhile, in order to further improve the accuracy of the signals, a baseline correction method is adopted to reduce the influence of signal drift, and the baseline correction time is 1s before a task.

As shown in fig. 3 and 4, which are respectively time-frequency diagrams of the whole brain electrode and the middle 17 conducting electrode, in the time-frequency diagrams, it can be seen that the ERD phenomenon of the middle 17 conducting electrode is more obvious than that of the whole brain electrode in the motor imagery time period, so that the middle 17 conducting electrode is selected as the initial electrode for tracing.

it should be noted that, the electroencephalogram features at different moments are different, so that the electroencephalogram tracing results at different time points are different, for example, the tracing results at the rest state and the motor imagery moment are relatively different, the difference is reflected on the distribution of the electroencephalogram tracing feature map, the distribution of the rest state tracing results is not regular, and the tracing results when the lower limb movement or the lower limb movement imagery are generally concentrated in the middle area of the brain, as shown in fig. 5. The color represents the current density value of the source, and the brighter the color (the larger the gray value) indicates the more active the electroencephalogram activity there.

The distribution of different time point sources and the amplitude of the sources are different in the motor imagery period, and the time point with the maximum amplitude is determined as the time point of electroencephalogram tracing by drawing time domain contour maps of different leads, so that the electroencephalogram signal is strongest, and the tracing result is more obvious. The time corresponding to the maximum amplitude is also different because the reaction time to the stimulus, that is, the time point at which the motor imagery starts, is different from one test to another. Fig. 6 shows a time domain contour diagram of a test, where t=0.302 s is the maximum amplitude.

In the embodiment of the invention, 20 tested electrodes are arranged, the time point with the maximum amplitude value on each tested time domain is calculated, the middle 17 conducting electrode is taken out from each tested electrode, the source tracing is carried out on each tested initial electrode, and the source tracing characteristic distribution diagram of each tested is obtained, as shown in fig. 7; meanwhile, by setting a unified amplitude threshold value to be 10% in the traceability feature distribution map, only the area with the amplitude higher than the maximum amplitude multiplied by 10% is displayed in each tested traceability feature distribution map.

after 20 tested traceability feature distribution diagrams are obtained, a superposition average diagram is obtained through superposition averaging, and the specific steps are as follows:

setting a blank template of the whole brain region, performing difference processing on each tested tracing characteristic distribution map and the blank template, and superposing and averaging all images subjected to the difference processing to obtain a superposition average map. In order to reduce the influence of the brain template during image processing, as shown in fig. 8, the source amplitude of the blank template is 0, each tested tracing feature distribution diagram is subjected to difference processing with the blank template diagram, and the image after difference processing is subjected to gray-scale processing, and the obtained image is shown in fig. 9.

in a specific embodiment, to effectively reduce feature dimensions, fully extract features, and properly reduce the computational effort of the algorithm, typically no more than 5 electrodes are used in the early selection of electrodes. Since the 64 electrodes are distributed over the brain surface in a relatively uniform area, the total area of the brain surface when 5 electrodes are selected is about 7.81%. Fig. 10 is a distribution diagram of 20 pixel distribution diagrams of the image superimposed average image after the test difference, in which the pixel values are integrated from high to low, and the pixel distribution accounts for 7.81% of the total distribution when the pixel value is higher than 173. Then, the pixel threshold is obtained by subtracting the pixel value 173 from each pixel point in fig. 11, and the distribution of the result obtained by subtracting the pixel value is more concentrated. The superimposed average map obtained by subtracting the pixel threshold is multiplied by 10 as a whole, and fig. 12 is a superimposed average map obtained by setting the pixel threshold. Further, in order to make the background white, this image is subtracted from a white template (the white template is an image template having a pixel value of 0), to obtain the image as shown in fig. 13. The feature distribution after the pixel threshold selection is a part with larger gray value, namely a part with larger brain current density and more active brain, and as can be seen from fig. 13, the feature distribution is concentrated in the middle area of the brain.

in one embodiment, an electrode-brain map is shown in fig. 14, and an electrode-brain map with traceable distribution is shown in fig. 15. Finally, the electrodes surrounded by the tracing result area are taken as the final electrodes, and as shown in fig. 16, the electrodes are 6 electrodes FC1, FC2, C1, cz, C2 and CP2 in the box respectively. The result is an electrode selected according to the characteristic distribution of the first trace, and at the same time, the electrode is used as an initial electrode in the second trace.

It should be noted that, after the first tracing is finished, the electrode selected by the first tracing result is used as the initial electrode of the second tracing, and meanwhile, the pixel threshold is also kept consistent with that of the first tracing, the pixel threshold is 173, as shown in fig. 17, to obtain a second tracing result graph, as shown in fig. 18, the second tracing result is distributed to the selected electrode, and is more concentrated in the middle area than the first tracing result distribution. The second tracing result graph is superimposed with the electrode-brain graph to obtain fig. 18. And selecting the electrodes FC1, FC2, C1, cz and C2 according to the tracing result to serve as initial electrodes in the third tracing.

The 3 rd tracing step is the same as the first 2 times, the pixel value threshold value is taken 173, the obtained final tracing result is shown in fig. 19, and is more concentrated than the 2 nd tracing result distribution, and is overlapped with the electrode-brain graph to obtain fig. 20. And selecting the electrodes FC1, FC2, C1, cz and C2 according to the tracing result, finding that the electrodes are the same as the electrodes obtained by the 2 nd tracing, ending the tracing, and ending the iteration.

According to the embodiment, through the 3 times of tracing, the optimal electrode which is finally selected according to the iteration result and is used as the lower limb motor imagery experiment is: FC1, FC2, C1, cz, C2.

The second aspect of the invention provides a lower limb movement BCI electrode selection system based on iterative tracing, which comprises: the device comprises a memory and a processor, wherein the memory comprises a lower limb movement BCI electrode selection method program based on iterative tracing, and the lower limb movement BCI electrode selection method program based on iterative tracing realizes the following steps when being executed by the processor: s1: selecting a tracing initial electrode;

Verification and analysis

In the embodiment of the invention, an SVM model classifier is established for classification recognition, four-domain fusion features are extracted based on FC1, FC2, C1, cz and C2 electrodes, and PSO-SVM is adopted for classification recognition. The 10 classification recognition results of the test set were averaged as final recognition results, as shown in table 1. The comparative analysis shows that: (1) The longitudinal results show that when the same electrode and the same classification method are used, the classification accuracy is improved by comparing the multi-domain fusion characteristics provided by the embodiment with the single CSP characteristics. (2) And the transverse result shows that when the same electrode and the same feature extraction method are used, the SVM algorithm after particle swarm optimization improves the recognition accuracy compared with the traditional SVM algorithm. (3) When the same feature extraction and classification algorithm is used, the FC1, FC2, C1, cz and C2 electrodes have higher recognition accuracy than the C3, C4 and Cz electrodes. (4) Compared with the whole table, the multi-domain fusion characteristics are extracted based on the FC1, FC2, C1, cz and C2 electrodes, and the highest accuracy 88.4274% is obtained by adopting a PSO-SVM classification and identification method, which is obviously higher than other methods.

Table 1 different methods classification recognition results table

As can be seen from Table 1, based on the extraction of multi-domain fusion characteristics of the FC1, FC2, C1, cz, C2 electrodes and the C3, C4 and Cz electrodes, the classification recognition accuracy by using the PSO-SVM is 88.4274% and 87.8205%, respectively. Therefore, the identification accuracy of the FC1, FC2, C1, cz and C2 electrodes selected by iterative tracing is higher than that of the traditional electrodes.

Through designing an electroencephalogram experiment and collecting electroencephalogram data, five electrodes FC1, FC2, C1, cz and C2 are selected based on an iterative tracing method, after experimental data are preprocessed, four-domain fusion characteristics of a time domain, a frequency domain, a time frequency and a space domain are extracted respectively for the five electrodes and the electrodes C3, C4 and Cz of a conventional motion area, finally, the extracted characteristics are classified and identified by adopting a PSO-SVM classification method, and the fact that the five electrodes selected through iterative tracing have higher identification precision, namely the electrodes which are more suitable for the electroencephalogram experiment of lower limb motor imagination are FC1, FC2, C1, cz and C2 is proved.

It is to be understood that the above examples of the present invention are provided by way of illustration only and not by way of limitation of the embodiments of the present invention. Other variations or modifications of the above teachings will be apparent to those of ordinary skill in the art. It is not necessary here nor is it exhaustive of all embodiments. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the invention are desired to be protected by the following claims.

Claims

1. The lower limb movement BCI electrode selection method based on iterative tracing is characterized by comprising the following steps of:

s1: selecting a tracing initial electrode;

the specific process is as follows: extracting brain electrical signals of the middle 17 leads and the whole brain leads respectively from all the tested brain electrical signals and performing short-time Fourier transform, and then performing baseline correction on time-frequency domain signals after the short-time Fourier transform to obtain time-frequency diagrams of the whole brain electrodes and the middle 17 leads respectively, wherein an electrode with obvious ERD phenomenon in a motor imagery time period in the time-frequency diagrams is an initial electrode;

the specific process is as follows: setting a blank template of a whole brain region, performing difference processing on each tested tracing characteristic distribution map and the blank template, and superposing and averaging all images subjected to the difference processing to obtain a superposition average map;

2. The method for selecting the lower limb movement BCI electrode based on iterative tracing according to claim 1, wherein the time for baseline correction is 1s before the task.

3. The method for selecting the lower limb movement BCI electrode based on iterative tracing according to claim 1, wherein in the tracing feature distribution map of step S2, only the region with the amplitude higher than the maximum amplitude multiplied by the amplitude threshold is displayed in each tested tracing feature distribution map by setting a uniform amplitude threshold.

4. The method for selecting the lower limb movement BCI electrode based on iterative tracing according to claim 1, wherein in step S2, the time point with the maximum amplitude is determined as the electroencephalogram tracing time point by drawing time domain contour diagrams of different leads.

5. The method for selecting the lower limb movement BCI electrode based on iterative tracing according to claim 1, wherein the method is characterized by setting a pixel threshold value, and processing the superposition average graph by using the pixel threshold value to obtain a tracing result graph, and comprises the following specific steps: setting a threshold pixel value, extracting the superposition average graph by using the pixel threshold value, and subtracting the extracted image by using a white template to obtain a tracing result graph.

6. The method for selecting the lower limb movement BCI electrode based on iterative tracing according to claim 5, wherein the white template is an image template with a pixel value of 0.

7. The method for selecting the lower limb movement BCI electrode based on iterative tracing of claim 1, wherein the pixel threshold value is the same every time tracing.

8. Lower limb movement BCI electrode selection system based on iterative tracing is characterized in that the system comprises: the device comprises a memory and a processor, wherein the memory comprises a lower limb movement BCI electrode selection method program based on iterative tracing, and the lower limb movement BCI electrode selection method program based on iterative tracing realizes the following steps when being executed by the processor:

s1: selecting a tracing initial electrode;