KR102215807B1 - Method for analysing ear brain wave using re-reference method - Google Patents

Abstract

The present invention relates to an ear EEG analysis method using a re-reference method, comprising attaching a reference electrode and a plurality of measurement electrodes to the body, and maintaining a high cognitive state by repeating for a preset time Measuring each ear EEG corresponding to a low cognitive state maintaining a state, removing a noise component for each measured ear EEG, and re-referenced to each of the ear EEG from which the noise component has been removed Applying the method and extracting the features of the frequency band using a common space pattern algorithm for each data to which the re-reference method was applied, and estimating the classification accuracy of the ear EEG measured in the high and low perception states It may include the step of.

Description

Method for analysing ear brain wave using re-reference method}

The present invention relates to an ear EEG analysis method using a re-reference method, and more particularly, to an ear EEG analysis method using a re-reference method capable of efficiently measuring ear EEG.

As is well known, electroencephalography (EEG), also referred to as electroencephalography, refers to a voltage difference between a signal measured at an electrode set as a reference electrode and a signal measured at each other measurement electrode.

Such EEG can be classified into, for example, Evoked Potential, EP or Event Related Potential (ERP) generated by external stimulation, and Spontaneous EEG, which occurs spontaneously without external stimulation.

EEG can be used not only in industrial fields for diagnosis, rehabilitation, clinical research, etc. of patients with brain-related mental disorders, but also in daily life in various ways. For example, neurofeedback refers to a process in which a user generates an EEG in a desired direction, which can be used in an education field such as brain training or a game field using EEG. In addition, the brain computer interface refers to an interface that detects the user's intentions or thoughts by measuring and analyzing brain waves, which is a speller (Typing) that recognizes the user's intention and types letters. Speller), etc.

Here, the position of the reference electrode is different for each measuring device in the ear EEG measured around or inside the ear.

In addition, the position of the reference electrode has a great influence on the measurement signal, and since the amplitude of the measured signal decreases as the reference electrode is closer to the measurement electrode, incorrect reference electrode setting results in a decrease in signal quality.

1. Korean Patent Registration No. 10-1791038 (registered on October 23, 2017) 2. Korean Patent Registration No. 10-1032913 (registered on April 27, 2011)

The present invention relates to an ear EEG analysis method using a re-standard method that is superior in ease of use and practicality than the conventional EEG measurement method in which electrodes are attached to the scalp surface.

In addition, the present invention relates to an ear EEG analysis method using a re-reference method capable of increasing the analysis accuracy of ear EEG through the re-reference method.

The objects of the embodiments of the present invention are not limited to the above-mentioned objects, and other objects not mentioned will be clearly understood by those of ordinary skill in the technical field to which the present invention belongs from the following description. .

According to an embodiment of the present invention, the steps of attaching a reference electrode and a plurality of measurement electrodes to the body, and a high cognitive state maintaining a high cognitive state and a low cognitive state maintaining a low cognitive state are performed repeatedly for a preset time. Measuring each corresponding ear EEG, removing a noise component for each of the measured ear EEG, applying a re-reference method to each of the ear EEG from which the noise component has been removed, and the And extracting a characteristic of a frequency band using a common space pattern algorithm for each data to which the re-reference method is applied, and estimating the classification accuracy of the ear EEG measured in the high perception and low perception states, respectively.

In addition, in the removing of the noise component, a change voltage at both ends of the waveform due to blinking of an eye is checked through a visual inspection, and a signal of the ear EEG corresponding to a case higher than an average value of the change voltage at both ends may be removed. .

In addition, the re-reference method includes: Common average reference, Contralateral, Ipsilateral, Bipolar-Contralateral, and bipolar-Ipsilateral , It may be characterized in that at least one of the overall reference (Bipolar-All) of the polarity difference.

In addition, the frequency band may be at least one of a delta wave, theta wave, an alpha wave, a beta wave, and a gamma wave.

In addition, the high awareness state may be characterized by performing mental arithmetic.

In addition, the low recognition state may be characterized by imagining a song.

In the present invention, by attaching the electrode to the skin around the ear, not the scalp, it is possible to improve the convenience and aesthetics of use when developing an EEG-based application technology.

In addition, the present invention can increase the analysis accuracy of ear EEG by reducing the influence of the position of the reference electrode through the re-reference method.

1 is a flow chart showing a process of analyzing an ear EEG through a re-reference method according to an embodiment of the present invention,
2 is a view showing a reference electrode and a measurement electrode attachment position for measuring an ear EEG according to an embodiment of the present invention,
3 is a diagram showing an experimental paradigm for measuring EEG data necessary to develop a re-reference method according to an embodiment of the present invention,
4 is a diagram showing a data processing method for each re-reference method according to an embodiment of the present invention,
5 is a diagram showing basic data for each experiment participant of an ear EEG analysis method using a re-reference method according to an embodiment of the present invention.
6 and 7 are diagrams showing classification accuracy for each re-reference method according to an embodiment of the present invention.

Advantages and features of the embodiments of the present invention, and a method of achieving them will become apparent with reference to the embodiments described later in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in a variety of different forms, and only these embodiments make the disclosure of the present invention complete, and are common knowledge in the technical field to which the present invention pertains. It is provided to completely inform the scope of the invention to those who have, and the invention is only defined by the scope of the claims. The same reference numerals refer to the same components throughout the specification.

In describing the embodiments of the present invention, if it is determined that a detailed description of a known function or configuration may unnecessarily obscure the subject matter of the present invention, a detailed description thereof will be omitted. In addition, terms to be described later are terms defined in consideration of functions in an embodiment of the present invention, which may vary according to the intention or custom of users or operators. Therefore, the definition should be made based on the contents throughout this specification.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a flow chart showing a process of analyzing an ear EEG through a re-reference method according to an embodiment of the present invention, and FIG. 2 is a reference electrode and a location of a measurement electrode for measuring an ear EEG according to an embodiment of the present invention. 3 is a diagram showing an experimental paradigm for measuring EEG data required to develop a re-reference method according to an embodiment of the present invention, and FIG. 4 is a re-reference method according to an embodiment of the present invention. Figure 5 is a diagram showing the basic data for each experiment participant in the ear EEG analysis method using the re-reference method according to an embodiment of the present invention, and Figures 6 and 7 are an embodiment of the present invention It is a diagram showing the classification accuracy by re-standard method according to an example.

Referring to FIG. 1, the ear EEG analysis method using the re-reference method includes the step 110 of attaching a reference electrode and a plurality of measurement electrodes to the body, and a high awareness of maintaining a high cognitive state by repeatedly performing for a preset time. A step 120 of measuring each ear EEG corresponding to a low cognitive state maintaining a state and a low cognitive state, removing a noise component for each measured ear EEG (130), and removing the noise component Step 140 of applying a re-reference method to each ear EEG, and extracting features of a frequency band using a common space pattern algorithm for each data to which the re-reference method was applied, and measured in high and low perception states, respectively. It may include the step 150 estimating the classification accuracy of the ear EEG.

A reference electrode and a plurality of measurement electrodes may be attached to the body (110).

Here, referring to FIG. 2, the reference electrode is attached to the front center of the scalp (FCz) based on the International 10-20 system proposed by the International Federation of Societies for Electroencephalograph and Clinical Neurophysiology. can do.

In addition, three measuring electrodes can be attached to each of the mastoid processes behind both ears.

Ear EEG corresponding to a high cognitive state maintaining a high cognitive state and a low cognitive state maintaining a low cognitive state may be measured by repeatedly performing for a preset time (120).

Here, the high cognitive state may be characterized by performing mental arithmetic, and the low cognitive state may be characterized by imagining a song.

In addition, referring to FIG. 3, it may include opening and resting eyes, receiving a task, performing a task, and resting after completing the task.

Here, at the stage of receiving the task, it may be presented to imagine an equation for mental arithmetic (eg, addition, subtraction, etc.) or a song (eg, an alphabet song).

In addition, the above-described steps of opening and resting eyes, receiving a task, performing a task, and resting after completing a task become one cycle, and each of the high and low cognitive states can be repeatedly performed 50 times, It can take 45-50 seconds per cycle.

Noise components for each measured ear EEG may be removed (130).

Here, a change voltage at both ends of the waveform due to blinking of an eye can be checked through a visual inspection, and a signal of an ear EEG corresponding to a case higher than the average value of the change voltage at both ends may be removed.

A re-reference method may be applied to each ear EEG from which noise components have been removed (140).

Here, the re-reference method is executed to remove the measurement influence by the position of the reference electrode. Common average reference, Contralateral, Ipsilateral, Bipolar-Contralateral ), the ipsilateral reference of the positive difference (Bipolar-Ipsilateral), and the overall reference of the polarity difference (Bipolar-All) may be characterized in that at least one.

2 and 4, the common average criterion is a method of subtracting an average of 6 signals to each of 6 electrodes when there are 6 ear electrodes (eg, L1-an average of all 6 electrodes).

The contralateral criterion is a method of subtracting the average of all three electrodes opposite to the target electrode from the signal measured at the target electrode (for example, L1-the average of the right three electrodes).

The ipsilateral criterion is a method of subtracting the average of all three electrodes on the same side (ipsilateral) as the target electrode from the signal measured at the target electrode (eg, L1-the average of the left three electrodes).

Bipolar refers to a difference signal between two electrodes, and the ear EEG analysis method using the re-reference method according to an embodiment of the present invention is divided according to whether the positions between the two electrodes are opposite or ipsilateral to each other. .

The opposite standard of the positive difference is a method of using the difference between the two electrodes in all possible pairs by ensuring that the two electrode pairs are located opposite each other (e.g., L1-R1, L1-R2, L1-R3, L2-R1, L2-R2, L2-R3, L3-R1, L3-R2, L3-R3).

The ipsilateral standard of the anode difference is a method of using the difference between the two electrodes in all possible pairs while keeping the two electrode pairs on the same side (e.g., L1-L2, L1-L3, L2-L3, R1-R2, R1- R3, R2-R3).

The overall criterion for the anode difference uses the difference between all pairs of electrodes that may occur between the six electrodes, and is equal to the sum of the contralateral criterion of the anode difference and the ipsilateral criterion of the anode difference.

For each data to which the re-reference method is applied, a common space pattern algorithm is used to extract a feature of a frequency band, and classification accuracy of ear EEG measured in high and low cognitive states can be estimated (150). .

Here, the common space pattern algorithm is an EEG feature extraction algorithm that applies Fukunaga-Koontz Transform (FKT), and finds features that best reveal differences between signals collected under two different conditions. .

In addition, the common space pattern algorithm is already widely known for its effectiveness in the brain-computer interface field, and is known to be particularly effective in solving the motor imagery task of classifying brain waves imagining motion.

In addition, brain waves are classified into delta waves (0.5 to 4 Hz), theta waves (4 to 8 Hz), alpha waves (8 to 13 Hz), beta waves (13 to 30 Hz), and gamma waves (30 to 50 Hz) according to the frequency band.

For the ear EEG analysis method using the re-reference method according to an embodiment of the present invention as described above, the classification accuracy for each re-reference method was experimentally verified, and the experimental process and results will be described in detail below.

1. Subject

5, this experiment was conducted from May 26, 2017 to July 7, 2017 targeting 15 adults (7 males and 8 females) who voluntarily expressed their intention to participate. All of the subjects had no mental or physical illness, and the average age was 24.5±2.7 years.

2. Experimental Protocol

Subjects were instructed to sit in comfortable chairs and remain stable. The subjects opened their eyes and rested, were presented with a task according to the signal, performed the task presented according to the signal, and repeated rest according to the signal.

And, referring to Figure 3, the task is presented to imagine subtraction mental arithmetic or alphabet song, close your eyes, open and rest your eyes according to the signal, receive a task according to the signal, perform the task presented according to the signal, and , Resting again according to the signal becomes one cycle, and subtraction mental arithmetic and alphabet song were repeated 50 times each.

3. Data collection and analysis

(1) In the EEG measurement,

The reference electrode is attached to the front center of the scalp (FCz) based on the International 10-20 system proposed by the International Federation of Societies for Electroencephalograph and Clinical Neurophysiology, and the measuring electrode is located behind both ears. Three were attached to each mastoid process.

(2) In data analysis,

The voltage at both ends of the waveform due to blinking was checked through visual inspection, and the signal was removed when it was higher than the average value of each subject (subtraction mental arithmetic data average: 126.67 ± 22.95 ㎶, alphabet song imaginary data average: 196 ± 51.34 ㎶) .

Then, subtraction mental arithmetic data and alphabet song imaginary data were extracted, respectively.

(3) In the performance evaluation,

Common average reference, contralateral, ipsilateral, bipolar-contralateral, ipsilateral of bipolar difference are re-referenced to each ear EEG from which noise components have been removed By applying the reference (Bipolar-Ipsilateral) and the overall reference (Bipolar-All) of the bipolar difference, and using a common space pattern algorithm for each data applied to each method, the frequency bands are divided into delta waves (0.5 to 4 Hz) and theta waves ( 4 to 8 Hz), alpha wave (8 to 13 Hz), beta wave (13 to 30 Hz), and gamma wave (30 to 50 Hz), respectively, and extracted features for each frequency band, measured in high and low perception states, respectively. The classification accuracy of the ear EEG was measured.

4. Experiment result

6 and 7, the common average criterion has an average classification accuracy of 72.7% and a standard deviation of 10.7, and the contralateral criterion has an average classification accuracy of 78.9% and a standard deviation of 9.2. The deviation was 10.2, the opposite standard of the positive difference was 68.6% on average, and the standard deviation was 10.2, and the ipsilateral standard of positive difference was 67.1% on average and 10.0 standard deviation, and the overall standard of positive difference was 60.9% on average and 10.1 standard deviation.

As a result, it was found that the classification accuracy of the contralateral criterion was statistically significantly higher than that of the other re-referenced methods (Friedmann test with Wilcoxon signed-rank sum test: Bonferroni corrected p <0.05).

5. Conclusion

In this experiment, unlike the conventional ear EEG analysis method, in which the experimental result is changed by the position of the reference electrode, the re-reference method was used to investigate the classification accuracy of the ear EEG. It was confirmed that the classification accuracy of more than 70% was maintained when the criteria and contralateral criteria were applied.

Therefore, in the present invention, by attaching the electrode to the skin around the ear, not the scalp, it is possible to improve the convenience and aesthetics of use when developing an EEG-based application technology.

In addition, the present invention can increase the analysis accuracy of ear EEG by reducing the influence of the position of the reference electrode through the re-reference method.

In the above description, various embodiments of the present invention have been presented and described, but the present invention is not necessarily limited thereto, and those of ordinary skill in the art to which the present invention pertains, within the scope of the technical spirit of the present invention. It will be easy to see that branch substitutions, modifications and changes are possible.

Claims (6)

Attaching a reference electrode and a plurality of measurement electrodes to the body,
Measuring ear brain waves corresponding to a high cognitive state for performing mental arithmetic by repeatedly performing for a preset time and a low cognitive state for imagining a song, and
Removing a noise component for each of the measured ear EEG,
Applying a re-reference method to each of the ear brain waves from which the noise component has been removed,
Including the step of extracting a feature of a frequency band using a common space pattern algorithm to each data to which the re-reference method is applied, and estimating classification accuracy of the ear EEG measured in the high and low perception states,
The reference electrode is attached to the front center of the scalp,
The measuring electrode is an ear EEG analysis method using a re-standard method attached to the mastoid process behind both ears.
The method of claim 1,
The step of removing the noise component,
Ear EEG analysis method using a re-reference method in which a change voltage at both ends of a waveform due to blinking is checked through a visual inspection, and a signal of the ear EEG corresponding to a case higher than the average value of the change voltage at both ends is removed.
The method of claim 1,
The above re-standard method,
Common average reference, Contralateral, Ipsilateral, Bipolar-Contralateral, Bipolar-Ipsilateral, Bipolar difference -All) ear EEG analysis method using a re-reference method, characterized in that at least one.
The method of claim 1,
The frequency band is,
Ear brain wave analysis method using a re-reference method, characterized in that at least one of delta wave, theta wave, alpha wave, beta wave, and gamma wave.
delete delete

Images