KR20150018679A - Apparatus and method for classifying brain wave - Google Patents

Abstract

Disclosed is a brainwave classifying apparatus capable of classifying brainwaves sensed by using a plurality of channels into a plurality of groups. The brainwave classifying apparatus selects any number of channels from among the channels, and determines whether classifying the selected channels into specific groups is appropriate. The brainwave classifying apparatus may also analyze brainwave responses with the sum of outputs of a plurality of groups when the brainwave responses caused by a plurality of stimuli are received. The brainwave classifying apparatus includes a sensing unit, a channel selection unit, and a determination unit.

Description

[0001] APPARATUS AND METHOD FOR CLASSIFYING BRAIN WAVE [0002]

The following embodiments are directed to techniques for classifying EEG, and more specifically, techniques for classifying EEG using correlation coefficients between EEGs.

The BCI system, which controls the mechanical devices and computers using EEG, which is an electrical characteristic that occurs when the brain state changes, is able to handle the device freely with the thought of the human being, and is receiving the spotlight as the next generation interface. In particular, it is noteworthy that patients who can not control EMG because of their difficulty in using muscles because they can control the device using only images can control the machine or the computer. Recently, BCI system using various kinds of EEG responses such as VEP (Visual Evoked Potential), P300, and mu rhythm has appeared, and various researches have been conducted such as moving a wheelchair or operating a robot in application field. In particular, the character input device through visual stimulation, which has been studied using various methods, has advanced to the point of being commercialized. In recent years, researches have been carried out in Korea, such as controlling a robot using a BCI system.

The most basic step in implementing a BCI system is the acquisition of signals. If the noise can be minimized and only the desired signal can be learned clearly, the problems that may arise in the subsequent signal processing and intention reasoning process can be reduced. However, since most of the methods for measuring EEG signals are non-invasive, electrodes are attached along the scalp to acquire brain waves, resulting in a problem of low spatial resolution and susceptibility to noise. Therefore, in order to develop a BCI system using EEG, it is necessary to sufficiently prepare for noise and to install many electrodes in order to increase spatial resolution. Based on these findings, previous researches have given one stimulus to the subject to operate the device, and only the changes of the EEG occurring at this time were explored and analyzed. Therefore, some of the measured channels, or other brain regions, The electrodes have been installed by selecting only the target channel without installing them. In this case, it has been disregarded what kind of changes in EEG occurring in other areas of the brain have been ignored so far, and it is difficult to measure the EEG effectively because only a part of the channels can be used.

The purpose of the following embodiments is to classify a plurality of sensed EEG channels into a plurality of groups.

According to an exemplary embodiment of the present invention, there is provided a channel estimation method comprising: a sensing unit sensing a brain wave using a plurality of channels; a channel selection unit selecting an arbitrary number of channels from among the plurality of channels; And a determination unit for determining whether to classify the selected channels into groups based on the correlation coefficients.

Here, the channel selection unit may select a channel combination composed of the arbitrary number of channels by a predetermined number of times, and the determination unit may determine the group among the channel combinations based on a correlation coefficient between the channels included in each channel combination You can choose.

And a memory unit indicating whether each of the plurality of channels is selected.

Also, the channel selection unit may check the memory units to see if the channels are previously selected, and may select the arbitrary number of channels according to whether the channels are previously selected.

Here, the determination unit may calculate the correlation coefficient according to the following equation (1).

[Equation 1]

는 채널간의 상관 계수를 구하는 함수이다. 또한,

는 각 채널을 나타낸다.Here, N represents the number of selected channels,

Is a function for obtaining correlation coefficients between channels. Also,

Represents each channel.

The apparatus may further include a receiver for receiving the EEG response and a reaction analyzer for analyzing the received EEG response as the sum of the responses of the plurality of classified groups.

In addition, the reaction analyzer may analyze the EEG response according to the following equation (2).

&Quot; (2) "

는 시간 t에서, g개의 그룹으로 표현된 반응 뇌파이고,

는 그룹의 개수이다.

는 시간 t에서, i번째 그룹

가 얼마나 활성화되었는지를 나타내는 계수이다.

는 시간 t에서 각 그룹

에 속하는 채널들이 가진 출력값을 의미함.here,

Is the reaction EEG expressed in g groups at time t,

Is the number of groups.

At time t, the i-th group

Is activated.

Lt; RTI ID = 0.0 > t &

And the output value of the channels belonging to the channel.

According to yet another exemplary embodiment, there is provided a method of detecting brain waves, comprising the steps of sensing an EEG using a plurality of channels, a channel step of selecting an arbitrary number of channels from the plurality of channels, a correlation step between the selected channels, And determining whether to classify the selected channels into groups based on the correlation coefficients.

Here, the selecting of the channel may include selecting a channel combination composed of the arbitrary number of channels by a predetermined number of times, and the determining may include determining a channel combination based on a correlation coefficient between channels included in each channel combination, It is possible to select the group.

In addition, the step of selecting the channel may store in the memory whether or not each of the plurality of channels is selected.

In addition, the step of selecting the channel may refer to the memory to check whether each of the channels has been previously selected, and to select the arbitrary number of channels according to whether each channel has been previously selected.

Here, the determining step may calculate the correlation coefficient according to the following equation (3).

&Quot; (3) "

는 채널간의 상관 계수를 구하는 함수이다. 또한,

는 각 채널을 나타낸다.Here, N represents the number of selected channels,

Is a function for obtaining correlation coefficients between channels. Also,

Represents each channel.

The method may further include receiving an EEG response and analyzing the received EEG response as a sum of responses of the plurality of classified groups.

In addition, the analyzing step may analyze the EEG response according to Equation (4).

&Quot; (4) "

는 시간 t에서, g개의 그룹으로 표현된 반응 뇌파이고,

는 그룹의 개수이다.

는 시간 t에서, i번째 그룹

가 얼마나 활성화되었는지를 나타내는 계수이다.

는 시간 t에서 각 그룹

에 속하는 채널들이 가진 출력값을 의미함.
here,

Is the reaction EEG expressed in g groups at time t,

Is the number of groups.

At time t, the i-th group

Is activated.

Lt; RTI ID = 0.0 > t &

And the output value of the channels belonging to the channel.

The purpose of the following embodiments is to classify a plurality of sensed EEG channels into a plurality of groups.

1 is a diagram showing classification of brain waves sensed using a plurality of channels into a plurality of groups according to an exemplary embodiment.
2 is a block diagram illustrating the structure of an EEG classifying apparatus according to an exemplary embodiment.
3 is a diagram illustrating a memory unit storing whether or not each channel is selected, according to an exemplary embodiment.
4 is a diagram illustrating analysis of an EEG response as a sum of the responses of channels included in several groups, according to an exemplary embodiment.
FIG. 5 is a flowchart illustrating steps of an EEG classifying method according to an exemplary embodiment.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings.

1 is a diagram showing classification of brain waves sensed using a plurality of channels into a plurality of groups according to an exemplary embodiment.

One of the interfaces between human and computer, research is being conducted on technologies to directly control computers using brain waves. The present stage is an initial stage of technology development, and research is focused on sensing brain waves from the user's 110 brain and analyzing the sensed brain waves.

According to one aspect, a plurality of electrodes are installed in various parts of the brain of the user 110. In FIG. 1 (a), a plurality of electrodes 111, 112, and 113 are provided on the scalp of the user 110 to sense the user's brain waves.

In order to sense the brain waves of the user 110, a stimulus is given to the user 110 to induce a specific reaction. Here, visual stimulation, auditory stimulation, and the like can be given to the user 110. The user 110 generates a specific pattern of brain waves in response to stimulation.

The brain wave classifying apparatus senses brain waves of the user 110 using a plurality of electrodes 111, 112, and 113. Electroencephalograms sensed using the electrodes 111, 112, and 113 correspond to respective channels. That is, 'channel' in the present specification refers to an EEG sensed by using different electrodes.

By analyzing all the brain waves sensed using each channel, it is possible to grasp the brain activity of the user 110 with respect to a specific stimulus. However, analyzing signals of all channels is complicated due to calculation amount, complexity, and the like.

Accordingly, a channel selection technique for dividing a plurality of channels into groups 121, 122, 123, 124, 125, 126 and reducing the number of signals to be analyzed has been proposed. In FIG. 1B, an embodiment in which a plurality of channels are classified into six groups 121, 122, 123, 124, 125 and 126 is shown. 1B, each of the groups 121, 122, 123, 124, 125, and 126 is divided according to the area of the scalp, but according to another embodiment, each channel or electrode may be differentiated .

2 is a block diagram illustrating the structure of an EEG classifying apparatus according to an exemplary embodiment.

The EEPROM 200 according to the exemplary embodiment includes a sensing unit 210, a channel selection unit 220, a memory unit 230, a determination unit 240, a reception unit 250, and a reaction analysis unit 260 .

The sensing unit 210 senses brain waves using a plurality of channels. Here, the channel means an EEG sensed using different electrodes. According to one aspect, the sensing unit 210 can sense EEG using about 60 channels.

The channel selection unit 220 selects an arbitrary number of channels from a plurality of sensed channels. According to one aspect, the channel selection unit 220 can select a channel a predetermined number of times. For example, if the predetermined number of times is 1000, the channel selection unit 220 may repeat the selection of an arbitrary number of channels 1000 times. In this case, a plurality of channels selected at the first time may be referred to as a channel combination. Accordingly, when the channel selector 220 selects an arbitrary number of channels 1000 times, the channel selector 220 can generate 1000 channel combinations. Each channel combination includes an arbitrary number of channels selected by the channel selection unit 220. Here, the channel selection unit 220 can arbitrarily determine the number of channels to be selected each time.

The determination unit 230 calculates a correlation coefficient between the selected channels. The correlation coefficient is a coefficient indicating whether the signals between the channels are similar. If the signals between the channels are similar, the correlation coefficient is a large value. If the signals between the channels are not similar, the correlation coefficient is a small value.

The correlation coefficient is defined between each two channels. Therefore, the number of correlation coefficients is also determined according to the number of selected channels. According to one aspect, the determination unit 240 may average correlation coefficients according to Equation (1).

[Equation 1]

는 채널간의 상관 계수를 구하는 함수이다. 또한,

는 각 채널을 나타낸다.Here, N represents the number of selected channels,

Is a function for obtaining correlation coefficients between channels. Also,

Represents each channel.

The determination unit 240 may determine whether to classify an arbitrary number of channels selected by the channel selection unit 220 into a group based on the averaged correlation coefficient. In addition, the determination unit 240 may select a combination to be grouped among the predetermined number of channel combinations selected by the channel selection unit 220. [

According to one aspect, the channel selection unit 220 can select a new channel with reference to a result of selecting a channel previously. For this, the channel selection unit 220 selects an arbitrary number of channels from among a plurality of channels, and the memory unit 230 stores whether or not each channel is selected.

Hereinafter, the operations of the memory unit 230 and the channel selection unit 220 will be described with reference to FIG.

3 is a diagram illustrating a memory unit storing whether or not each channel is selected, according to an exemplary embodiment.

The horizontal axis in FIG. 3 indicates whether each channel is selected, and the vertical axis indicates the number of times the channel is selected. In addition, the number inside each square indicates whether or not the corresponding channel is selected. If the number is '1', it means that the corresponding channel is selected. If the number is '0', it means that the corresponding channel is not selected.

Referring to FIG. 3, the channel selection unit 220 selects four channels of the second channel, the fourth channel, the sixth channel, and the eighth channel in the first selection. Accordingly, the first combination includes the second channel, the fourth channel, the sixth channel, and the eighth channel. For convenience of explanation, it is assumed in FIG. 3 that the first to fourth channels are classified into one group and the sixth to eighth channels are classified into different groups. The determination unit 240 calculates correlation coefficients between the selected four channels, and averages the calculated correlation coefficients as shown in Equation (1).

The channel selection unit 220 selects the second channel, the fourth channel, the fifth channel, and the seventh channel in the second selection. Accordingly, the second combination includes the second channel, the fourth channel, the fifth channel, and the seventh channel. The determination unit 240 calculates correlation coefficients between the selected four channels, and averages the calculated correlation coefficients as shown in Equation (1).

The channel selection unit 220 compares the average value of the correlation coefficients for the first combination with the average value of the correlation coefficients for the second combination. In the second combination, the fifth channel and the seventh channel were selected instead of the sixth channel and the eighth channel. The channel selection unit 220 selects a next channel in consideration of a point where the channel is changed and an average value of the correlation coefficient is changed.

In the third selection, the channel selection unit 220 can select the second channel, the third channel, the fourth channel, and the seventh channel. The third combination includes a second channel, a third channel, a fourth channel, and a seventh channel. Since three channels are selected from the first to fourth channels, which are preferably classified into the same group, the average value of the correlation coefficients for the third combination may be increased rather than the average value of the correlation coefficients for the first combination or the second combination .

In the third combination, the third channel was newly selected, resulting in an increase in the average value of the correlation coefficient for the third combination. Therefore, the channel selection unit 220 can maintain the third channel in the fourth selection. Alternatively, the channel selection unit 220 may set the possibility that the third channel is maintained to be higher than the possibility that another channel is maintained.

In the fourth selection, the channel selection unit 220 can select the first to fourth channels. The fourth combination includes the first channel to the fourth channel. Since the first to fourth channels, which are preferably classified into the same group, are all selected, the average value of the channel coefficients for the fourth combination is the maximum value.

Even if the channel selection unit 220 repeats the selection of the channel, the average value of the channel coefficients for the fourth combination is the maximum value.

Accordingly, when the channel selection unit 220 repeats channel selection a predetermined number of times, the determination unit 240 can classify the channels based on the average value of the correlation coefficients for each combination.

In the embodiment shown in FIG. 3, if the average value of the correlation coefficients for the fourth combination is the largest, the determination unit 240 may classify the channels included in the fourth combination into groups. Accordingly, the first to fourth channels are classified into specific groups.

According to one aspect, the memory unit may store only information on a limited number of combinations. The channel selection unit 220 selects a new combination of channels, and the determination unit 240 can calculate an average value of correlation coefficients for the newly selected combination. In addition, the determination unit 240 may compare the average value of the correlation coefficients for the new combination channels with the average value of the correlation coefficients for the previous combination channels. If the average value of the correlation coefficients for the new combination channels is larger than the average value of the correlation coefficients for the previous combination channels, the memory unit 230 deletes the combination having the smallest average value of the correlation coefficients among the previous combinations, A new combination can be stored in the memory unit.

After the channel selection unit 220 selects all combinations of the predetermined number of times, the determination unit 240 may classify the combinations having the largest average value of the correlation coefficients among the predetermined number of combinations into groups.

The receiving unit 250 receives a user's brain wave response to a specific stimulus. Here, a specific stimulus may be an auditory stimulus, a visual stimulus, a tactile stimulus, or the like, or two or more stimuli may be superimposed.

The reaction analyzer 260 analyzes the received EEG response as a sum of the responses of the respective groups.

4 is a diagram illustrating analysis of an EEG response as a sum of the responses of channels included in several groups, according to an exemplary embodiment.

As described above, each channel can be classified into a plurality of groups. Each group may include a channel sensed using an electrode located in a specific region of the brain. For example, the first group includes channels sensed using electrodes located in the first region 441 of the brain, and the second group includes channels sensed by using the electrodes located in the second region 442 of the brain, . ≪ / RTI > An embodiment is shown in which channels are classified into groups corresponding to the six regions 441, 442, 443, 444, 445, and 446. [

The human brain produces an EEG response in response to stimuli given internally or externally. In particular, the EEG response to specific stimuli has a tendency to appear in specific areas of the brain. According to one aspect, when the user 410 is exposed to the visual stimulus 420, the first region 441 of the brain actively reacts to generate a strong output brain wave. In addition, when the user 410 is exposed to the auditory stimulus 430, the second region 442 of the brain actively reacts to generate strong output brain waves.

4, the user 410 may be exposed to an overlapping stimulus in which the visual stimulus 420 and the auditory stimulus 430 are superimposed. As shown in FIG. 4, when multiple stimuli 420 and 430 are superimposed on the user, the EEG response to the superimposed stimuli assumes that the individual responses generated by the individual stimuli are superimposed and do not interfere with each other can do. That is, when the visual stimulus 420 and the auditory stimulus 430 are overlapped and exposed to the overlapping stimulus as shown in FIG. 4, the first region 441 and the second region 442 of the brain can actively react.

In this case, the EEG generated by the nested stimuli follows the principle of superposition, and the total response of the EEG can be thought of as a linear combination of responses to the individual stimuli.

Therefore, the number of stimuli applied to the user and the number of responses to the stimuli are the same, and each reaction appearing as a separate group is regarded as a separate group based on a specific point in time. .

This is expressed by Equation (2) below.

&Quot; (2) "

는 시간 t에서, g개의 그룹으로 표현된 반응 뇌파이고,

는 그룹의 개수이다. 집중도 계수

는 시간 t에서, i번째 그룹

가 얼마나 활성화되었는지를 나타내는 계수이다.

는 시간 t에서 각 그룹

에 속하는 채널들이 가진 출력값을 의미함.
here,

Is the reaction EEG expressed in g groups at time t,

Is the number of groups. Concentration coefficient

At time t, the i-th group

Is activated.

Lt; RTI ID = 0.0 > t &

And the output value of the channels belonging to the channel.

는 하기 수학식 3에 따라서 정의될 수 있다.
Here, the concentration coefficient

Can be defined according to the following equation (3).

&Quot; (3) "

Referring to Equation (3), the concentration coefficient means a ratio of power between the entire channel and the corresponding group. A group with a high concentration coefficient can be regarded as an important group in which the user is most concentrated at the time t. Also, groups with low concentration coefficients can be regarded as relatively inactive groups.

는 시간에 따라 변하는 값이지만, 특정 시간 동안 사용자가 자극에 집중할 수 있다면, 그 시간 동안은 상수로 볼 수 있다.Concentration coefficient

Is a time varying value, but if the user can focus on the stimulus during a certain time, it can be seen as a constant during that time.

Analysis of the response EEG according to Equation (2) reveals that, when exposed to two or three stimuli, the channels responding to the stimuli are activated at the same time as when one stimulus is given.

FIG. 5 is a flowchart illustrating steps of an EEG classifying method according to an exemplary embodiment.

In step 510, the brain wave classifying apparatus senses brain waves using a plurality of channels.

In step 520, the EEG apparatus selects an arbitrary number of channels from among the plurality of sensed channels. Selections Any number of channels constitute channel combinations.

In step 530, the brain wave classifying apparatus stores information about the selected channel. Here, the information on the selected channel may indicate whether or not each channel is selected. According to one aspect, the brain wave classifying apparatus may include a memory. In this case, the brain wave classifying apparatus can store information on the selected channel in the memory.

In step 540, the EEG apparatus calculates a correlation coefficient for the selected channels. According to one aspect of the present invention, the brain wave classifying apparatus can average the calculated correlation coefficients by the above Equation (1) to calculate an average value of correlation coefficients.

In step 550, the EEG apparatus may determine whether or not any number of channels have been selected a predetermined number of times. If the number of times that an arbitrary number of channels are selected is smaller than the predetermined number, the EEG classifying apparatus reselects an arbitrary number of channels in step 520.

According to one aspect, the brain wave classifier can refer to the memory to check whether each channel has been previously selected, and to reselect an arbitrary number of channels depending on whether each channel was previously selected.

According to one aspect, the memory may only store information for a limited number of combinations. In this case, the brain wave classifying apparatus can calculate an average value of correlation coefficients for the newly selected combination. In addition, the EEG device can compare the average value of the correlation coefficients for the new combination channels with the average value of the correlation coefficients for the previous combination channels. If the average value of the correlation coefficients for the channels of the new combination is larger than the average value of the correlation coefficients for the channels of the previous combination, the EEG classifier deletes the combination having the smallest average value of the correlation coefficients from the previous combinations, Can be stored in the memory.

If at step 550 the number of times that an arbitrary number of channels are selected is equal to or greater than a predetermined number, the EEG classifier classifies at least one of the selected channels into groups, at step 560. According to one aspect, the brain wave classifier can select a group among channel combinations based on the correlation coefficient between the channels included in each channel combination. For example, the brain wave classifier can classify the channels included in the combination having the largest average value of the correlation coefficients among the selected combinations as a group.

At step 560, the brain wave classifier receives the user's brain wave response to a particular stimulus. Here, a specific stimulus may be an auditory stimulus, a visual stimulus, a tactile stimulus, or the like, or two or more stimuli may be superimposed.

In step 570, the brain wave classifier analyzes the received EEG response as the sum of the responses of the plurality of groups classified in step 560. According to one aspect of the present invention, when an EEG is applied to a user by overlapping a plurality of stimuli, an EEG response to the stimuli superimposed on each other is assumed to be a superposition of individual responses caused by individual stimuli, have.

In this case, the EEG generated by the nested stimuli follows the principle of superposition, and the total response of the EEG can be thought of as a linear combination of responses to the individual stimuli.

Therefore, the number of stimuli applied to the user and the number of responses to the stimuli are the same, and each reaction appearing as a separate group is regarded as a separate group based on a specific point in time. .

According to one aspect, the brain wave classifier can analyze the EEG response according to Equation (2) as a sum of individual groups.

The method according to an embodiment may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

210: sensing unit
220: Channel selection unit
230: memory unit
240:
250: Receiver
260: reaction analysis section

Claims (15)

A sensing unit sensing a brain wave using a plurality of channels;
A channel selector for selecting an arbitrary number of channels among the plurality of channels; And
A determination unit for determining whether to classify the selected channels into groups based on the calculated correlation coefficients,
And an electroencephalogram classifier.
The method according to claim 1,
Wherein the channel selection unit selects a channel combination composed of the arbitrary number of channels by a predetermined number of times,
Wherein the determination unit selects the group among the channel combinations based on a correlation coefficient between channels included in each channel combination.
3. The method of claim 2,
A memory unit for storing whether or not each of the plurality of channels is selected;
Further comprising:
The method of claim 3,
Wherein the channel selection unit refers to the memory unit to confirm whether the channels are previously selected and to select the arbitrary number of channels according to whether or not each channel is previously selected.
The method according to claim 1,
Wherein the determination unit calculates the correlation coefficient according to the following equation (1).

[Equation 1]

Here, N represents the number of selected channels,

Is a function for obtaining correlation coefficients between channels. Also,

Represents each channel.
The method according to claim 1,
A receiving unit for receiving an EEG response; And
And analyzing the received EEG response as the sum of the responses of the plurality of classified groups,
Further comprising:
The method according to claim 6,
Wherein the reaction analyzer analyzes the EEG response according to the following equation (2).

&Quot; (2) "

here,

Is the reaction EEG expressed in g groups at time t,

Is the number of groups.

At time t, the i-th group

Is activated.

Lt; RTI ID = 0.0 > t &

And the output value of the channels belonging to the channel.
Sensing an EEG using a plurality of channels;
A channel step of selecting an arbitrary number of channels among the plurality of channels; And
Calculating correlation coefficients between the selected channels, and determining whether to classify the selected channels into groups based on the calculated correlation coefficients
And a brain wave.
9. The method of claim 8,
Wherein the selecting of the channel comprises: selecting a channel combination composed of the arbitrary number of channels by a predetermined number of times;
Wherein the determining comprises selecting the group from among the channel combinations based on a correlation coefficient between the channels included in each channel combination.
10. The method of claim 9,
Wherein the step of selecting the channel stores in the memory whether or not each of the plurality of channels is selected.
11. The method of claim 10,
Wherein the step of selecting a channel further comprises the steps of: determining whether each of the channels is previously selected by referring to the memory; and selecting the arbitrary number of channels according to whether or not each of the channels is previously selected.
9. The method of claim 8,
Wherein the determining step calculates the correlation coefficient according to the following equation (3).

&Quot; (3) "

Here, N represents the number of selected channels,

Is a function for obtaining correlation coefficients between channels. Also,

Represents each channel.
9. The method of claim 8,
Receiving an EEG response; And
Analyzing the received EEG response as the sum of the responses of the plurality of classified groups
Further comprising the steps of:
14. The method of claim 13,
Wherein analyzing the response includes analyzing the EEG response according to Equation (4).

&Quot; (4) "

here,

Is the reaction EEG expressed in g groups at time t,

Is the number of groups.

At time t, the i-th group

Is activated.

Lt; RTI ID = 0.0 > t &

And the output value of the channels belonging to the channel.
A computer-readable recording medium having recorded thereon a program for executing the method according to any one of claims 8 to 14.

Images