JP2006006665A - Brain function analyzing system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To analyze a brain function appropriately even when a pasteless electroencephalograph, or the like, having an unstable contact state is used by correcting a brain wave signal according to a contact state between an electrode and the scalp. <P>SOLUTION: A brain function analyzing system 1 is provided which comprises at least one brain signal detection section arranged in contact with the scalp for detecting the brain wave signal S1, a parameter calculation section 7 for calculating a parameter S2 that shows the contact state between each brain wave signal detection section and the scalp on the basis of the detected brain wave signal S1, a contact state determination section 8 for comparing the parameter S2 calculated by the parameter calculation section 7 and a predetermined threshold value S5 to determine whether or not the contact state is good, a brain wave signal correction section 9 for correcting the brain wave signal S1 when the contact state is determined to be bad, and a brain function determination section 10 for determining the brain function state on the basis of a brain wave signal S3 corrected by the brain wave correction section 9. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a brain function analysis system.

  Conventionally, analysis of brain function states has been performed for the purpose of early detection of Alzheimer's disease, and various methods for analyzing brain function states have been proposed (see, for example, Non-Patent Document 1 and Non-Patent Document 2). . In this analysis method, an electroencephalogram signal is detected, and the determination is made based on the smoothness of the spatial transmission and the standard deviation. When brain neuron activity is normal, the electroencephalogram signal is transmitted smoothly and there is little variation. It is lost and its standard deviation tends to increase.

  In order to perform such an analysis with high accuracy, it is indispensable to detect an electroencephalogram signal with high accuracy. Electroencephalogram signals are detected by bringing multiple electrodes into contact with the scalp, but in order to achieve a reliable contact between the electrode and the scalp due to the presence of hair and the unevenness of the shape of the head, the electrodes must be applied with paste. It is common to adhere to the scalp. However, when using a paste, it takes a long time to start the inspection, and the burden on the person to be inspected is great. Also, when removing the electrode after the inspection, the hair can be removed to remove the paste. There is an inconvenience that work is required and the burden on the inspected person is large.

  On the other hand, pasteless electroencephalogram signal detection methods that do not use paste have also been proposed (see, for example, Patent Document 1 and Patent Document 2). These detection methods are methods in which a fiber piece in a state where water is absorbed in a conductive solution or a water-containing gel impregnated with conductivity is impregnated into a foaming material, and each electrode is brought into contact with the scalp, and no paste is used. Therefore, there is an advantage that the time required for mounting and dismounting can be shortened.

Brain Function Research Institute, Inc., “DIMENSION, a new EEG analysis method for detecting functional deterioration of brain cortical neurons in early Alzheimer's disease”, [online], [March 8, 2004 search], Internet <URL: http : //www.bfl.co.jp/abst-32.html> Brain Function Laboratory Co., Ltd., “Method of Estimating Deterioration of Brain Cortical Neuron Function”, [online], [Search March 8, 2004], Internet <URL: http://www.bfl.co.jp/ abst-31.html> JP-A-10-165386 (second page, etc.) Japanese Utility Model Publication No. 2-63811 (FIG. 1 etc.)

  However, the pasteless method that does not use the paste cannot be denied that the contact state between the electrode and the scalp is unstable, although it is simpler than the case where the paste is used. Then, the inspection is performed in a state where the contact is not reliably made, and there arises a problem that an appropriate inspection result cannot be obtained or the inspection needs to be performed again. A general electroencephalograph is equipped with a means for detecting contact impedance and confirming it on a monitor, but it is difficult to say that the contact impedance necessarily represents a contact state, and an appropriate test result cannot be obtained. There is an inconvenience.

  The present invention has been made in view of the above-described circumstances, and even when an electroencephalograph with an unstable contact state such as a pasteless method is employed, an electroencephalogram signal is generated according to the contact state between the electrode and the scalp. An object of the present invention is to provide a brain function analysis system capable of correcting and appropriately analyzing brain function.

In order to achieve the above object, the present invention provides the following means.
The present invention relates to at least one electroencephalogram signal detection unit arranged in contact with the scalp to detect an electroencephalogram signal, and a parameter for calculating a parameter indicating a contact state between each electroencephalogram signal detection unit and the scalp based on the detected electroencephalogram signal A calculation unit, a contact state determination unit that compares the parameter calculated by the parameter calculation unit with a predetermined threshold value to determine whether the contact state is good, and an electroencephalogram signal when the contact state is determined to be bad A brain function analysis system is provided that includes a brain wave signal correction unit that corrects a brain function and a brain function determination unit that determines a brain function state based on the brain wave signal corrected by the brain wave signal correction unit.

  According to the present invention, from the electroencephalogram signal detected by the electroencephalogram signal detection unit, the parameter calculation unit calculates the parameter indicating the contact state between the electroencephalogram signal detection unit and the scalp, and the contact state determination unit sets the parameter to a predetermined value. The quality of the contact state is determined by comparing with the threshold value. When it is determined that the contact state is bad, the electroencephalogram signal is corrected by the operation of the electroencephalogram signal correction unit, and the brain function state is determined based on the corrected electroencephalogram signal. Therefore, not only when the contact state is good, but also when the contact state is bad, the brain function state can be properly determined.

In the said invention, it is preferable that the said parameter is calculated based on the noise contained in an electroencephalogram signal.
When the contact state between the electroencephalogram signal detection unit and the scalp is poor, low-frequency noise is often included in the electroencephalogram signal. Therefore, by calculating the noise included in the electroencephalogram signal, it is possible to accurately determine the contact state and correct the electroencephalogram signal appropriately.

  In the above invention, the parameter may be a signal-to-noise ratio included in an electroencephalogram signal. As described above, since the noise in the electroencephalogram signal increases when the contact state is poor, the signal-to-noise ratio is a parameter that accurately represents the contact state. Therefore, it is possible to appropriately determine the brain function state by calculating the signal-to-noise ratio and correcting the electroencephalogram signal based on the signal-to-noise ratio.

In the above invention, the signal-to-noise ratio may be a ratio of an α wave component signal to a signal of other components.
The brain wave signal used to determine brain function is in the frequency band of the α wave, and the noise component that increases due to the contact state is a lower frequency band, so the signal of the α wave component and the signal of the other components It is possible to obtain a signal-to-noise ratio that accurately represents the contact state.
In this case, the α wave component signal is preferably a signal in a wavelength band of about 8 to 13 Hz.

  In the above invention, the signal-to-noise ratio may be an average value of the signal-to-noise ratios included in the electroencephalogram signals detected by the plurality of electroencephalogram signal detection units. In general, the detection of the electroencephalogram signal is performed in a plurality of electroencephalogram signal detection units, and the contact state of each detection unit varies. Therefore, by using the average of those values, the overall contact state is detected. It is possible to determine pass / fail.

In the above invention, the parameter may be a variation in a signal-to-noise ratio included in an electroencephalogram signal. When the variation is large, the contact state is often bad, and the quality of the overall contact state can be accurately determined by seeing the variation.
In this case, the variation of the signal-to-noise ratio may be a standard deviation of the signal-to-noise ratio, and is a variation coefficient obtained by dividing the standard deviation of the signal-to-noise ratio by the average value of the signal-to-noise ratio. It is good as well.
Further, both the average value of the signal-to-noise ratio included in the electroencephalogram signal and the variation in the signal-to-noise ratio may be used as parameters.

In the above invention, an amplitude value of an electroencephalogram signal may be used as the parameter. When the contact state between the electroencephalogram signal detection unit and the scalp becomes poor, the amplitude often increases momentarily. Therefore, the deterioration of the contact state can be determined by calculating the amplitude value.
In this case, it is preferable that the threshold value is set to a value larger than the maximum amplitude value of the α wave.

In the above invention, the electroencephalogram signal correction unit removes the electroencephalogram signals detected by all the electroencephalogram signal detection units in the time domain determined to be in poor contact based on the parameter, and is removed. The brain wave signals before and after the time domain may be combined.
By removing the EEG signal in the time domain that is determined to be in poor contact and combining it before and after it, the noise contained in the EEG signal can be removed, making it possible to properly determine the brain function status It becomes.

Further, in the above invention, the electroencephalogram signal correction unit identifies an electroencephalogram signal detection unit determined to have a poor contact state based on the parameter, and the electroencephalogram signal detected by the identified electroencephalogram signal detection unit It is also possible to remove over the time region determined to be in a poor contact state and interpolate based on the electroencephalogram signals before and after the time region or the electroencephalogram signals detected by other electroencephalogram signal detectors.
Eliminate noise by removing EEG signals in the time domain determined to be in poor contact, and interpolate that area based on EEG signals before and after the time domain or EEG signals from other EEG signal detectors Thus, the brain function state can be more appropriately determined by supplementing the brain wave signal in a pseudo manner.

In the above invention, it is preferable that the electroencephalogram signal correction unit attenuates the electroencephalogram signal over a time domain determined to be in a poor contact state based on the parameter by multiplying the electroencephalogram signal by a predetermined coefficient.
When the amplitude value of the electroencephalogram signal increases locally due to the deterioration of the contact state, the noise is removed from the electroencephalogram signal by simply applying a predetermined coefficient to the electroencephalogram signal in the time domain and attenuating it. It is possible to determine the brain function state.

In the above invention, the electroencephalogram signal correction unit identifies an electroencephalogram signal detection unit determined to have a poor contact state based on the parameter, and the electroencephalogram signal detected by the identified electroencephalogram signal detection unit It is good also as determining a brain functional state based on the electroencephalogram signal detected by the other electroencephalogram signal detection part excepting.
By excluding the electroencephalogram signal detection unit identified as having a poor contact state, it is possible to remove noise contained in the electroencephalogram signal and make a determination as to an appropriate brain function state.

Moreover, in the said invention, it is good also as determining a brain functional state using the pseudo electroencephalogram signal estimated based on the electroencephalogram signal detected by the other electroencephalogram signal detection part instead of the excluded electroencephalogram signal.
The excluded brain wave signal can be supplemented with the pseudo brain wave signal, so that a more appropriate brain function state can be determined.

  Moreover, in the said invention, a brain function determination part is good also as determining a brain function state based on the smoothness of the spatial transmission of an electroencephalogram, and temporal fluctuation. By observing the relationship between the smoothness of the spatial transmission of brain waves and the temporal fluctuation, it is possible to easily determine whether the disease is Alzheimer's disease or normal.

  According to the present invention, since the electroencephalogram signal is corrected by the parameter that accurately represents the contact state between the electroencephalogram signal detection unit and the scalp, it is possible to determine an appropriate brain function state. As a result, an appropriate analysis result can be obtained even if a pasteless electroencephalogram signal detection unit that does not use paste is used. That is, it is possible to reduce the time required for wearing and removing, eliminating the need for washing hair, and reducing the burden on the subject.

A brain function analysis system according to an embodiment of the present invention will be described below with reference to FIGS.
As shown in FIG. 1, the brain function analysis system 1 according to the present embodiment includes an electroencephalogram detection apparatus 2, a brain function analysis apparatus 3, and a monitor 4 that displays an analysis result.

The electroencephalogram detection apparatus 2 is an electroencephalograph, and includes a plurality of electrodes 5 arranged in contact with the scalp A (FIG. 1 shows five electrodes 5 for ease of illustration, but usually 21 electrodes. In this embodiment, it is assumed that there are 21 electrodes 5.) And an electroencephalograph main body 6 that outputs an electric signal detected by the electrodes 5 as an electroencephalogram signal.
Further, as shown in FIG. 2, the electroencephalogram analysis apparatus 3 receives the electroencephalogram signal S1 output from the electroencephalograph main body 6 and calculates a predetermined parameter S2, and the parameter calculation unit 7 Based on the calculated parameter S2, the electroencephalogram which corrects the electroencephalogram signal S1 output from the electroencephalograph main body 6 based on the contact state determination unit 8 which determines the quality of the contact state between the electrode 5 and the scalp A and the determination result. A signal correction unit 9 and a brain function determination unit 10 that determines a brain function state based on the corrected electroencephalogram signal S3 and outputs a determination signal S4 are provided.

  The parameter calculation unit 7 calculates a signal-to-noise ratio (hereinafter also referred to as SN ratio S2) as the parameter S2. The SN ratio S2 is calculated for each of the plurality of electrodes 5. The electroencephalogram signal S1 detected by each electrode 5 includes a component that can be used for determination of brain function in the frequency band of the α wave, and includes many noise components that vary depending on the contact state in other frequency bands. Therefore, for example, as shown in FIG. 3, the contact state is determined by taking a ratio of the brain wave signal S in the frequency band (8 Hz to 13 Hz) of the α wave and the brain wave signal N in the other frequency band. A good SN ratio S2 can be obtained. Therefore, the parameter calculation unit 7 obtains the frequency distribution of the electroencephalogram signal S1 output from the electroencephalograph main body 6, and then the integrated value of the electroencephalogram signal S in the frequency band of the α wave and the electroencephalogram signal N in the other frequency bands. The SN ratio S2 is obtained by calculating a ratio with the integrated value of.

  In place of the electroencephalogram signal S shown in FIG. 3, as shown in FIG. 4, for example, a Gaussian distribution having a half value at 8 Hz and 13 Hz is used as the electroencephalogram signal S, and the ratio with the other electroencephalogram signals N It is good also as SN ratio S2 by taking. Further, without being limited to 8 Hz and 13 Hz, for example, a Gaussian distribution having a half value at ± 3 Hz around the peak of the electroencephalogram signal may be used as the electroencephalogram signal S, and the others may be used as the electroencephalogram signal N.

  The contact state determination unit 8 determines the quality of the contact state based on the SN ratio S2 calculated by the parameter calculation unit 7. Specifically, with respect to the electroencephalogram signal S1 obtained from all the electrodes 5, the time change of the SN ratio S2 is calculated and the average value thereof is calculated. An example is shown in FIG.

The contact state determination unit 8 further compares the average value of the SN ratio S2 at each time with a threshold value S5 that is determined in advance and stored in the memory 11. As a result, when the period t ₁ t ₂ in which the SN ratio S 2 is low, that is, it is determined that the noise is large with respect to the signal, the signal S 6 specifying the period t ₁ t ₂ is output. Has become

The electroencephalogram signal correction unit 9 corrects the electroencephalogram signal S1 based on the electroencephalogram signal S1 received from the electroencephalograph main body 6 and the information S6 received from the contact state determination unit 8 during the period t ₁ t _{2 with} a low SN ratio. It is supposed to be. Specifically, as shown in FIG. 6, the contact state determination unit 8 deletes the electroencephalogram signal during the period t ₁ t ₂ in which it was determined that the S / N ratio is low (FIG. 6 (a)). Signals before and after the period t ₁ t ₂ are combined so as to eliminate the portion (FIG. 6B). The corrected electroencephalogram signal S3 is output to the brain function determination unit 10.

  The brain function determination unit 10 determines the brain function state based on the corrected electroencephalogram signal S3 input from the electroencephalogram signal correction unit 9. Specifically, a Dα value indicating the smoothness of spatial transmission of an electroencephalogram and a standard deviation (that is, temporal fluctuation) Dσ value are calculated. The relationship between the Dα value and the Dσ value is as shown in FIG. By plotting the calculated Dα value and Dσ value on the graph shown in FIG. 7, whether the subject has Alzheimer's disease, is normal, is a boundary between them, or how much is progressing Can be determined. The determination signal S4 is output to the monitor 4 and displayed.

  According to the brain function analysis system 1 according to the present embodiment configured as described above, the SN ratio that favorably indicates the contact state between the electrode 5 serving as the electroencephalogram signal detection unit and the scalp A is calculated as the parameter S2, and Since the electroencephalogram signal S1 is corrected based on this, it is possible to exclude the electroencephalogram signal S1 when the contact state is bad from the information that is the basis of the judgment and perform an appropriate judgment. As a result, even if a pasteless method in which the contact state between the electrode 5 and the scalp A is unstable is employed, the brain function can be properly analyzed.

In the present embodiment, the contact state determination unit 8 uses the average value of the SN ratios of all the channels as the parameter S2, but instead of this, a value indicating the variation of the SN ratios of all the channels, for example, A standard deviation or a variation coefficient may be used as the parameter S2. If the contact state between the electrode 5 and the scalp A in some of the plurality of channels deteriorates, the variation in the SN ratio increases. Therefore, as shown in FIG. The change may be calculated and compared with a predetermined threshold value, and a period t ₁ t ₂ in which the variation is larger than the threshold value may be output as a period to be excluded.

Further, in the above embodiment, the electroencephalogram signal correction unit 9 excludes the electroencephalogram signals S1 of all channels in the period t ₁ t ₂ in which the average value of the SN ratio is low. For example, as shown by the chain line X in FIG. 9, the channel where the SN ratio deteriorates in the period t ₁ t ₂ (here, channel 2 and The channel 21) may be specified, and the brain wave signal of the period t ₁ t ₂ may be excluded from only the specified channel and may be interpolated with a pseudo brain wave signal.

As an interpolation method, based on the electroencephalogram signal S1 in a period other than the excluded period t ₁ t ₂ of the specified channels (for example, CH2 and CH21), the electroencephalogram signal S1 in the excluded period t ₁ t ₂ is changed. A method of estimation, a method of estimation based on the electroencephalogram signal S1 within the period t ₁ t ₂ in channels other than the specified channels (for example, CH2 and CH21), and the like can be considered. For example, it compares the brain wave signal S1 with each other for a predetermined time before and after the excluded period t ₁ t _2, may be generating a pseudo electroencephalogram signal according to the degree of the matching by calculating the degree of matching.

In addition, the electroencephalogram signals S1 before and after the period t ₁ t ₂ during which the average value of the S / N ratio is below the threshold value are simply combined. In this case, however, as shown in FIG. Since it may become a continuous waveform and the accuracy of the subsequent analysis may be lowered, it may be corrected using a window function. The continuous electroencephalogram signal S1 waveform as shown in FIG. 10B can be obtained by the correction. As the window function, any known window function such as a rectangular window, a Hanning window, a Hamming window, or a Blackman window can be used.

Further, without performing interpolation of the excluded period t ₁ t ₂ , the brain function parameters Dαi and Dσi of the slices are calculated using the electroencephalogram signal S1 of the other period, and weighted according to the time according to the following equation: , Dα, Dσ may be calculated.

ここで、Ｔ_１，Ｔ_２，Ｔ_３はそれぞれ、ＳＮ比が良好な期間の時間を示し、Ｄαｉ、Ｄσｉは、それぞれそれらの期間における脳波の伝達の滑らかさと標準偏差とを示している。

Here, T ₁ , T ₂ , and T ₃ each indicate the time of a period with a good S / N ratio, and Dαi and Dσi indicate the smoothness and standard deviation of electroencephalogram transmission in those periods, respectively.

Further, in the above embodiment, the average time variation of the SN ratio S2 of all channels is observed, and the electroencephalogram signal S1 of the period t ₁ t ₂ that falls below the predetermined threshold value S5 is excluded from the basis of judgment. It was decided. Instead, as shown in FIG. 11, the S / N ratios S2 of all the channels at each instant during the detection of the electroencephalogram signal S1 are individually observed, and the S / N ratios S2 below the threshold value at each instant The electroencephalogram signal S1 may be excluded and interpolated.

For example, at the moment of FIG. 11, since the SN ratio S2 between the channel 7 and the channel 18 is below the threshold value S5, the contact state between the electrode 5 and the scalp A at the position of the channels 7 and 18 shown in FIG. It turns out that it is bad. Therefore, the electroencephalogram signals S1 of these channels 7 and 18 at this moment are excluded, and instead of this, for example, interpolation may be performed using the average value of the electroencephalogram signals S1 of the surrounding channels. For example, in the case of channel 7, the average value of channels 5, 9, 13, 15, 19, 20, and 21, and in the case of channel 18, channels 1, 2, 3, 4, 5, 6, 17, 19 The average value may be adopted.
Further, instead of the above modification, the entire electroencephalogram signal S1 having a period during which the SN ratio S2 is lower than the threshold value is excluded from the basis of the brain function determination, and is interpolated by the average value of the electroencephalogram signals S1 of the surrounding channels. It is good as well.

Moreover, in the said embodiment, SN ratio was employ | adopted as parameter S2 which shows the contact state of the electrode 5 and the scalp A favorably. Instead, the amplitude value of the electroencephalogram signal S1 in each channel may be adopted as the parameter S2. When the contact state between the electrode 5 and the scalp A becomes poor, a low-frequency signal appears, and the amplitude of the electroencephalogram signal S1 locally increases as shown in FIG. 13 (in this example, the channel 7 ). Therefore, when such a signal appears, the brain function may be determined by excluding the signals of all channels in the period t ₁ t ₂ during which the signal continues.

In the example of FIG. 13, the entire electroencephalogram signal S1 of the channel 7 whose amplitude is locally increased may be excluded from the basis of the determination of the brain function.
Further, the electroencephalogram signal S1 of the excluded channel 7 may be interpolated by the average value of channels in the vicinity thereof, for example, the channels 5, 9, 13, 15, 19, 20, 21 in the example of FIG.
Also, for the period t ₁ t _{2 in} which the amplitude value exceeds the threshold value, the electroencephalogram signal S1 of the channel 7 is excluded, and the excluded electroencephalogram signal S1 is replaced with a pseudo electroencephalogram signal estimated from another channel or the like. The brain function state may be analyzed. In this case, the threshold value is preferably set to a value larger than the maximum amplitude value of the α wave.

Furthermore, the amplitude value is attenuated by multiplying an appropriate coefficient only for the period t ₁ t _{2 in} which the amplitude value exceeds the threshold value, and the brain function analysis is performed using the attenuated electroencephalogram signal. Good.

1 is an overall configuration diagram showing a brain function analysis system according to an embodiment of the present invention. It is a block diagram which shows the internal structure of the brain function analysis apparatus of the brain function analysis system of FIG. It is a figure which shows an example of the frequency distribution of an electroencephalogram signal. It is a figure which shows the other example of the frequency distribution of an electroencephalogram signal. It is a figure which shows the relationship between a parameter at the time of employ | adopting the average value of SN ratio as a parameter, and an electroencephalogram. It is a figure which shows an example of the processing method of the period when the average value of SN ratio was less than the threshold value in FIG. It is a figure which shows the relationship between the smoothness of the spatial transmission of the electroencephalogram used for a brain function determination, and its standard deviation. FIG. 9 is a first modification of the brain function analysis system of FIG. 1 and is a diagram showing a time change of a parameter when a variation in SN ratio is adopted as a parameter. It is a 2nd modification of the brain function analysis system of FIG. 1, Comprising: It is a figure which shows the case where the time change of SN ratio is observed for every channel. FIG. 6 is a diagram showing one means for eliminating the discontinuity of the signal after processing in FIG. 5. It is a 3rd modification of the brain function analysis system of FIG. 1, Comprising: It is a figure which shows the spatial dispersion | variation in a signal to noise ratio. It is a figure which shows the quality of the SN ratio in each electrode position in the case of FIG. FIG. 9 is a diagram showing a fourth modification of the brain function analysis system in FIG. 1 and a case where the amplitude of an electroencephalogram signal is adopted as a parameter.

Explanation of symbols

A Scalp S1 EEG signal S2 Signal to noise ratio (parameter)
S3 Corrected electroencephalogram signal S5 Threshold value 1 Brain function analysis system 5 Electrode (electroencephalogram signal detector)
7 Parameter calculation unit 8 Contact state determination unit 9 EEG signal correction unit 10 Brain function determination unit

Claims (18)

At least one electroencephalogram signal detector that is arranged in contact with the scalp and detects an electroencephalogram signal;
Based on the detected electroencephalogram signal, a parameter calculator that calculates a parameter indicating a contact state between each electroencephalogram signal detector and the scalp;
A contact state determination unit that determines the quality of the contact state by comparing the parameter calculated by the parameter calculation unit with a predetermined threshold;
An electroencephalogram signal correction unit that corrects an electroencephalogram signal when it is determined that the contact state is poor;
A brain function analysis system comprising: a brain function determination unit that determines a brain function state based on an electroencephalogram signal corrected by an electroencephalogram signal correction unit.   The brain function analysis system according to claim 1, wherein the parameter is calculated based on noise included in an electroencephalogram signal.   The brain function analysis system according to claim 2, wherein the parameter is a signal-to-noise ratio included in an electroencephalogram signal.   The brain function analysis system according to claim 3, wherein the signal-to-noise ratio is a ratio of a signal of an α wave component and a signal of other components.   The brain function analysis system according to claim 4, wherein the α-wave component signal is a signal in a wavelength band of approximately 8 to 13 Hz.   The brain function analysis system according to any one of claims 3 to 5, wherein the signal-to-noise ratio is an addition average value of signal-to-noise ratios included in the electroencephalogram signals detected by the plurality of electroencephalogram signal detection units.   The brain function analysis system according to claim 2, wherein the parameter is a variation in a signal-to-noise ratio included in an electroencephalogram signal.   The brain function analysis system according to claim 7, wherein the variation in the signal-to-noise ratio is a standard deviation of the signal-to-noise ratio.   The brain function analysis system according to claim 7, wherein the variation in the signal-to-noise ratio is a variation coefficient obtained by dividing the standard deviation of the signal-to-noise ratio by the average value of the signal-to-noise ratio.   The brain function analysis system according to claim 2, wherein the parameters include an average value of a signal-to-noise ratio included in an electroencephalogram signal and variations in the signal-to-noise ratio.   The brain function analysis system according to claim 2, wherein the parameter is an amplitude value of an electroencephalogram signal.   The brain function analysis system according to claim 11, wherein the threshold value is set to a value larger than a maximum amplitude value of the α wave.   The electroencephalogram signal correction unit removes the electroencephalogram signals detected by all the electroencephalogram signal detection units in the time domain determined to be in poor contact based on the parameter, and the electroencephalogram signals before and after the removed time domain The brain function analysis system according to any one of claims 1 to 12, wherein:   Based on the parameter, the electroencephalogram signal correction unit identifies an electroencephalogram signal detection unit determined to have a poor contact state, and determines an electroencephalogram signal detected by the identified electroencephalogram signal detection unit to have a poor contact state The brain function analysis according to any one of claims 1 to 12, wherein the brain function analysis is performed on the basis of the electroencephalogram signals before and after the time domain, or detected by another electroencephalogram signal detection unit. system.   The brain function analysis system according to claim 11 or 12, wherein the electroencephalogram signal correction unit multiplies the electroencephalogram signal over a time region determined to be in a poor contact state based on the parameter by a predetermined coefficient to attenuate the electroencephalogram signal. .   The electroencephalogram signal correction unit identifies an electroencephalogram signal detection unit that is determined to be in a poor contact state based on the parameter, and excludes the electroencephalogram signal detected by the identified electroencephalogram signal detection unit. The brain function analysis system according to any one of claims 1 to 12, wherein a brain function state is determined based on an electroencephalogram signal detected by a detection unit.   The brain function analysis system according to claim 16, wherein the brain function state is determined using a pseudo brain wave signal estimated based on a brain wave signal detected by another brain wave signal detection unit instead of the excluded brain wave signal.   The brain function analysis system according to any one of claims 1 to 17, wherein the brain function determination unit determines a brain function state based on smoothness of spatial transmission of brain waves and temporal fluctuation.

Images