JP2593625B2 - Biological information automatic identification device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for automatically identifying biological information, and more particularly to an apparatus for measuring human brain waves and analyzing neural activity in the brain in the field of medical electronics (ME). .

[0002]

2. Description of the Related Art Human thought, recognition, memory recall, pleasure / discomfort, mental fatigue, tension, etc., already reflect the electrical activity of many neurons in the brain. Widely known.

For this reason, the thinking process, recognition, recall of memory,
In addition, coordination occurs in the activity of neurons with a certain emotion, and the coordination appears as a potential on the scalp surface, which can be measured as an electroencephalogram.

Accordingly, the present inventor has already filed an application for the title of "Multidimensional Correlation Analysis of EEG" in Japanese Patent Application No. 4-317568, and has conducted qualitative analysis of cranial nerve activity.

The present invention will be described in brief. A predetermined number (for example, 21) of electrodes (electroencephalogram sensors) are provided on a human head (plan view), and output signals of these electrodes are subjected to Fourier transform. FIG. 6 shows the positions of the electrodes where the peaks of various frequency components obtained by
For example, as shown in FIG.

That is, the case shown in FIG. 6 shows a state where the correlation between the electrode outputs is strong when the subject listens to comfortable music, for example.

[0007] In this example, the detection is particularly performed based on the peak of a typical α wave in brain waves.

[0008]

In the conventional example in which the analysis of brain waves is performed by the correlation analysis method as described above, the screen is displayed as shown in FIG. However, there is a problem that it is difficult to quantify.

SUMMARY OF THE INVENTION Accordingly, an object of the present invention is to provide an apparatus for automatically identifying biological information which can be easily identified by quantifying human emotions and intellectual work contents.

[0010]

In order to achieve the above object, an automatic biological information identifying apparatus according to the present invention comprises a plurality of sensors attached to a body of a subject to detect a characteristic amount of the subject; An amplifier for amplifying the output signal of the sensor;
Each output signal of the amplifier is converted into a digital signal, and at the same time, each digital signal is Fourier-transformed to obtain a spectral power for each of a plurality of divided frequency bands within a desired frequency band. Learning and calculating and storing the coefficients and biases of the neural network so that a plurality of emotional elements or intellectual work content elements of the attached subject become standard values, and thereafter output signals and coefficients of each sensor. And a computing device that computes the level of each emotional element or intellectual work content element of the subject from a value when the bias is applied to the neural network, and a display device that displays the computation result of the computing device. I have.

Further, in the above-mentioned present invention, it is sufficient that the characteristic amount is a scalp potential or a myoelectric potential.

Further, the arithmetic unit may use a cross-correlation coefficient between sensor outputs instead of the scalp potential or the myoelectric potential as the characteristic amount.

Further, the arithmetic unit may use at least one of the number of heartbeats, eye movement and blink frequency instead of or in addition to the spectral power.

Preferably, the desired frequency band is at least one of an α band and a β band.

[0015]

In the present invention, the cerebral nerve activity of a target human is quantified, for example, with respect to emotions and intellectual work contents as follows.

That is, the arithmetic unit inputs an output signal output from a predetermined number of brain wave sensors and amplified by an amplifier.

The input analog signal is converted into a digital signal, and the digital signal is Fourier-transformed for each sensor output to divide the digital signal into desired frequency bands (for example, at least one of the α band and the β band).

Then, the spectrum power for each frequency band divided in this manner, that is, the variance on the time axis (variance) within the frequency band is obtained.

These spectral powers are input to a neural network and processed, and each coefficient in the neural network is set to a standard value representing various emotional elements or intellectual work content elements in a subject to which the sensor is attached. And bias are obtained by learning and stored.

Thereafter, by applying the coefficients and bias learned by the neural network to the sensor outputs of the subject stored in this way, to the above-mentioned feature values indicated by the subject in each situation, A value (analog value) in which the value calculated by the network indicates the level of the emotion element or the intellectual work content element.
By displaying this value, the current emotion of the subject can be determined.

By focusing on the fact that the characteristics of cranial nerve activity differ from person to person and creating a data file for each individual, accurate identification of emotions or intellectual work can be performed with high accuracy. Become.

[0022]

FIG. 1 shows the configuration of an embodiment of an automatic biological information identification apparatus according to the present invention. In the figure, reference numeral 1 denotes a subject for the electroencephalogram analysis, and number of EEG sensors (electrodes) 2 ₁ to 2 ₆ is mounted by a paste.

Note that a sensor may be attached to the head of the subject 1 so as to detect not only the potential distribution on the scalp but also the myoelectric potential as the characteristic amount indicating the cranial nerve activity of the subject. Further, instead of or in addition to the potential distribution on the scalp and the myoelectric potential, at least one of the number of heartbeats, eye movements, and blink frequency may be detected by a sensor.

[0024] After the six sensors 2 ₁ to 2 _6, which is once amplified by the preamplifier 3, after being further amplified by the main amplifier 4 to a predetermined level, given to the arithmetic unit 5, a final identification result It is given to the display device 6 and displayed. The preamplifier 3 and the main amplifier 4 constitute an amplifier.

FIG. 2 shows the processing procedure of the arithmetic unit 5 shown in FIG. 1. Hereinafter, the arithmetic processing in the arithmetic unit 5 will be described with reference to the flowchart of FIG.

[0026] First, the arithmetic unit 5 converts into digital signals by inputting the output analog signals of the sensors 2 ₁ to 2 ₆ from the main amplifier 4 (Step S1).

Then, the output of each of these sensors is subjected to a fast Fourier transform (FFT) to convert time information into frequency information (step S2).

Here, the Fourier transform will be briefly described. Assuming that the length of the time series is T seconds, the Fourier transform value is the spectral power in the frequency spectrum every 1 / T [Hz] per sensor. As shown in FIG. In this case, the frequency becomes discrete every 1 / T [Hz]. The highest frequency is 1 / 2τ [Hz] (τ is a sampling interval when A / D conversion is performed on brain waves).

As a result, a total of (1 / 2τ) / (1 /
T) = T / 2τ Fourier transform values will be generated. There are a real part and an imaginary part for each frequency, and a total of (1 / 2τ) × 2 = T / τ variables are generated.

In FIG. 3, as the Fourier transform waveform, α wave band (8 to 13) which is important information for estimating “emotional” (element) or “intelligent work contents” (element) in the brain wave.
Hz), but other δ waves (1-3H)
The same applies to z), θ wave (4 to 7 Hz), or β wave (13 Hz or more). In the following description, α
A description will be given using only the wave as an example.

In the Fourier transform waveform of the α-wave band obtained in this way, for example, three partial frequency bands α ₁ , α _{2 and} α ₃ are particularly band-pass filtered (BPF).
(Step S3). In addition, as an example of the partial frequency bands α ₁ , α _2, α ₃ , α ₁ = 8 to 9.5 Hz, α ₂ = 9.5 to
11 Hz, α ₃ = 11 to 13 Hz.

The use of the partial frequency bands α ₁ , α _2, α ₃ in the following processing as described above is performed when the processing including the above-mentioned various brain waves other than the α wave causes an excessive amount of information, resulting in discrimination. This is because it becomes difficult.

The partial circumference of each sensor output obtained in this way
Wavenumber band α₁, α_2,α_ThreeThe spectral power of each
Is the frequency band α₁, α_2,α_ThreeIs six sensors 2 each
₁~ 2 ₆X₁ ¹, X₁ ^Two, X₁ ^Three, X
₁ ^Four, X₁ ^Five, X₁ ⁶; x_Two ¹, X_Two ^Two, X_Two ^Three, X_Two ^Four, X_Two ^Five,
x_Two ⁶; x_Three ¹, X_Three ^Two, X_Three ^Three, X_Three ^Four, X_Three ^Five, X_Three ⁶Consisting of 6
× 3 = 18 spectral powers will be obtained.
(Step S4).

Incidentally, instead of the above spectral power,
"Cross correlation" between the values of the sensor outputs may be used.
When six sensors are used as described above, a total of ₆ C ₂ = 15 correlation values can be obtained for each frequency band.

The thus obtained spectral power x
By providing the neural network provided by the software in ₁ ¹ ~x ₃ ⁶ an arithmetic unit 5, an emotion or intellectual work when the output value of the neural network has listened a plurality of sounds, for example, in the subject 1 A set of a neural network coefficient and a bias that becomes a standard value to be represented is obtained by learning (step S5).

Here, the learning method of the neural network will be described with reference to FIG. 4 and the following equations showing the contents of FIG. Although the neural network uses an algorithm known as a back propagation method, other algorithms may be used.

Firstly, as shown in the following equation (1) with respect to the spectral power x ₁ ¹ ~x ₃ ^6, performs weighting by the weighting factor (C), further adding the bias (B) to the multiplication result. Then, define the calculation result _{(y 1, y 2, y} 3).

[0038]

(Equation 1)

Although the weighting coefficient (C) is set to the number of neurons N = 3, the number N of neurons can be appropriately selected as described later.

The operation result (y _1, y _2, y ₃ ) is further converted to a value (z _1, z _2, z ₃ ) by a non-linear function F as shown in FIG. ).

[0041]

(Equation 2)

[0042] And further turn, performs weighting with the value (z _1, z _2, z ₃₎ different weighting coefficients as shown in the following equation (3) with respect to (D), further bias the result of the multiplication ( B ′) is added. Then, the calculation result is expressed as (w _1, w
_2, w ₃ ).

[0043]

(Equation 3)

Note that also in this case, the number of neurons N = 3
Is assumed.

The calculation result (w _1, w _2, w ₃ ) is further converted to a value (v _1, v _2, v ₃ ) by a linear function G as shown in FIG. (Compressed).

[0046]

(Equation 4)

When the value (v _1, v _2, v ₃ ) obtained in this way is applied to, for example, “emotion” as neural activity of the brain as shown in the following equation (5), "Pleasure"
Emotions when the subject 1 listens to three sounds that are known in advance to indicate "anger" and "sorrow"
To standard values (or reference values) (100), (010), (0
01), and the weighting coefficients (C) and (D) and the biases (B) and (B ′) at this time are obtained by iterative learning (Iteration).

[0048]

(Equation 5)

In the above learning, it is necessary to use three linear function parts G in order to correspond to three emotions. However, before this linear function part G, an appropriate number of weighting coefficients (C ), (D), bias (B) and non-linear function (F).

After calculating the set of the weighting coefficients (C) and (D) and the biases (B) and (B ') for the value representing the emotion when the plurality of predetermined sounds are heard by the subject 1, Are stored in a memory (not shown) built in the arithmetic unit 5 (step S5).

The set of the weighting coefficients (C) and (D) and the biases (B) and (B ′) stored in the memory indicates the unique characteristics of the subject 1 regarding the three emotions. relative spectral power x ₁ ¹ ~x ₃ ⁶ obtained from each sensor output after this (step S4), and 4 and the above-described formula (1) to (5) weighting coefficients stored in the memory as shown in ( C) and (D) and the set of biases (B) and (B ′) apply.

As a result, the values (v _1, v
_2, v ₃ ) is not a binary value as in equation (5), but is output as an analog value, for example, “0.5”, “0.3”, “0.1”, respectively. The state can be displayed on the display device 6 using a bar chart as shown in FIG. 5, a pie chart, a symbol, a color, or the like (step S6).

[0053] Here, it described in the simplest case the neural network learning methods, and it is sufficient with only the output z ₁ in FIG.

That is, the weighting coefficient (C) is (c _1, c _2, ...)
By setting only c ₁₈ ) and the bias (B) to only b _1, only the value y ₁ is obtained as shown in the following equation (6).

[0055]

(Equation 6)

Then, this value y ₁ is converted into one nonlinear function F
, F (y ₁ ) = z ₁ is obtained. This value z _1, the value indicating the unpleasant "0" determined in advance a value indicating pleasure as "1".

For example, the subject 1 most dislikes the “noise” for the values “0” and “1” representing emotions in advance.
And the music that test subject 1 prefers, or
The subject 1 is set to select a key of a keyboard provided in the arithmetic unit 5 when such music or sound is heard.

Substituting these “1” and “0” and learning
The weighting factor (c _1,c_2,... c₁₈) And ba
Ias b₁Can be calculated, and these are stored.
Applied to each sensor output in the above equation (6)
Then, the ratio of the discomfort and the pleasure of the subject's emotion 1
Can be displayed on the display device 6.

Although the above embodiment has been described by classifying emotions into three types, it is possible to consider many more emotions, and the values of the emotions corresponding to these are also limited as described above. It is not something to be done.

In the above description, emotions have been described as examples, but the present invention can be similarly applied to intellectual work contents such as mental arithmetic, graphic recognition, and voice recognition.

[0061]

As described above, according to the apparatus for automatically identifying biological information according to the present invention, the digital output signals of a plurality of sensors attached to the body of a subject and detecting the characteristic amount of the subject are subjected to Fourier transform. Determine the spectral power for each of a plurality of divided frequency bands in the desired frequency band,
Learning and storing coefficients and biases of the neural network so that the spectral power becomes a standard value for a plurality of emotional elements or intelligent work content elements of the subject to which the sensor is attached by the neural network. Since each level of the subject's emotional element or intellectual work content element is calculated and determined from the output signal of each sensor and the value when each coefficient and bias are applied to the neural network, the following is performed. Automatic identification becomes possible.

(1) For example, when performing mental arithmetic, it is possible to identify how much concentration was made in mental arithmetic, what kind of thinking form was taken, and how much thinking was rested. it can.

(2) It is possible to identify and quantify what kind of emotions have occurred in addition to pleasure when receiving a stimulus that should provide pleasure.

(3) If the same result is obtained every time when the state of mind when the same comfortable stimulus is given for a long period of time, for example, every week, is obtained every time, the emotion of the person Can be said to be very stable, otherwise there is some emotional instability, and the instability can classify the degree of emotional instability.

(4) If the same work is repeated, fatigue gradually accumulates. In such a case, if the mental state is automatically identified, the content of the mental state changes as the work time elapses. That is, since the mental response to the external stimulus gradually changes, the accumulated state of fatigue can be monitored from such information. In other words, when some kind of relaxation treatment is performed in order to eliminate the fatigue state, it is possible to quantitatively measure the degree of the relaxation effect of the treatment.

(5) In addition, if the difference in how to receive the same sensitivity is classified and the sensitivity type of each person is classified, not only each individual but also the type of each individual can be determined. It helps classify your talents and even help you choose which profession is right for you.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of a biological information automatic identification device according to the present invention.

FIG. 2 is a flowchart showing a processing procedure of an arithmetic unit used in the automatic biological information identifying apparatus according to the present invention.

FIG. 3 is a waveform diagram showing frequency components of an α-wave band as a predetermined frequency band in the biological information automatic identification device according to the present invention.

FIG. 4 is a block diagram showing a concept of neural network processing by software in an arithmetic unit in the automatic biological information identification apparatus according to the present invention.

FIG. 5 is a bar graph showing an example of emotion identification in the automatic biological information identification apparatus according to the present invention.

FIG. 6 is a graph showing an electroencephalogram state of a subject due to an α-wave peak identified by a conventional example.

[Explanation of symbols]

DESCRIPTION OF REFERENCE NUMERALS 1 test subject 2 _{1 to} 2 ₆ sensor 3 preamplifier 4 main amplifier 5 arithmetic unit 6 display device

Claims (6)

(57) [Claims] 1. A plurality of sensors attached to a body of a subject to detect a characteristic amount of the subject, amplifiers for amplifying output signals of the sensors, and converting each output signal of the amplifier into a digital signal. Fourier transform each digital signal to determine spectral power for each of a plurality of divided frequency bands within a desired frequency band, and the spectral power is further changed by a neural network to a plurality of emotional elements or intellectual work contents of a subject to which the sensor is attached. For the element
Subsequent output signals and each coefficient and the bias of each sensor is stored in search to learn coefficients and bias of the neural network such that the standard value each from the value when applied to the neural network of the subject Each emotion
An automatic biological information identification device, comprising: a calculation device for calculating the level of an element or an intellectual work content element ; and a display device for displaying a calculation result of the calculation device. 2. The biological information automatic identification apparatus according to claim 1, wherein the arithmetic unit is an scalp potential or a myoelectric potential as the feature amount. 3. The biological information automatic identification apparatus according to claim 1, wherein the arithmetic unit uses a cross-correlation coefficient between the outputs of the sensors instead of the spectrum power. 4. The apparatus according to claim 1, wherein the arithmetic unit uses at least one of the number of heartbeats, eye movement and blink frequency instead of the scalp potential or the myoelectric potential as the feature amount. Biological information automatic identification device. 5. The apparatus according to claim 1, wherein the arithmetic unit uses at least one of the number of heartbeats, eye movements, and blinking frequency in addition to the scalp potential or the myoelectric potential as the feature amount. Biological information automatic identification device. 6. The apparatus according to claim 1, wherein the desired frequency band is at least one of an α band and a β band.