JP5118230B2 - Brain activity measuring device - Google Patents

Description

  The present invention relates to a brain activity measuring apparatus, and more particularly to an apparatus for measuring a brain activity state in order to determine a brain disease such as senile dementia disorder.

  There are statistics that brain diseases, especially nearly 30% of people aged 90, have dementia, which is becoming a major problem for the aging society.

  Along with this, it is desirable to detect such cognitive impairment as early as possible and treat it before reaching a severe cognitive state. A plurality of brain potential sensors attached to the subject's head and measuring the brain potential of the subject, and converting the alpha wave component in the output signal of each brain potential sensor into numerical data, The dipole degree is obtained, and the mean value and variance of the square error between the brain potential by the equivalent dipole at the time of occurrence of the peak of the dipole degree and the measured brain potential within a certain time range and the variance are obtained for each brain potential sensor. A brain activity measuring device comprising: a computing device that creates a contour map relating to the distribution of the average value or variance on the scalp or the brain surface corresponding thereto; and an output device that outputs the contour map The already proposed (e.g. see Patent Document 1.).

In the above-mentioned Patent Document 1, the presence of a decrease in neuron function somewhere in the brain is detected by analyzing the brain wave potential distribution of alpha waves, and thus has the following problems.
B) There are quite a few people (10-15%) who show little alpha wave.
B) Alpha waves are suppressed and very unstable in the open eye state.
C) Alpha waves are greatly affected by sensitivity.
D) Measurement of the degree of cerebral dysfunction using only alpha waves cannot identify cerebral dysfunction sites.

  As described above, since Patent Document 1 performs measurement based on alpha waves to the last, it does not detect a neuronal function-decreasing site, and the degree of brain disease is unknown. There is a problem in that it is unclear whether it has recovered.

  Therefore, the present inventors have previously developed a brain activity measuring device that can clearly indicate the degree of brain disease and the reduced / recovered portion of neuronal function without being limited to alpha waves (eg, Patent Documents). (See 2.)

  That is, at this point, the present inventors have discovered that neuronal activity becomes unstable when neuronal function in the cerebral cortex decreases. This effect appears as fluctuations in local brain potential power (T. Musha, T. Asada, F. Yamashita, T. Kinoshita, H. Matsuda, M. Uno, Z. Chen and WRShankle, "A new EEG method for controlling cortical neuronal impairment that is sensitive to early stage Alzheimer's disease, "Clinical Neurophysiology, 113 (2002) 1052-1058). It was confirmed that this property is not limited to alpha waves, but covers a wider range of brain potential frequencies (for example, 2 to 40 Hz).

  The brain activity measuring device developed based on this is equipped with a plurality of sensors attached to the head of the subject to measure the brain potential of the subject, and the frequency of the alpha wave of the brain potential output from each sensor. The predetermined frequency band is further divided into a predetermined number of frequency banks having a constant frequency width, and the data of each divided frequency bank is divided into segments of a predetermined time width on the time axis, and the normalized power variance obtained for each segment is obtained. (Normalized Power Variance: hereinafter simply referred to as NPV) and the average value and standard deviation between normalized power variances obtained in advance in the same manner for a given normal population in the same manner. A Z-score, and an arithmetic unit that maps the average value of the Z-score for each brain potential sensor over the entire frequency bank on the brain surface. It was to create a local function decline map of emissions.

  Since this map shows how much the subject's brain activity is deviated from the normal person, that is, the “abnormality” of the neuron activity, the Z score of the subject is calculated. There are two types of abnormalities, and if the Z score> 0, the fluctuation of neuronal activity is larger than that of the normal population, indicating that it is “unstable”, whereas the Z score <0 In some cases, it can be shown that neuronal activity is more “inactive” than the normal population.

  In this way, in order to avoid artifacts such as blinking and induction from commercial AC power, all brain potentials of 2 Hz to 40 Hz were targeted, so the frequency range can be expanded from alpha waves, regardless of whether eyes are open or closed It will be applicable.

  Furthermore, if a standard template for various brain diseases is created clinically, differential diagnosis for those diseases can be performed, and the details of the therapeutic effect for each brain disease can also be known from the changes in this map. It was something that could be done.

Japanese Patent No. 3581361 Patent No. 4145344

However, the above Patent Document 2 has the following problems.
1) Since the Z score is only mapped on the brain surface, it is difficult to identify brain diseases that differ only in the depth of neuronal abnormal activity from the brain surface.
2) When calculating the normalized power variance (NPV), the brain potential obtained from each sensor (channel) on the scalp, that is, the time series data of brain potential, is divided into frequency banks by fast Fourier transform (FFT). Since NPV was calculated after performing inverse fast Fourier transform (IFFT) on time series data, the number of computations became excessive and delay time occurred as both FFT and inverse FFT operations were performed Was. Therefore, when data processing is performed intensively by a server via the Internet, traffic congestion on the line is expected due to an increase in the number of users of the system, and signal processing is greatly delayed. .

  Accordingly, an object of the present invention is to provide a brain activity measuring apparatus that realizes three-dimensional identification of a region where neuronal activity is abnormal and shortens its processing time.

  The present inventor found that the statistical fluctuation of the NPV value calculated for each segment is large, and even after averaging the segments, the signal-to-noise ratio is about 1, and this is because the NPV is obtained for each segment. I found out that there was a cause.

  Therefore, the brain activity measuring device according to the present invention includes a plurality of sensors attached to the subject's head and measuring the brain potential of the subject, a brain potential output from each sensor, and a grid set in advance in the standard brain Obtain the current component in the x, y, or z direction estimated at the grid point from the predetermined coordinates of the points and the predetermined coordinates of each sensor, or the magnitude of the combined current of the current components; The size is divided into segments of a predetermined time width on the time axis, and Fourier transform is obtained for each segment within a predetermined frequency range to obtain Fourier coefficients, and the average of the squares of absolute values over all segments for each Fourier coefficient A value is obtained and divided into frequency banks so as to include a plurality of adjacent average values among them; normalization power is calculated using the average value of the squares of the absolute values of the adjacent Fourier coefficients for each frequency bank. Obtaining the variance and obtaining the average value of the normalized power variance over all frequency banks; the subject from the average value and the standard deviation between the normalized power variances obtained in advance in the same manner for a predetermined normal population Or obtain the subject's Y score from the average value between the standardized power variances obtained from all grid points of the subject and the standard deviation thereof instead of using the normal population as a reference; And an arithmetic unit for converting the Z score or Y score corresponding to the lattice point on the horizontal plane into a contour map.

  The predetermined frequency range is, for example, 2 to 40 Hz, and the two adjacent plural bands include a frequency band of 0.78 Hz, and the segment width is, for example, 2.56 seconds.

Furthermore, in the present invention, the sensor is provided in a terminal device, the arithmetic device is provided in a calculation center, the terminal device and the calculation center are connected via a communication line, and brain potential data is stored in the calculation center. It is collected and stored as a database, and the database can be referenced from the terminal device.

  Furthermore, in the present invention, brain potentials output from a plurality of sensors attached to the subject's head and measuring the brain potential of the subject, predetermined coordinates of lattice points set in advance in the standard brain, predetermined coordinates of each sensor, To determine the current component in the x, y, or z direction estimated from the grid points or the combined current magnitude of each current component; Dividing into segments, Fourier transform is performed for each segment within a predetermined frequency range to obtain Fourier coefficients, and for each Fourier coefficient, the average value of the squares of absolute values over all segments is obtained, and adjacent to each other from these Dividing into frequency banks so as to include a plurality of the average values; obtaining a normalized power variance using the average value of the squares of the absolute values of the adjacent Fourier coefficients for each frequency bank and the entire circumference An average value of the normalized power variances over several banks is obtained; a Z score of the subject is obtained from an average value between the normalized power variances obtained in advance in the same manner for a predetermined normal population and its standard deviation Or, instead of using the normal population as a reference, obtain the Y score of the subject from the average value between the standardized power variances obtained from all grid points of the subject and the standard deviation thereof; on the specified horizontal plane There are provided a program for causing a computer to execute a procedure for converting the Z score or Y score corresponding to a lattice point into a contour map, and a computer-readable recording medium on which the program is recorded.

  In the present invention, abnormal sites of neuronal activity can be displayed inside the brain, so that brain disease diagnosis can be realized at low cost, non-invasiveness, high sensitivity, and high reliability. Therefore, it is expected to spread to small and medium medical institutions. This also serves as a means of suppressing Alzheimer's disease, which requires early detection for prevention, and can also be used as a diagnosis and monitor for children with recently increasing depression and developmental disorders. large.

1 is a block diagram showing a configuration example [1] of a brain activity measuring apparatus according to the present invention. FIG. FIG. 3 is a block diagram showing a configuration example [2] of a brain activity measuring apparatus according to the present invention. FIG. 6 is a flowchart showing a procedure for displaying a Z score as a map as an operation example I of the brain activity measuring apparatus according to the present invention. FIG. 4 is a flowchart showing the flow chart shown in FIG. 3 in an easy-to-understand manner. It is the figure which showed the example of arrangement | positioning of the lattice point in a brain used in each operation | movement Example of the brain activity measuring apparatus based on this invention. FIG. 3 is a contour map diagram of the Z score of the x-direction current component at each grid point on a certain horizontal plane obtained in the operation example 1 of the brain activity measuring apparatus according to the present invention. It is a contour map figure of the Z score of the y direction current ingredient in each lattice point on a certain horizontal plane obtained in operation Example 1 of the brain activity measuring device concerning the present invention. FIG. 3 is a contour map diagram of the Z score of the z-direction current component at each grid point on a certain horizontal plane obtained in the operation example 1 of the brain activity measuring apparatus according to the present invention. Operation of the brain activity measuring apparatus according to the present invention The Z score obtained for the magnitude of the current obtained by synthesizing the current components in the x, y, and z directions at each lattice point on a horizontal plane obtained in the second embodiment is displayed as a map. It is the flowchart figure which showed the procedure. FIG. 4 is a contour map of a Z score obtained for the magnitude of current obtained by combining the current components in the x, y, and z directions at each grid point on a certain horizontal plane obtained in the second embodiment of the brain activity measuring apparatus according to the present invention. is there. It is the flowchart figure which showed the procedure of carrying out the map display of Y score as operation Example 3 of the brain activity measuring apparatus which concerns on this invention.

* Configuration example: Figs. 1 and 2
FIG. 1 shows a configuration example [1] of a brain activity measuring apparatus according to the present invention. In this configuration example, first, (but it referred electrode both hereinafter, occasionally in the sensor.) For example Brain potential of 21 before and after the sensor or MEG sensors on the head 1 2 1 _{to 2} 21 _{(hereinafter,} reference numeral 2 The brain potential based on intracerebral nerve activity is measured, or a cap or a helmet in which these sensors are arranged in advance is worn. In this case, the sensor 2 is disposed at a position determined by the International 10-20 standard, and a sensor (not shown) is also attached to the right earlobe, for example, as a reference potential.

  The brain potential measured by the sensor 2 is supplied to an analog / digital converter (A / D) 5 via an amplifier 3 and a multiplexer 4, and the digitized measured brain potential) data is input interface (I / F) 15. To be supplied to the computer 10. Note that the input interface 15 may pass the data as it is, or only outputs a component having a predetermined frequency band (for example, a predetermined frequency band wider than the frequency of the alpha wave) after performing digital filtering. May be.

  In the computer 10, a CPU 11 is connected to a ROM 13, a RAM 14, an input interface 15, and an output interface 16 via a bus 12. The ROM 13 is a read-only storage medium, and the RAM 14 is a memory that stores brain potential data from the keyboard 24 and the A / D converter 5 during calculation.

  The input interface 15 is connected to an external storage device 25 storing a program and the like, and the output interface 16 is displayed on the display device 31 such as a CRT for displaying the calculation result of the computer 10. A printer 32 for storing data and waveforms is connected as an output device. Note that the program or the like may be stored in advance in the ROM 13 without using the external storage device 25.

  Here, as shown in the configuration example [2] of FIG. 2, the above brain potential data is transmitted from the interface 17 of the clinical site computer 10 operating as a data transfer terminal device via a communication line 41 such as the Internet. Then, it is sent to a calculation center 42 as an arithmetic unit, and the result analyzed here is sent back to the computer 10 again via the communication line 41, and is output at an output device such as the CRT 31 or the printer 32. You may comprise as follows. In this case, the server device, the program, and the recording medium are installed in the calculation center 42, and the server device or the like collects and accumulates brain potential, that is, data relating to brain potential, and accumulates data relating to a specific brain disease. From the computer 10, data for each specific brain disease stored in a database can be referred to.

Hereinafter, operation examples of generation of a neuron abnormality display map and abnormality identification will be described with reference to FIGS.
* Operation example 1: Contour map display example of Z score of current component in x direction (or y direction or z direction) at each grid point on a certain horizontal plane (FIGS. 3 to 8)
An operation example 1 in the above-described configuration example will be described below with reference to FIG. 3 and a flowchart of FIG. 4 shown at a different angle corresponding thereto. Note that after the sensor group 2 shown in the head 1 is arranged, a power supply (not shown) is turned on to set the computer 10 to an initial state. In addition, the brain potential is measured based on intracerebral nerve activity from the ₂₁ sensors 2 ₁ to 2 ₂₁ placed on the head 1 at regular sampling time intervals.
(1) Creation of normal population database (solid arrow route)
Step S1a, S2:
First, a group of normal people in advance, such as MRI (Magnetic Resonance Imaging) and SPECT (Single Photon Emission Computing Tomography) measurements using the Mini Mental State Test (MMSE) method to evaluate existing cognitive functions The brain potential (scalp potential) is measured from the sensor 2 one by one.

In this case, the potential signals of the sensors 2 ₁ to 2 ₂₁ are sampled every 5 ms. The frequency range is set to a constant frequency band (for example, 2 to 40 Hz) wider than the frequency band of the α wave by bandpass filter processing.

Step S3:
By subtracting a constant from each measured value so that the average of 21 brain potentials becomes zero for each sample, a deviation from the average value of 21 brain potentials is calculated.

Step S4:
As shown in FIGS. 5 (1) to (3), 1544 grid points (j) set in advance in the standard brain are set. This is a three-layer skull model (the standard brain is divided into three geometric regions: “skull”, “cerebrospinal fluid layer”, and “brain tissue”, and their conductivity is 1/80: 3: 1. If a cubic lattice with a 1 cm interval is assembled in the standard head model and its position coordinates are determined, the number of lattice points is 1,544. Then, current sensor dipole (hereinafter simply referred to as current) vector component values (1544 × 3 = 4632) (1544 × 3 = 4632) at each lattice point are measured in advance using 21 sensor coordinates and sensor 2 _1. calculating from the measured brain potentials of to 2 _21.

The method of calculating the three-dimensional distribution of electromotive force in the brain from the potential distribution on the scalp is as sLORETA, eLORETA (Low Resolution Brain Electromagnetic Tomography).
The algorithm is public.

  The processing of the LORETA output data is executed 1544 times × 3 (x, y, z) = 4632 times at all grid points j.

Step S5:
The signal time series of the sensor (channel) is divided into s (for example, s = 70) segments on the time axis. In this case, the segment length is 2.56 seconds (= 5 ms × 512).
Step S6:
For the frequency range to be analyzed (for example, 2 to 40 Hz), discrete Fourier transform is performed for each segment (step S6a), and a Fourier coefficient H _jm having a frequency of an integral multiple mf ₀ of the fundamental frequency f ₀ for each segment is obtained ( The same is calculated), and the average value <| H _jm | ² > of the square of the absolute value over all the segments is calculated (step S6b), and M (in order to include n (for example, two) values adjacent to each other) For example, 96) frequency banks are configured (step S6c).

For example, if the sampling frequency is 200 Hz (5 ms) and the segment length is 2.56 seconds as described above, the discretized frequency is an integer multiple of the fundamental frequency f ₀ (= 1 / 2.56 = 0.39 Hz), and the lowest frequency is stored in bank # 1. 6f ₀ = 2.34Hz and 7f ₀ = 2.73Hz, the next bank contains 7f ₀ and 8f ₀ , and the highest frequency bank # 96 is distributed 101f ₀ = 39.41Hz and 102f ₀ = 39.8Hz, It is divided into 96 frequency banks in total.

Step S7:
As described above, for each frequency bank # 1 to # 96 to which two frequency components are assigned, each grid point j and each frequency bank m (in this example, m starts from “6”, and frequency bank # 1 has m = 6 The standardized power variance NPV _jm is calculated and stored in the RAM 14. The calculation process of NPV _jm is as follows.

Fourier-expand x _n, n = 0 to N−1 _, where one segment of length T sec contains N data samples. Let the Fourier coefficient be H _m . The frequency is mf ₀ and f ₀ = 1 / T is the fundamental frequency.

  here

In equation (1), only two frequency components, that is, mf ₀ and If only (m + 1) f ₀ is left, the nth sample data is as follows.

  cc means the complex conjugate term. On average across all segments

  In the same way

  On the other hand, NPV is expressed by the following formula.

  Therefore, when Expressions (4) and (5) are substituted into Expression (6), the following Expression (7) is obtained.

Step S8:
By repeating the above steps S1a to S7 for all normal subjects, the group mean <NPV _{NL, jm} > for the normal population and the standard deviation σ _{NL, jm within the group} are calculated, and the result is Stored in the RAM 14 as a database.
(2) Creation of subject's Z score map (double solid arrow route)
Step S1b:
The brain potential from the sensor 2 attached to the subject is measured.

Steps S2 to S7:
As in the case of the normal population, steps S2 to S7 are executed to calculate the normalized power variance NPV _jm of the subject.

Step S9:
The subject's Z score is determined from the subject's normalized power variance NPV _jm , the average normalized power variance <NPV _{NL, jm} > of the normal population, and the standard deviation σ _{NL, jm} according to the following equation.

Step S10:
For the standard model of the head, specify the shape of the horizontal cross section at 1 cm intervals and the coordinates of the lattice points so as to include the designated lattice points.

Step S11:
Draw contour lines calculated by interpolation from the subject's Z-score values related to currents in the x, y, or z directions derived at grid points on each horizontal section.

  Contour lines in the x direction are shown in FIG. 6, contour lines in the y direction are shown in FIG. 7, and contour lines in the z direction are shown in FIG. In both cases, the position in the vertical direction when the horizontal plane of the brain is sliced is indicated by z.

  The determination of the contour lines indicates that the more white the image is, the larger the positive value of the Z score is, and the blacker the value is, the larger the negative value of the Z score is, indicating an abnormal part of the neuron.

That is, the Z score takes a negative value when the normalized power variance NPV _j takes a value smaller than the average value <NPV _j > for the lattice point j of the normal population. In other words, it means that the power fluctuation of the subject's lattice point is smaller than the power fluctuation of the corresponding lattice point j of the normal person, which indicates that the neuronal activity is inactive compared to the normal person. .

On the other hand, a positive value of the Z score represents an anomaly that the fluctuation of neuronal activity is larger and more “unstable” than the normal population. And the larger the absolute value, the more abnormal.
<Explanation of high-speed calculation processing>
When each segment includes only two Fourier coefficients H _f1 and H _f2 as shown in step S6c of FIG. 4, the NPV for each segment is calculated as shown in equation (7). In this case, the calculation time is shortened to about 1% compared to the case where these two Fourier coefficients are changed in time series by inverse Fourier transform.

  In the case of the present invention, when the frequency bank includes two frequency components (average value of the square of the absolute value of the Fourier coefficient), the time-series data for 5 minutes is divided into about 117 segments, and Fourier Since the NPV value can be calculated directly from the coefficient, it is not necessary to perform the inverse FFT operation. The number of FFT operations is 117. In the case of the above-mentioned Patent Document 2, it is necessary to perform the inverse FFT operation after the FFT operation. Therefore, the calculation processing time is remarkably shortened as compared with Patent Document 2.

Furthermore, in the present invention, the square value of the absolute value of the Fourier coefficient is averaged over 70 segments and then used to calculate the NPV, so that the relative statistics contained in the NPV are calculated. Noise is reduced to (117) ^-0.5 times, or 10%. On the other hand, since the relative noise of the Fourier coefficient is eight times the relative noise inherent in the Fourier coefficient from the above equation (7), the relative noise inherent in the NPV is that of the noise inherent in the Fourier coefficient. It will reach 2 × 8 = 16 times, and the noise reduction effect is great.
* Operation example 2: Contour map display example of Z score of magnitude of combined current in x direction, y direction and z direction at each lattice point on a certain horizontal plane (FIGS. 9 and 10)
The difference between the present embodiment and the operation embodiment 1 described above is that step S4 in the flowchart of FIG. 3 is replaced with step S4a in FIG.

That is, in the case of Example 1, the contour map shown in FIGS. 6 to 8 from the value of the Z score of the subject related to the current component in the x direction, the y direction, or the z direction derived to the grid points on each horizontal section. However, the present Example 2 is different in that the contour score map is displayed by obtaining the Z score from the magnitudes of the combined currents in the x direction, the y direction, and the z direction at each lattice point. Therefore, the processing of the output data in FIG. 4 is 1544 times. Others are the same as in the first embodiment.
* Operation Example 3: Contour map display example of the Y score of the current component in the x direction (or y direction or z direction) at each lattice point on a certain horizontal plane (FIG. 11)
The difference between the present embodiment and the operation embodiment 1 described above is that step S9 in the flowchart of FIG. 3 is replaced with step S9a in FIG.

That is, in the case of Example 1, the contour map shown in FIGS. 6 to 8 from the value of the Z score of the subject related to the current component in the x direction, the y direction, or the z direction derived to the grid points on each horizontal section. However, in the case of Example 3, the difference is that the contour map is generated by obtaining the “Y score” of the subject related to the current component in the x direction, the y direction, or the z direction at each lattice point. . Others are the same as in the first embodiment. However, although the generated contour map itself is different from the contour map of the first embodiment, the illustration is omitted here.

The Y score Y _jm in this embodiment is calculated from the average value <NPV _{, jm} > of the NPV _{jm at} each lattice point and the standard deviation σ _jm as follows.

* Operation Example 4: Contour map display example of Y score of magnitude of combined current in x direction, y direction, and z direction at each lattice point on a certain horizontal plane (not shown)
The difference between this embodiment and the above-described operation embodiment 2 is that the contour map is generated by obtaining the “Y score” of the subject instead of the Z score generating step S9 in the flowchart of FIG. is there. The contour map is not shown here.

  It should be noted that the present invention is not limited to the above-described embodiments, and various modifications can be made by those skilled in the art based on the description of the scope of claims, and the present invention is also applicable to depression as a brain disease. It is clear.

1 Subject's head
2, 2 _{1 to} 2 ₂₁ Brain potential (magnetoencephalogram) sensor
3 Amplifier
4 multiplexer
5 A / D converter
10 computers
11 CPU
13 ROM
14 RAM
15, 16, 17 interface
24 keyboard
25 External storage
31 CRT
32 Printer
41 Communication line
42 Calculation Center In the figure, the same symbols indicate the same or corresponding parts.

Claims (5)

A plurality of sensors attached to the subject's head to measure the brain potential of the subject;
Current component in x-direction, y-direction or z-direction or current component estimated at the grid point from the brain potential output from each sensor, the predetermined coordinates of the grid point set in advance in the standard brain, and the predetermined coordinates of each sensor The current component or current magnitude is divided into segments of a predetermined time width on the time axis, and Fourier transform is performed for each segment within a predetermined frequency range to obtain a Fourier coefficient. For each Fourier coefficient, the average value of the square of the absolute value over the entire segment is obtained, and divided into frequency banks so as to include a plurality of adjacent average values from each other; A standardized power variance is obtained using an average value of the squares of the absolute values of the Fourier coefficients, and an average value of the standardized power variance over all frequency banks is obtained; In the same way, the Z score of the subject is obtained from the average value between the standardized power variances obtained in advance and the standard deviation thereof, or obtained from all grid points of the subject instead of using the normal population as a reference. An arithmetic unit that obtains a Y score of the subject from an average value between standardized power variances and a standard deviation thereof; and calculates a contour map of the Z score or Y score corresponding to a lattice point on a specified horizontal plane;
A brain activity measuring device comprising: In claim 1,
2. The brain activity measuring apparatus according to claim 1, wherein the predetermined frequency range is 2 to 40 Hz, the two adjacent to each other includes a frequency band of 0.78 Hz, and the segment width is 2.56 seconds. In claim 1,
The sensor is provided in a terminal device, the computing device is provided in a calculation center and the terminal device and the calculation center are connected via a communication line, and data relating to brain potential is collected and accumulated as a database in the calculation center, A brain activity measuring device characterized in that the database can be referred to from the terminal device.   From the brain potential output from a plurality of sensors attached to the head of the subject and measuring the brain potential of the subject, the predetermined coordinates of the lattice points set in advance in the standard brain, and the predetermined coordinates of each sensor at the lattice points Determine the estimated current component in the x-direction, y-direction or z-direction or the combined current magnitude; divide each current component or current magnitude into segments of a predetermined time width on the time axis, A Fourier coefficient is obtained by Fourier transform for each segment within a predetermined frequency range, an average value of squares of absolute values over all segments is obtained for each Fourier coefficient, and a plurality of the average values adjacent to each other are obtained. In each frequency bank, the normalized power variance is obtained by using the average value of the squares of the absolute values of the adjacent Fourier coefficients for all the frequency banks. An average value of the normalized power variance is obtained; a Z score of the subject is obtained from an average value between standardized power variances obtained in advance in the same manner for a predetermined normal population and its standard deviation, or the normal person The Y score of the subject is obtained from the average value between the standardized power variances obtained from all the grid points of the subject instead of the group as a reference, and the standard deviation thereof; the grid corresponding to the grid point on the designated horizontal plane A program for causing a computer to execute a procedure for converting a Z score or a Y score into a contour map.   A computer-readable recording medium on which the program according to claim 4 is recorded.

Images