JP2005287641A - Brain function degradation degree estimating apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an easy-to-handle and inexpensive brain function degradation degree estimating apparatus using the fewest possible sensors. <P>SOLUTION: This apparatus stores NPV (the variance of power within the fixed period of the output signals of the sensor made into numerical data normalized by the square value of the average of the power) thresholds for an Alzheimer's patient and a normal person respectively obtained from sensitivity and specificity obtained for the NPV obtained by attaching the sensor to the forehead parts of the Alzheimer's patient diagnosed beforehand and the normal person, compares the NPV obtained by attaching one sensor to the forehead part of a subject with the NPV threshold, judges that the subject is the Alzheimer's patient when the NPV of the subject is equal to or more than the NPV threshold of the specificity, and decides that he/she is the normal person when it is equal to or less than the NPV threshold of the sensitivity. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a brain function deterioration level estimation device, and more particularly to a device for estimating a brain function deterioration level such as senile dementia disorder.

  Regarding senile dementia, there are statistics that nearly 30% are demented at the age of 90, which is becoming a big problem for the aging society in the future.

  Accordingly, it is desired to detect such dementia disorders as early as possible and treat them before reaching a severe dementia state. The measurement (determination) of this dementia disorder has been conventionally performed as shown below. It is done in a different way.

(1) Hasegawa-style simplified intelligence evaluation scale (HDS)
(2) National Seken-style dementia screening test (3) N-type mental function test (4) Mental States Questionnaire (MSQ)
(5) Mini-mental state test (MMSE) method (6) Erasawa-style "clinical criteria for senile intelligence"
(7) Functional assessment staging (FAST) method (8) Clinical dimension rating (CDR) method (9) GBS scale (10) N-type mental state scale for the elderly (NM scale)
All the conventional methods for measuring dementia as described above have the following problems because all doctors employ a format in which the subject (patient) is inquired.

  1) Since the listener is involved, the dependence on the specific human relationship between the listener and the subject is large, and the answers are not always accurately obtained objectively, and the judgment results vary.

  2) The subject may learn the contents of the interview while repeating the test, and objective judgment results cannot be obtained.

  3) The subject may refuse to answer.

  4) The ability to identify early dementia is low.

  In addition, methods using SPECT (Single Photon Emission Computing Tomography), PET, etc., inject radioactive materials into blood vessels and use them as clues for measuring radiation doses radiated in the brain. The diagnostic cost becomes very high.

  Therefore, the present inventor has already proposed a brain activity measurement (determination) device that is inexpensive, non-invasive, highly sensitive, highly reliable, and easy to operate (see, for example, Patent Document 1). This will be described below.

  When neurons in the brain cortex are activated, an electromotive force is generated, a current flows perpendicularly to the cortical surface, and a potential distribution is generated on the scalp. This potential distribution can be approximated by one current dipole assumed in the brain. A dipole in which the value obtained by averaging the square error value between the potential generated at the sensor position by the dipole and the measured potential for all sensors is minimized is called an equivalent dipole.

  The equivalent dipole for a limited-band electroencephalogram creates a smooth potential distribution on the scalp through the potential distribution recorded by the sensor (electrode). The dipole degree represents the approximation of the equivalent dipole potential and minimizes the mean value of the square error between these two potentials at the sensor position. Therefore, the degree of dipole is an indicator of the smoothness of the scalp potential. When the intracortical neuronal activity is uniform, the dipole degree is close to 1, but when the neuronal activity is non-uniform, the dipole degree decreases. A decrease in the degree of dipole represents a decrease in neuronal activity. Since the dipole degree of the narrow-band EEG fluctuates approximately with respect to time, the average of the peak values is called the average dipole degree.

  And it became clear that there is a threshold for the average dipole degree, and normal and dementia can be distinguished from that threshold, and based on this, quantification of dementia, especially Alzheimer type dementia, is possible, It became possible to distinguish between normal and dementia at a certain accuracy rate.

  The peak value of the dipole degree fluctuates with time, and its standard deviation increases with neuronal function deterioration. Such a standard deviation also has a threshold value. When the standard deviation is larger than the threshold value, it can be diagnosed as Alzheimer-type dementia.

The average dipole degree D _alpha about Alpha waves decreases with the progress of Alzheimer's disease, SPECT that cerebral blood flow from either side lobe toward parietal lobe is reduced with a positive correlation with the average dipole degree D _alpha Confirmed by. This is a characteristic trend of early Alzheimer's disease.

Assuming the subject's head is spherical, the calculation of the dipole degree is very simple. This method is very simple to calculate, and there are few variations in the determination results as in the MMSE method. Furthermore, the cost can be greatly reduced compared to the SPECT method, etc., and the sensitivity for discriminating between normal and abnormal Will improve.
Japanese Patent Laid-Open No. 2002-248087 (summary, FIG. 1)

  However, in the case of the above-mentioned Patent Document 1, for example, many 21 sensors must be attached to the subject, and there is a problem that handling is difficult and cost is high.

  Therefore, an object of the present invention is to provide a brain function deterioration degree estimating apparatus that is easy to handle and inexpensive, using as few sensors as possible.

  In order to solve the above problems, the present inventor attempted to determine the degree of brain deterioration of a subject using a single sensor in order to reduce the number of sensors as much as possible. For this reason, first, it is possible to determine how much the fluctuation of the electroencephalogram power (square value) differs between a normal person and an Alzheimer's disease patient, and based on this, it can be determined whether the subject is an Alzheimer's disease patient or a normal person I paid attention to it. However, although the power variance (fraction) itself depends on the electroencephalogram amplitude, the amplitude itself is not directly related to the deterioration of the brain function, so normalization was performed in order to obtain a relative change.

  That is, as shown in the following formula (1), a variance PV of power of an electroencephalogram amplitude within a certain time is first calculated for one sensor.

そして、下記の式(2)に示すように、このパワーバリアンスPVを、そのパワーの平均値の二乗値で規格化したパワーバリアンスNPVを求めるものである。

Then, as shown in the following equation (2), the power variance NPV obtained by normalizing the power variance PV by the square value of the average value of the power is obtained.

そして、このような規格化パワーバリアンスNPVを、一個のセンサを頭部に装着して求める。この一個のセンサを装着する箇所の候補は、図1に示すような、例えば21箇所(International １0-２0 standardによって決められた位置)である。

Such a normalized power variance NPV is obtained by mounting one sensor on the head. For example, 21 positions (positions determined by the International 10-20 standard) as shown in FIG.

  A single sensor is sequentially attached at each of these 21 locations, and the attachment target is first a group of normal subjects and Alzheimer's disease patients. The group of these normal persons and Alzheimer's disease patients (hereinafter sometimes referred to as AD) is assumed to be known in advance as a group of persons determined by a known method as described above, for example. For each of the above, one sensor is sequentially attached at the position shown in FIG. 1, and the normalized value of the power variance expressed by the above equation (2) is obtained.

  Then, the distribution (ratio) of the normalized power variance NPV is obtained for each group of normal subjects and Alzheimer's disease patients.

  FIG. 2 shows probability density distributions A and B for NPV for normal and Alzheimer's disease patients. For distribution A, the area on the left side of the given NPV indicated by the dashed line C is defined as specificity, and for distribution B, the area on the right side of the given NPV is defined as sensitivity.

3, portions 2 ₁ showing a single sensor in FIG. 1 Normal's specificity curves A and Alzheimer's disease when mounted (hereinafter, this sensor itself. That if there is referred to as a sensor 2 _1, etc.) The sensitivity curve B of a patient is shown.

  Here, the “specificity” of the normality characteristic curve A of the normal person shown by a solid line is 82% when NPV = 2.5, for example, and this is in the region of NPV ≦ 2.5, for example, 100 normal person population It shows that 82 people belong to. The “sensitivity” of the sensitivity characteristic curve B of Alzheimer's disease patients is, for example, 77% in the case of NPV = 2.2, and 77 of 100 AD patients are in the region of NPV ≧ 2.2, for example. It shows that it belongs.

Moreover, refers to an NPV of sensitivity and specificity are equal to the cutoff value, cutoff NPV _CO = 2.28 next in the case of FIG. 3 (and FIG. 24), NPV and the value at the boundary is, than NPV _CO Alzheimer's disease can be determined as large if it is large, and it can be determined as normal if it is smaller than NPV _CO , but the discrimination ability at this time is 65%. In addition, two thresholds are introduced for NPV in order to improve discrimination ability. For example, when NPV> 2.58, the subject is determined to have Alzheimer's disease. At that time, since the specificity is 90% or more, 10% of normal subjects may be erroneously determined to have Alzheimer's disease.

  On the other hand, when NPV <2.08, the subject is determined to be normal, but because the sensitivity is 10% or less, 10% of Alzheimer's disease patients may be erroneously determined to be normal. In the case of 2.08 <NPV <2.58, the determination is not made, but the electroencephalogram is measured again several months later, and it is determined whether or not it is normal by increasing or decreasing the NPV.

Figure 4 shows the sensitivity characteristic B of a normal person singular distribution curves A and Alzheimer's disease patients were obtained in the same manner as FIG. 3 by the sensor 2 ₂ shown in FIG. Similarly, FIGS. 5 to 23 respectively show the sensitivity characteristic B of each patient normals specificity characteristics A and Alzheimer's disease in a sensor 2 _{3-2 21.}

  Thus, as shown in the characteristic curve B of FIG. 3 to FIG. 23, the worse the condition in Alzheimer's disease patients, the greater the value of normalized power variance NPV. Decrease.

This, referring to FIGS 23 for the other sensors 2 _{2-2 21,} sensitivity and specificity when the cutoff NPV _CO, the sensor 2 ₁ or 2 ₁₇ most% high, i.e., the head It is shown that by attaching one sensor to the forehead part of the part, it can be more clearly discriminated from other sensors whether it is an Alzheimer's disease patient or a normal person.

Therefore, in the present invention is obtained by a attaching a single sensor forehead, especially places 2 _1.

  In this way, the brain function deterioration degree estimating apparatus according to the present invention has a single sensor and the sensitivity obtained with respect to NPV obtained by attaching the sensor to the forehead of Alzheimer's disease patients and normal persons known in advance. And the NPV threshold value for Alzheimer's disease patients and normal persons obtained from the specificity, respectively, and the NPV value obtained by attaching the sensor to the subject's forehead and the NPV threshold value are compared. When the NPV threshold of specificity is exceeded, it is determined that the patient is an Alzheimer's disease, and when the sensitivity is below the NPV threshold of the sensitivity, a device that is normal, and a device that outputs the determination result, the NPV, The variance of the power of the electroencephalogram signal within a certain period of the output signal of the sensor converted into numerical data is normalized by the square value of the average value of the power.

  Note that the above-described determination device can perform a necessary calculation in a calculation center connected via a communication circuit, and send the obtained calculation result back to the output device via the communication line.

  Further, in the present invention, in order to automatically estimate the degree of brain function deterioration, it is obtained from the sensitivity and specificity obtained for NPV by attaching one sensor to the forehead of Alzheimer's disease patients and normal persons known in advance. NPV threshold values for Alzheimer's disease patients and normal subjects are stored, and the NPV threshold value obtained by attaching the sensor to the subject's forehead is compared with the NPV threshold value. There is provided a program for causing a computer to execute a procedure for determining that the patient is Alzheimer's disease when exceeding the threshold and determining that the subject is normal when the sensitivity is below the NPV threshold.

  Furthermore, the present invention also provides a computer-readable recording medium characterized by recording the above program.

  As described above, according to the present invention, by attaching a single sensor to the forehead part, it is possible to determine whether the subject is an Alzheimer's disease patient or a normal person. The degree can be estimated, and an easy-to-handle and inexpensive device can be realized.

  FIG. 25 shows an embodiment of the brain function deterioration degree estimating apparatus according to the present invention.

Do this embodiment, first, measuring the potential based on the brain neuronal activity by attaching to the forehead portion 2 ₁ shown in FIG. 1 of the head 1 such as EEG sensors or MEG sensors 2 _1, or, in advance Cover the cap with the sensor.

The measured potential from the sensor 2 ₁ is supplied to an analog / digital converter (A / D) 5 via an amplifier 3, the digitized measured potential (EEG) data through an input interface (I / F) 15 It is supplied to the computer 10. The input interface 15 can perform the following processing by extracting only the component having the frequency band (for example, beta wave) designated in advance by performing digital filtering.

  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 medium that stores arithmetic programs and the like, and the RAM 14 is a memory that stores brain wave data from the digitizer 23, the keyboard 24, and the A / D converter 5.

  As shown in FIG. 26, the electroencephalogram data is sent from the interface 17 of the computer 10 operating as a data transfer terminal device only to this case to the calculation center 42 as an arithmetic device via a communication line 41 such as the Internet. The result analyzed here may be sent back to the computer 10 at the clinical site again via the communication line 41 and output from an output device such as the CRT 31 or the printer 32 so that the doctor can use it as a diagnostic material. In this case, the program and its recording medium are installed in the calculation center.

  Further, the input interface 15 is connected to an external storage device 25 storing the characteristic data of the graph shown in FIG. 24, and the output interface 16 is a display device 31 such as a CRT for displaying the calculation result of the computer 10 and the display device. A printer 32 that stores data and waveforms displayed on 31 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.

  The operation of the present embodiment in the above configuration will be described below with reference to the flowchart shown in FIG.

First, for example, a group of 100 normal people and a group of 100 Alzheimer's disease patients are obtained in advance based on the existing method as described above. Then, attach the sensor 2 ₁ (forehead) shown in FIG. 1 each person of the normal person group and Alzheimer's disease patient populations. Then, in each person, the normalized power variance NPV from the output of the sensor 2 ₁ in a certain period for example 120 seconds is calculated based on the above equation (2) (Step S1).

  With respect to the normalized power variance NPV calculated in step S1, the characteristics of the normal population, that is, the specificity characteristics are obtained, and similarly, the characteristics of the Alzheimer's disease patient group, that is, the sensitivity characteristics are obtained (step S2). This is a characteristic curve graph similar to that shown in FIG. 3 or FIG. 24 (1).

Then, as shown in FIG. 24 (1), threshold values NPV _L = 2.08 and NPV _U = 2.58 corresponding to a desired value = 90% are set as an example for characteristic curves A and B (step S3).

In this way, the threshold NPV _L and NPV _U about normal person group and Alzheimer's patient population is set, similarly attach the sensor 2 ₁ to the subject, as well as calculate the normalized power variance NPV in 120 seconds (Step S4).

Since the specificity when NPV = NPV _U is 90% as shown in FIG. 24 (1), only 10% of normal subjects belong to the region of NPV> NPV _U (step S5) ((2 (See the shaded area to the right of).) Therefore, it is determined that the subject in this area has Alzheimer's disease (step S6), and at this time, the possibility of misdiagnosis of Alzheimer's disease for 10% of normal subjects remains.

On the other hand, when NPV = NPV _L , the sensitivity is 90% as shown in the figure (1). Therefore, only 10% of Alzheimer's disease patients are included in the region of NPV <NPV _L (step S7). (Refer to the hatched portion on the left side of FIG. 2). This subject is determined to be a normal person (step S8). At this time, only 10% of Alzheimer's disease patients are misdiagnosed as normal persons.

When it is found that it is between the threshold values NPV _L and NPV _U in FIG. 24 via step S5 and step S7, it is uncertain whether it is determined to be a normal person or an Alzheimer's disease patient. A re-inspection is performed (step S9).

The above thresholds NPV _L and NPV _U improve the discriminating ability by providing an indeterminate region. However, when the NPV cut-off value NPV _CO is used, the discriminating ability is 65%. One threshold may be used.

In the brain function degradation level estimation apparatus according to the present invention, it is a plan view showing a measurement position set to find out where to install a single sensor on the head. It is the figure which showed the specificity characteristic curve of the normal person with respect to normalized power variance NPV, and the sensitivity characteristic of an Alzheimer's disease patient. Is a graph showing the NPV of specificity characteristics and sensitivity characteristics of Alzheimer's disease patients was measured using a sensor 2 _1. Is a graph showing the NPV of specificity characteristics and sensitivity characteristics of Alzheimer's disease patients was measured using a sensor 2 _2. Sensor 2 ₃ is a graph showing the specificity characteristics and sensitivity characteristics of Alzheimer's disease patients NPV measured using. Sensor 2 ₄ is a graph showing the specificity characteristics and sensitivity characteristics of Alzheimer's disease patients NPV measured using. FIG. 5 is a graph showing the specificity characteristics of NPV measured using sensor ₂₅ and the sensitivity characteristics of Alzheimer's disease patients. ₆ is a graph showing the specificity characteristics of NPV and the sensitivity characteristics of Alzheimer's disease patients measured using sensor 26. FIG. Is a graph showing the NPV of specificity characteristics and sensitivity characteristics of Alzheimer's disease patients was measured using the sensor 2 _7. Is a graph showing the NPV of specificity characteristics and sensitivity characteristics of Alzheimer's disease patients was measured using a sensor 2 _8. Is a graph showing the NPV of specificity characteristics and sensitivity characteristics of Alzheimer's disease patients was measured using a sensor 2 _9. Is a graph showing the NPV of specificity characteristics and sensitivity characteristics of Alzheimer's disease patients was measured using a sensor 2 _10. Sensor 2 ₁₁ is a graph showing the specificity characteristics and sensitivity characteristics of Alzheimer's disease patients NPV measured using. Is a graph showing the NPV of specificity characteristics and sensitivity characteristics of Alzheimer's disease patients was measured using a sensor 2 _12. Sensor 2 ₁₃ is a graph showing the specificity characteristics and sensitivity characteristics of Alzheimer's disease patients NPV measured using. Sensor 2 ₁₄ is a graph showing the specificity characteristics and sensitivity characteristics of Alzheimer's disease patients NPV measured using. It is a graph showing the NPV of specificity characteristics and sensitivity characteristics of Alzheimer's disease patients was measured using the sensor _215. Is a graph showing the NPV of specificity characteristics and sensitivity characteristics of Alzheimer's disease patients was measured using a sensor 2 _16. Is a graph showing the NPV of specificity characteristics and sensitivity characteristics of Alzheimer's disease patients was measured using a sensor 2 _17. Sensor 2 ₁₈ is a graph showing the specificity characteristics and sensitivity characteristics of Alzheimer's disease patients NPV measured using. Is a graph showing the NPV of specificity characteristics and sensitivity characteristics of Alzheimer's disease patients was measured using a sensor 2 _19. Is a graph showing the NPV of specificity characteristics and sensitivity characteristics of Alzheimer's disease patients was measured using a sensor 2 _20. Is a graph showing the NPV of specificity characteristics and sensitivity characteristics of Alzheimer's disease patients was measured using a sensor 2 _21. Is a diagram showing characteristics and the threshold in the case of adopting the sensor 2 _1. It is the block diagram which showed one Example of the brain function degradation degree estimation apparatus which concerns on this invention. It is the block diagram which showed the modification of the brain function degradation degree estimation apparatus which concerns on this invention. It is the flowchart figure which showed the process sequence of the arithmetic unit (determination apparatus) used for this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS ₁ Test subject's head 2,2 ₁ EEG (magnetoencephalogram) sensor 3 Amplifier 5 A / D converter 10 Computer 11 CPU
13 ROM (program recording medium)
14 RAM
15, 16, 17 Interface 24 Keyboard 25 External storage device 31 CRT
32 Printer 41 Communication line 42 Calculation center In the figure, the same reference numerals indicate the same or corresponding parts.

Claims (4)

One sensor,
Storing the NPV thresholds for Alzheimer's disease patients and normal persons respectively obtained from the sensitivity and specificity obtained for the NPV obtained by attaching the sensor to the forehead of Alzheimer's disease patients and normal persons known in advance; The NPV obtained by attaching the sensor to the forehead of the subject is compared with the NPV threshold, and when the subject's NPV exceeds the NPV threshold of the specificity, it is determined that the patient is Alzheimer's disease, and the sensitivity of the NPV A device that determines that the person is normal when the threshold value is below,
A device for outputting the determination result,
A brain function deterioration degree estimation apparatus, wherein the NPV is obtained by standardizing a power variance within a certain period of an output signal of the sensor converted into numerical data by a square value of an average value of the power. In claim 1,
Estimating the degree of brain function deterioration, wherein the determination device performs a necessary calculation in a calculation center connected via a communication circuit, and sends the obtained calculation result back to the output device via the communication line apparatus. To automatically estimate the degree of brain function deterioration,
Attach one sensor to the forehead of Alzheimer's disease patients and normal persons who have been known in advance and memorize the NPV thresholds for Alzheimer's disease patients and normal persons obtained from the sensitivity and specificity obtained for NPV, respectively. The NPV obtained by attaching the sensor to the forehead is compared with the NPV threshold, and when the subject's NPV exceeds the NPV threshold of the specificity, it is determined as an Alzheimer's disease patient, and the NPV threshold of the sensitivity is determined. A program that causes a computer to execute a procedure for determining a normal person when the number is below. A computer-readable recording medium in which the program according to claim 3 is recorded.

Images