JPH10262941A - Electroencephalogram processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To shorten the measuring time for removing an electrocardiogram waveform from an electroencephalogram waveform by detecting the appearing time point of the featured waveform of a pulse wave corresponding to the featured waveform of the electrocardiogram waveform, thereby facilitating the segment division of the electroencephalogram waveform for obtaining the estimated waveform of the electrocardiogram waveform. SOLUTION: An amplifier 3 amplifies the potential difference between an active electrode 1a and a reference electrode 1b and the potential of pulse wave sensors 2a and 2b and an analog/digital converter(A/D) converts it into digital signals to store in a storage part 5. A pulse wave extreme detecting part 6 directly detects the extreme of the pulse wave waveform with periodicity and the time point of appearing the extreme from the display picture of the pulse wave waveform. An electroencephalogram dividing start point setting part 7 sets the start point of the division of the electroencephalogram waveform, a noise contamination estimating part 8 successively adds segments obtained by dividing the electroencephalogram waveform to obtain an average to obtain an estimated waveform, and an electroencephalogram signal extracting part 9 subtracts the estimated waveform from the electroencephalogram waveform to extract an electroencephalogram signal removed or the contamination of noise.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an electroencephalogram processing apparatus, and more particularly to an electroencephalogram processing apparatus for removing noise due to an electrocardiographic waveform from a waveform of an electroencephalogram in which noise due to an electrocardiographic waveform is mixed.

[0002]

2. Description of the Related Art FIG. 9 is a block diagram showing an example of a conventional electroencephalogram processing apparatus.

A conventional brain wave processing apparatus is disclosed in
An electroencephalogram processing device disclosed in Japanese Patent No. 7824 is known.

[0004] As shown in FIG.
Two brain wave measurement electrodes 91a and 91b, a brain potential amplifier 93 that receives signals from the brain wave measurement electrodes 91a and 91b and amplifies a potential difference therebetween, and converts an analog signal from the brain potential amplifier 93 into a digital signal. An analog-to-digital converter 94 for converting and an analog-to-digital converter 9
Temporary storage unit 95 for storing the output signal of ECG 4 and two electrocardiogram electrodes 92a and 92b for detecting an electrocardiographic waveform.
And an ECG amplifier 96 for receiving signals from the ECG electrodes 92a and 92b and amplifying a potential difference therebetween,
A trigger signal generator 99 for generating a trigger signal from an output signal of the ECG amplifier 96; and a trigger signal generator 9
And an arithmetic unit 98 for estimating and subtracting an electrocardiographic waveform mixed as noise into the electroencephalogram waveform stored in the temporary storage unit 95 based on the signal from the temporary storage unit 95 and extracting a noise-free electroencephalogram waveform.

The electroencephalogram processing apparatus constructed as described above comprises electroencephalogram measurement electrodes 91a and 91b and an electrocardiogram electrode 92a.
And 92b simultaneously detect an electroencephalogram and an electrocardiographic waveform. The electroencephalogram is amplified by a brain potential amplifier 93, converted from an analog signal to a digital signal by an analog-to-digital converter 94, and stored in a temporary storage unit 95. On the other hand, the potential difference of the electrocardiogram waveform is amplified by an electrocardiogram amplifier 96, a trigger signal is generated by a trigger signal generation unit 99, and the trigger signal is sent to a temporary storage unit 95. As the trigger signal, for example, a signal relating to a point in time when a characteristic waveform such as an R-wave appears is generated and sent to the temporary storage unit 95.
The arithmetic unit 98 divides the waveform of the brain wave into individual segments, starting from a point in time when the trigger signal from the trigger signal generating unit 99 traces back for a predetermined time, and sequentially adds each segment to obtain an average value. The estimated waveform of the electrocardiogram waveform mixed in the electroencephalogram waveform was obtained, and the electrocardiogram waveform was removed by subtracting this estimated waveform from the original electroencephalogram waveform containing the electrocardiogram waveform while synchronizing the starting point. EEG waveforms are extracted.

[0006]

Since the conventional electroencephalogram processing device as described above must simultaneously detect an electrocardiographic waveform in addition to detecting an electroencephalogram, it is necessary to mount electrodes for detecting the electrocardiographic waveform. It is necessary, and therefore, the preparation for the examination is troublesome, and there is a problem that the subject is limited and the applicable range is limited.

[0007] An object of the present invention is to solve the above-mentioned problems of the conventional electroencephalogram processing apparatus, and to utilize a pulse train in which a sensor can be mounted more easily than an electrode for detecting an electrocardiographic waveform. By detecting the time when the characteristic waveform of the pulse corresponding to the characteristic waveform of the electrocardiographic waveform is detected, the segmentation of the electroencephalogram waveform for obtaining the estimated waveform of the electrocardiographic waveform is facilitated. An object of the present invention is to provide an electroencephalogram processing device capable of shortening a measurement time for removing a radio wave shape and facilitating electroencephalogram processing.

[0008]

A first electroencephalogram processing device according to the present invention comprises an electroencephalogram sensor having an active electrode and a reference electrode for detecting an electroencephalogram waveform as a potential difference between two electrodes, and a pulse from an earlobe or a fingertip. A pulse sensor for detecting a pulse waveform, an amplifier for amplifying a potential difference between the active electrode and the reference electrode of the electroencephalogram sensor and a potential of the pulse sensor, and converting an analog signal output from the amplifier into a digital signal. An analog-to-digital converter, a storage unit for storing an output signal of the analog-to-digital converter, and detecting, from the signal stored in the storage unit, an extreme value of a pulse waveform having periodicity and a time point at which the extreme value appears. A pulse wave extremum detecting unit, and an electroencephalogram division starting point setting unit that sets a point in time when the output signal of the pulse wave extremal value detecting unit goes back for a predetermined time as a starting point of the division of the electroencephalogram waveform, Estimating an electrocardiographic waveform mixed as noise into the electroencephalogram waveform by dividing the electroencephalogram waveform into individual segments from the start point set by the electroencephalogram division start point setting unit and sequentially adding the segments to obtain an average value A noise mixing estimating unit for obtaining a waveform, and an electroencephalogram signal extracting unit for extracting an electroencephalogram signal from which noise has been mixed by subtracting the estimated waveform obtained by the noise mixing estimating unit from the brain wave waveform by synchronizing the starting point. In particular, as the pulse wave extremum detection unit, a pulse wave waveform stored in the storage unit is displayed on a screen of a display device, and an extreme value of a pulse wave waveform having periodicity from the screen. Either has a waveform display unit that directly detects the time point when the extreme value appears, or a pulse waveform preprocessing unit that performs smoothing processing of the pulse waveform stored in the storage unit, Calculating a first-order difference value at each point in time from the signal subjected to the smoothing processing in the wave-form preprocessing unit, and when the first-order difference value is equal to a sufficiently small value or smaller than the predetermined value; Or a difference type pulse wave extremum automatic detection unit for detecting a time point when the extremum appears as an extremum, or before performing a smoothing process of a pulse waveform stored in the storage unit. A processing unit, a wavelet transform type pulse wave extremum for obtaining an extremum by performing a wavelet process on the signal subjected to the smoothing process in the pulse wave waveform preprocessing unit and detecting a time point at which the extremum appears. A pulse wave pre-processing unit having an automatic detection unit, or performing a smoothing process of a pulse waveform stored in the storage unit, and at each time from a signal smoothed by the pulse wave pre-processing unit. Calculating the second-order difference value and detecting the time point at which the inflection point appears as an inflection point when the second-order difference value is equal to a sufficiently small predetermined value or smaller than the predetermined value. And a pulse wave inflection point automatic detection unit.

A second electroencephalogram processing apparatus according to the present invention comprises an electroencephalogram sensor having an active electrode and a reference electrode for detecting an electroencephalogram waveform as a potential difference between two electrodes, and a pulse for detecting a membrane wave from an earlobe or a fingertip. A group sensor, an amplifier that amplifies the potential difference between the active electrode and the reference electrode of the brain wave sensor and the potential of the pulse sensor, and an analog-to-digital converter that converts an analog signal output from the amplifier into a digital signal. A storage unit for storing an output signal of the analog-to-digital converter, and a pulse wave extremum detection for detecting, from the signal stored in the storage unit, an extremum of a periodic membrane wave having a periodicity and a time when the extremum appears. A pulse wave amplitude value calculating unit that calculates a pulse wave amplitude value as a difference between adjacent extreme values detected by the pulse wave extreme value detecting unit, and a pulse wave amplitude value detecting unit that is detected by the pulse wave extreme value detecting unit. A determination unit having an addition segment determination unit that selects an extreme value whose pulse wave amplitude value calculated by the pulse wave amplitude value calculation unit from within the extreme value is within a predetermined range and the present time of the extreme value; An electroencephalogram division starting point setting section that sets a point in time that is retroactive by a predetermined time by the output signal from the determination section as a starting point of the electroencephalogram waveform division, and individually calculates the electroencephalogram waveform from the starting point set by the electroencephalogram division starting point setting section. And a noise mixing estimating unit for obtaining an estimated waveform of an electrocardiographic waveform mixed as noise in the brain wave waveform by sequentially adding the segments to obtain an average value. An EEG signal extraction unit for extracting an EEG signal in which noise is removed by subtracting the estimated waveform from the EEG waveform while synchronizing the starting point. In particular, as the pulse wave extremum detection unit, the pulse wave stored in the storage unit is displayed on a screen of a display device, and the extreme value of the pulse wave waveform having periodicity from the screen and the time when the extreme value appears Has a waveform display unit that directly detects, or a pulse waveform pre-processing unit that performs smoothing processing of the pulse waveform stored in the storage unit, and each of the signals that have been smoothed by the pulse waveform pre-processing unit. Calculating a first-order difference value at a point in time, and when the first-order difference value is equal to a sufficiently small predetermined value or smaller than the predetermined value, the value is regarded as an extreme value and the extreme value appears Or a pulse type waveform pre-processing unit for performing a smoothing process of a pulse type waveform stored in the storage unit, and a pulse type waveform pre-processing unit for performing a smoothing process on the pulse type waveform stored in the storage unit. For the processed signal A wavelet conversion type pulse wave extremum automatic detection unit that determines the extremum by performing a wavelet process and detects the time when the extremum appears, or of the pulse waveform stored in the storage unit. A pulse waveform pre-processing unit that performs a smoothing process, and calculates a second-order difference value at each time from the signal that has been subjected to the smoothing process in the pulse waveform pre-processing unit, and the second-order difference value is a predetermined value having a sufficiently small value. A pulse wave inflection point automatic detection unit that detects a point in time when the inflection point appears as an inflection point when the pulse wave extremum value is equal to or smaller than the predetermined value. It is.

[0010]

Next, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing a first embodiment of the present invention. FIG. 10 is a waveform diagram showing an example of actually measured brain wave data, electrocardiographic waveform data and pulse wave data, FIG. 11 is a waveform diagram showing one cycle of the pulse wave data in FIG. 10, and FIG. FIG. 13 is a waveform diagram showing an example of received pulse wave data, and FIG. 13 is a waveform diagram showing an example of a waveform change in pulse wave data affected by a mental state.

The embodiment of FIG. 1 shows an electroencephalogram sensor having an active electrode 1a and a reference electrode 1b for detecting an electroencephalogram waveform as a potential difference between two electrodes, a pulse wave waveform (fingertip pulse wave) attached to a fingertip. ), A pulse sensor 2b attached to an earlobe and detecting a pulse waveform (earlobe pulse wave), a potential difference between the active electrode 1a and the reference electrode 1b of the electroencephalogram sensor, and a pulse sensor. An amplifier (amplifier) 3 for amplifying the potential of 2a or 2b, an analog / digital converter (A / D) 4 for converting an analog signal output from the amplifier into a digital signal, and an output signal of the analog / digital converter 4 are stored. And a waveform for displaying the pulse waveform stored in the storage unit 5 on the screen of the display device, and directly detecting the extreme value of the periodic pulse waveform and the time when the extreme value appears from the screen. Display 6 A pulse wave extremum detecting unit 6 having an EEG, an electroencephalogram division starting point setting unit 7 for setting a point in time when the output signal of the pulse wave extremal value detecting unit 6 goes back for a predetermined time as a starting point of the electroencephalogram waveform division, A noise for obtaining an estimated electrocardiographic waveform mixed as a noise in an electroencephalogram waveform by dividing the electroencephalogram waveform into individual segments from the starting point set by the setting unit 7 and sequentially adding the segments to obtain an average value. A brain wave signal extraction unit that extracts a brain wave signal from which the noise is removed by subtracting the estimated waveform obtained by the noise contamination estimation unit from the brain wave waveform by synchronizing the above-mentioned starting point; ing.

Next, the operation of the electroencephalogram processing apparatus configured as described above will be described.

This embodiment relates to three types of data, ie, electroencephalogram data 15, electrocardiogram data 16, and pulse wave data 17 shown in FIG. EEG data 15 of these
Are the active and reference electrodes (eg, Ag / AgCl
An amplifier (amplifier) (for example, SYNAX1 manufactured by NEC Medical Systems Co., Ltd.)
200), and an electrocardiographic data 16 is applied to a set of three electrodes (chest V lead) using an amplifier (for example, a PO medical device manufactured by NEC Medical Systems Co., Ltd.) that amplifies the potential difference between the electrodes.
LYGRAPH360), and the pulse wave data 17 is
A pulse sensor for detecting pulse wave data as a potential difference (for example, a pulse wave pick-up manufactured by NEC Medical Systems Co., Ltd.) and an amplifier for amplifying the detected potential difference (for example, PO manufactured by NEC Medical Systems Co., Ltd.)
LYGRAPH360) is connected and measured, and the analog signal of the output of each amplifier is recorded by a recorder (for example, OMNICORDER 8M14 manufactured by NEC Medical Systems Co., Ltd.).
FIG. 7 is a diagram showing a waveform output to FIG. The pulse wave data 17 among these has the same periodicity as the electrocardiogram data 16 under the influence of the electrocardiogram data 16. In this one cycle (about 700 msec), there are several extreme values such as the maximum values 17a and 17c and the minimum values 17b and 17d. When the pulse wave data 17 is compared with the electrocardiogram data 16,
For example, the interval between the present time of the peak value 17a of the pulse wave data 17 and the present time of the R wave 16a of the electrocardiographic data 16 is constant (about 340 msec). Therefore, if the present point of the maximum of the pulse wave data 17 is known, the extremum of the pulse wave data 17 is used, and the time point retroactive from the time point by a certain time is used as a starting point, without using the electrocardiogram data 16. The electroencephalogram data 15 can be divided into individual segments.

In the present embodiment, the electroencephalogram data 15 is measured by an electroencephalogram sensor having an active electrode 1a and a reference electrode 1b, and the pulse wave data 17 is measured by a pulse sensor 2a or a pulse sensor 2b. Potential difference between 1a and reference electrode 1b and pulse sensor 2a or 2b
Is amplified by an amplifier (amplifier) 3 and an analog signal output from the amplifier 3 is converted into an analog-to-digital converter (A).
/ D) 4 to convert the signal into a digital signal, and store the output signal of the analog-to-digital converter 4 in a storage unit 5 such as a hard disk of a personal computer.

The pulse wave extremum detection unit 6 displays the brain wave data and the pulse data stored in the storage unit 5 as analog waveforms on the screen of the waveform display unit 6a. The time when the maximum value 17a of the group data appears is read. The reading of the extremum of the pulse data at the present time can be performed by outputting the waveforms of the electroencephalogram data and the pulse data to the recorder as described above, and reading from the output recording paper.

An electroencephalogram division starting point setting section 7 sets a point in time when the waveform of the electroencephalogram data 15 is divided by a point in time retroactive by a predetermined time set by an output signal of the pulse wave extreme value detecting section 6. Specifically, the following data processing is performed.

That is, assuming that the detected extreme value present time is t _i ^* (i = 1, 2,..., L) (L is the number of detected extreme value present times), the electroencephalogram data 15 The starting point of the waveform division is t _i = t _i ^* + T _retro (i = 1, 2,..., L) (T _retro is the time to be _{traced back} ).

[0019] In the example of FIG. 10, when using the occurrence time of the maximum value 17a of the pulse wave data 17, the T _retro = T _{A over R} + T _R (T _{A over R} is R wave 16a of the electrocardiograph data 16 the difference between the appearance time of the maximum value 17a occurrence time and the pulse wave data 17, T _R is the time to be retroactively from the appearance point of the R-wave 16a of the electrocardiograph data 16). Specific numerical values of T _R, 200 meters
sec.

The noise contamination estimating section 8 has a starting point t _i (i = 1, 2,...) Set by the brain wave dividing starting point setting section 7.
L), the EEG data 17 is divided into individual segments,
The estimated waveform of electrocardiogram data mixed as noise into the electroencephalogram waveform is obtained by sequentially adding these segments to obtain an average value. Specifically, the following data processing is performed.

That is, assuming that the value of the digital data of the electroencephalogram at time t is y (t), the starting point t _i (i =
Each segment of the electroencephalogram data based on 1, 2,..., L) is represented by y _j ^* (τ + t _j ) = y (τ + t _j ) [u (τ) −u (τ
− (T _{j + 1} −t _j ))].

Where u (τ) is a step function,
[U (τ) −u (τ− (t _{j + 1} −t _j ))] is a rectangular wave type window function. Furthermore,

As a result, an estimated value of the mixed noise at the time point τ can be obtained. However, in the above equation, since it is an estimated value at time τ from each starting point, in order to return to the time t of the original brain wave data,

Is performed.

The electroencephalogram signal extraction unit 9 extracts an electroencephalogram signal by subtracting the estimated waveform obtained by the noise contamination estimation unit 8 from the original electroencephalogram data containing noise. That is,

Thus, an electroencephalogram signal is extracted.

FIG. 2 is a block diagram showing a second embodiment of the present invention.

In the embodiment of FIG. 2, instead of the pulse wave extremum detecting unit 6 of the embodiment of FIG. 1, a pulse wave pre-processing unit 36a for performing a smoothing process of the membrane wave stored in the storage unit 5; When the first-order difference value at each time is calculated from the signal subjected to the smoothing processing by the pulse waveform pre-processing unit 36a, and the first-order difference value is equal to or smaller than a sufficiently small predetermined value. And a pulse-wave extremal-value detection unit 36 having a differential-type pulse-wave extremum-automatic detection unit 36b that detects the time when the extremum appears as an extreme value. Other components are the same as those in the embodiment of FIG.

Next, regarding the operation of the electroencephalogram processing device configured as described above, the operation of the portion performing the same operation as the first embodiment will be omitted, and the operation of the portion different from the first embodiment will be described.

The pulse waveform preprocessing unit 36a performs smoothing processing of pulse data stored in the storage unit 5. There are various methods for the smoothing process. For example, the smoothing process can be performed by applying a low-pass digital filter to the pulse data or applying a moving average method.

The differential type pulse wave extremum automatic detection unit 36b calculates the first-order difference value at each time point from the signal smoothed by the pulse waveform pre-processing unit 36a, and the first-order difference value is a sufficiently small value. If the value is less than or equal to the specified value, the value is taken to the extreme value (maximum value or minimum value)
At the time when the extreme value appears. Specifically, the following data processing is performed.

That is, the digital data of the pulse train subjected to the smoothing process is defined as x _i, and the first-order difference value at the i-th time point is defined as x _i.

The definition is as follows. At this time, the fact that x _i is a local maximum value means that

By the way, the fact that x _i is a minimum value means that

Thus, each can be recognized.

For example, when applied to the waveform 17n of one cycle of the pulse wave data 17 in FIG. 11, the maximum value 17a and the minimum value 1
7b, the maximum value 17c, and the minimum value 17d can be sequentially detected.

FIG. 3 is a block diagram showing a third embodiment of the present invention.

In the embodiment shown in FIG. 3, instead of the pulse wave extremum detection unit 6 of the embodiment shown in FIG. 1, a pulse waveform pre-processing unit 56a for performing a smoothing process of the pulse waveform stored in the storage unit 5, A wavelet transform type pulse wave extremum automatic detection unit 56b which obtains an extremum by performing a wavelet process on the signal smoothed by the pulse waveform pre-processing unit 56a and detects a time when the extremum appears A pulse wave extremum detection unit 56 having the following is provided. Other components are the same as those in the embodiment of FIG.

Next, as for the operation of the electroencephalogram processing device configured as described above, the operation of a portion different from that of the first embodiment will be described.

The pulse waveform pre-processing unit 56a performs smoothing processing of the pulse waveform stored in the storage unit 5 in FIG. 2, as in the second embodiment of the present invention.

The wavelet conversion type pulse wave extremum automatic detection unit 56b includes a pulse wave pre-processing unit 56a.
By performing a wavelet conversion process on the signal subjected to the smoothing process in step (1), a further clarified extremum and a time point at which the extremum appears are detected. Specifically, the following data processing is performed.

Wavelet of arbitrary real-valued function f (t)
The t-transform is generally

However, ψ indicates a mother wavelet (Mot
Her Wavelet is a general term for various functions representing “waves” that are localized in space and time, and various functions include a Haar Wavelet.
et), Gabor Wave
elet), Malver Wavelet (Malver)
Wavelet), Moray Wavelet (Moll)
eye Wavelet), Mexican Wavelet (Mexican Wavelet), French Wavelet (French Wavelet), Shannon Wavelet (Shannon Wavelet),
Meyer Wavelet
t), Dove cheese wavelet (Daubechi)
es Wavelet), Simlet Wavelet, Corefret Wavelet, Cosplet Wavelet, Spline Wavelet
t) and the like are known. A and b are arbitrary real numbers, and along with two integers j and k,

## EQU3 ## By putting, the discrete expression of the function f (t) is

Can be obtained. Furthermore,

If the signal f (t) is regarded as f ₀ (t) at this time, the above equation becomes:

Can be written as

The wavelet transform process is used for various purposes such as periodic signal decomposition, pulse decomposition, noise decomposition, abnormal wave detection, and peak detection.
In the present embodiment, for example, the fourth floor B-spline wavelet (SplineWavelet) may be used as a mother wavelet (Mother Wavelet), and f _-2 (t) may be obtained for the pulse waveform data f (t). Further, from the waveform converted in this way, the extreme value and the time when the extreme value appears can be detected.

FIG. 4 is a block diagram showing a fourth embodiment of the present invention.

In the embodiment of FIG. 4, instead of the pulse wave extremum detection unit 6 of the embodiment of FIG. 1, a pulse wave pre-processing unit 76a for performing a smoothing process of the membrane wave stored in the storage unit 5; When the second-order difference value at each time point is calculated from the signal subjected to the smoothing process in the pulse waveform pre-processing unit 76a, and the second-order difference value is equal to or smaller than a sufficiently small predetermined value. Is provided as an inflection point, and a pulse wave inflection point automatic detection unit 76b for detecting a time point at which the inflection point appears is provided. Other components are the same as those in the embodiment of FIG.

Next, the operation of the electroencephalogram processing device configured as described above will be described, focusing on the operation of the portion different from that of the first embodiment.

The pulse waveform pre-processing unit 76a performs the smoothing process on the membrane waveform stored in the storage unit 5, as in the second embodiment.

The pulse wave inflection point automatic detecting section 76b calculates the second-order difference value at each time from the signal smoothed by the pulse waveform pre-processing section 76a, and the second-order difference value is a sufficiently small value. The time when the inflection point appears is detected by using the case where the value is equal to or smaller than the predetermined value as the inflection point. Specifically, the following data processing is performed.

That is, the digital data of the pulse train subjected to the smoothing process is denoted by x _i, and the second-order (retreat) difference value at the i-th time point

Is calculated. For example, in the case of the pulse wave data 17 in FIG. 11, four inflection points 17e, 17f, 17g,
17h, and the inflection points 17e and 17g are

The inflection points 17f and 17h are

Can be detected.

In this embodiment, for example, the retroactive time when using the inflection point 17e is different from the retroactive time when using the maximum value 17a in the second embodiment.

FIG. 5 is a block diagram showing a fifth embodiment of the present invention.

The embodiment shown in FIG. 5 is arranged between the pulse wave extremum detection unit 6 and the electroencephalogram division starting point setting unit 7 of the embodiment shown in FIG. A pulse wave amplitude value calculating unit 10a for calculating a pulse wave amplitude value as a difference between the pulse wave amplitude value calculating unit 10a and a pulse wave amplitude calculated by the pulse wave amplitude value calculating unit 10a from the extreme values detected by the pulse wave extreme value detecting unit 6. A judgment unit 10 having an extreme value whose value is within a predetermined range and an addition segment judging unit 10b for selecting the present time is provided. Other components are the same as those in the embodiment of FIG. is there.

Next, regarding the operation of the electroencephalogram processing device configured as described above, the operation of the portion different from that of the first embodiment will be described.

The pulse wave amplitude value calculating section 10a calculates a pulse wave amplitude value as a difference between adjacent extreme values detected by the pulse wave extreme value detecting section 6.

The addition segment judging section 10b determines, from among the extreme values detected by the pulse wave extreme value detecting section 6, an extreme value whose pulse wave amplitude value calculated by the pulse wave amplitude value calculating section 10a is within a predetermined range. And its current time to elect. Specifically, the following data processing is performed.

That is, first, the calculation of the pulse wave amplitude value is performed as follows.

One of the pulse wave data 17 as shown in FIG.
The maximum value and the minimum value of the pulse wave potential value are searched within the time width (window width) of the period 17n, and the difference between them is defined as the pulse wave amplitude value. In the case of FIG. 11, the pulse wave amplitude value = | x (t _A ) −x (t _D ) |. However, t _A is the time when the maximum value 17a appears, t _D
Is the time when the minimum value 17d appears.

The pulse wave amplitude value calculated in this way is affected. That is, as shown in FIG. 12, when the mental state is nervous, the amplitude decreasing section 1 in FIG.
As shown in FIG. 8A, when the amplitude value decreases and the mental state is relaxing, it is well known that the amplitude value increases as shown in an amplitude gradual increase section 18b of FIG. 12B.
The influence of such a psychological state may be reflected on the electrocardiogram data, and such a fluctuation of the electrocardiogram data may cause a fluctuation of the starting point of the division of the pulse wave data. Therefore, it is necessary to set the starting point of the division of the pulse wave data based on the data in the stable mental state. In order to detect the stable mental state, the pulse wave amplitude value is obtained as follows.

FIG. 13 is a waveform diagram showing pulse wave data when the mental state is changing. In FIG. 13, when the processing in the pulse wave extremum detection unit 6 is applied to the pulse wave data 20, for example, the time points t ₁ , t ₂ ,
.., T ₁₆ are detected. At these moments, if the pulse wave amplitude value is set to be equal to or greater than the predetermined value ε in a stable mental state, t ₁ = t ₁ ⁺ , t ₂ = t ₂ ⁺ , t ₃ =
^{_{t 3 +, t 4 = t}} 4 +, t 5 = t 5 +, t 13 = t 6 +, t 14 = t
₇ ⁺ , t ₁₅ = t ₈ ⁺ and t ₁₆ = t ₉ ⁺ can be selected.

FIG. 6 is a block diagram showing a sixth embodiment of the present invention.

In the embodiment of FIG. 6, instead of the pulse wave extremum detection unit 6 of the embodiment of FIG. 5, a pulse wave pre-processing unit 36a for performing a smoothing process of the membrane wave stored in the storage unit 5, When the first-order difference value at each time is calculated from the signal subjected to the smoothing processing by the pulse waveform pre-processing unit 36a, and the first-order difference value is equal to or smaller than a sufficiently small predetermined value. And a pulse-wave extremal-value detection unit 36 having a differential-type pulse-wave extremum-automatic detection unit 36b that detects the time when the extremum appears as an extreme value. Other components are the same as those in the embodiment of FIG. Therefore, the operation of the present embodiment is different from the operation of the pulse wave extremum detector 6 of the operation of the embodiment of FIG. 5 in that the pulse wave pre-processing unit 36a of the embodiment of FIG. This is a replacement of the operation with the detection unit 36b.

FIG. 7 is a block diagram showing a seventh embodiment of the present invention.

In the embodiment of FIG. 7, instead of the pulse wave extremum detection unit 6 of the embodiment of FIG. 5, a pulse wave pre-processing unit 56a for performing smoothing processing of the membrane wave stored in the storage unit 5, Wavelet conversion type pulse wave extremum automatic detection unit 56b for obtaining an extremum by performing wavelet processing on the signal smoothed by the pulse waveform preprocessing unit 56a and detecting the time when the extremum appears And a pulse wave extremum detection unit 56 having the following. Other components are the same as those in the embodiment of FIG. Therefore, the operation of the present embodiment is different from the operation of the pulse wave extremum detection unit 6 of the operation of the embodiment of FIG. 5 in that the pulse wave pre-processing unit 56a of the embodiment of FIG. This is a replacement of the operation with the automatic value detection unit 56b.

FIG. 8 is a block diagram showing an eighth embodiment of the present invention.

In the embodiment of FIG. 8, instead of the pulse wave extremum detection unit 6 of the embodiment of FIG. 5, a pulse wave pre-processing unit 76a for performing a smoothing process of the membrane wave stored in the storage unit 5, When the second-order difference value at each time point is calculated from the signal subjected to the smoothing process in the pulse waveform pre-processing unit 76a, and the second-order difference value is equal to or smaller than a sufficiently small predetermined value. Is provided as an inflection point, and a pulse wave inflection point automatic detection unit 76b for detecting a time point at which the inflection point appears is provided. Other components are the same as those in the embodiment of FIG. Therefore, the operation of the present embodiment is different from the operation of the pulse wave extremum detecting unit 6 of the operation of the embodiment of FIG. 5 in that the pulse waveform preprocessing unit 76a of the embodiment of FIG.
And the operation of the pulse wave inflection point automatic detection unit 76b is replaced.

[0074]

As described above, the electroencephalogram processing device according to the present invention includes an electroencephalogram sensor having an active electrode and a reference electrode for detecting an electroencephalogram waveform as a potential difference between two electrodes, and a pulse wave from an earlobe or a fingertip. A pulse sensor that detects a waveform, an amplifier that amplifies the potential difference between the active electrode and the reference electrode of the electroencephalogram sensor and the potential of the pulse sensor, and an analog-to-digital converter that converts an analog signal output from the amplifier into a digital signal A storage unit for storing an output signal of the analog-to-digital converter, and a pulse wave extremum detection unit for detecting, from the signal stored in the storage unit, an extremum of a pulse waveform having periodicity and a time point at which the extremum appears A brain wave division starting point setting unit that sets a point in time that has been traced back by a predetermined time by an output signal of the pulse wave extreme value detection unit as a starting point of the division of the electroencephalogram waveform, and a brain wave division starting point setting unit. A noise mixing estimator that divides an electroencephalogram waveform into individual segments from the starting point and sequentially adds the segments to obtain an average value, thereby obtaining an estimated waveform of an electrocardiogram waveform mixed as noise into the electroencephalogram waveform; By attaching an electrode for detecting an electrocardiographic waveform by providing an electroencephalogram signal extraction unit for extracting an electroencephalogram signal from which noise has been removed by synchronizing the starting point of the estimated waveform obtained by the estimating unit and subtracting the estimated waveform from the electroencephalogram waveform, Estimation of the electrocardiographic waveform because it is possible to use the pulse that is easier to attach the sensor than to detect the characteristic waveform of the pulse corresponding to the characteristic waveform of the electrocardiographic waveform. The effect is that the segmentation of the electroencephalogram waveform for obtaining the waveform can be easily performed, and the measurement time for removing the electrocardiogram waveform from the electroencephalogram waveform can be shortened to facilitate the electroencephalogram processing.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a first embodiment of the present invention.

FIG. 2 is a block diagram showing a second embodiment of the present invention.

FIG. 3 is a block diagram showing a third embodiment of the present invention.

FIG. 4 is a block diagram showing a fourth embodiment of the present invention.

FIG. 5 is a block diagram showing a fifth embodiment of the present invention.

FIG. 6 is a block diagram showing a sixth embodiment of the present invention.

FIG. 7 is a block diagram showing a seventh embodiment of the present invention.

FIG. 8 is a block diagram showing an eighth embodiment of the present invention.

FIG. 9 is a block diagram illustrating an example of a conventional electroencephalogram processing device.

FIG. 10 is a waveform chart showing an example of actually measured brain waves, electrocardiographic waveforms, and pulse waves.

FIG. 11 is a waveform diagram showing one cycle of the pulse wave of FIG. 10;

FIG. 12 is a waveform chart showing an example of actual measurement data of a pulse wave affected by a mental state.

FIG. 13 is a waveform chart showing an example of waveform fluctuation in pulse wave data affected by a mental state.

[Explanation of symbols]

 1a Active electrode 1b Reference electrode 2a ・ 2b Pulse sensor 3 Amplifier (Amplifier) 4 Analog / Digital converter (A / D) 5 Storage unit 6.36.56.76 Pulse wave extremum detection unit 6a Waveform display unit 7 EEG division Starting point setting unit 8 Noise mixing estimation unit 9 EEG signal extraction unit 10 Judgment unit 10a Pulse wave amplitude value calculation unit 10b Addition segment judgment unit 15 EEG data 16 Electrocardiogram data 17 Pulse wave data 17a / 17c Maximum value 17b / 17d Minimum value 17e 17h Inflection point 18a Amplitude decreasing section 18b Amplitude increasing section 20 Pulse wave data 36a / 56a. 76a pulse wave pre-processing unit 36b differential type pulse wave extreme value automatic detection unit 56b wavelet conversion type pulse wave extreme value automatic detection unit 76b pulse wave inflection point automatic detection unit 91a / 91b EEG measurement electrodes 92a / 92b for electrocardiogram Electrode 93 Brain potential amplifier 94 Analog-to-digital converter 95 Temporary storage unit 96 ECG amplifier 98 Operation unit 99 Trigger signal generation unit

Claims (10)

[Claims] 1. An electroencephalogram sensor having an active electrode and a reference electrode for detecting an electroencephalogram waveform as a potential difference between two poles, a pulse sensor for detecting a pulse waveform from an earlobe or a fingertip, and the electroencephalogram sensor. An amplifier for amplifying a potential difference between an active electrode and the reference electrode and a potential of the pulse sensor, an analog-to-digital converter for converting an analog signal output from the amplifier to a digital signal, and an output signal of the analog-to-digital converter A pulse wave extremum detection unit that detects an extreme value of a periodic membranous waveform having a periodicity from the signal stored in the storage unit and a time point at which the extremum appears, and a pulse wave extremum detection An electroencephalogram division starting point setting unit that sets a point in time that has been traced back by a predetermined time by the output signal of the unit as the starting point of the division of the electroencephalogram waveform; A noise mixing estimating unit for obtaining an estimated waveform of an electrocardiographic waveform mixed as noise into the brain wave waveform by dividing the brain wave waveform into individual segments and sequentially adding the segments to obtain an average value; and An electroencephalogram processing device comprising: an electroencephalogram signal extraction unit configured to extract an electroencephalogram signal from which noise is removed by synchronizing the estimated waveform obtained by the contamination estimation unit with the start point and subtracting the estimated waveform from the electroencephalogram waveform. . 2. A waveform display for displaying a pulse waveform stored in the storage unit on a screen of a display device, and directly detecting an extreme value of the periodic pulse waveform and a time point at which the extreme value appears from the screen. The electroencephalogram processing device according to claim 1, further comprising the pulse wave extremum detection unit having a unit. 3. A pulse waveform preprocessing unit for performing a smoothing process on the pulse waveform stored in the storage unit, and calculating a first-order difference value at each time from the signal smoothed by the pulse waveform preprocessing unit. When the first-order difference value is equal to a sufficiently small predetermined value or smaller than the predetermined value, the difference type pulse wave pole detecting the time when the extreme value appears as the extreme value The electroencephalogram processing apparatus according to claim 1, further comprising the pulse wave extremum detection unit having an automatic value detection unit. 4. A pulse wave pre-processing unit for performing a smoothing process on a pulse waveform stored in the storage unit, and performing a wavelet process on the signal smoothed by the pulse wave pre-processing unit. The electroencephalogram processing according to claim 1, further comprising the pulse wave extremum detection unit having a wavelet transformation type pulse wave extremum automatic detection unit that obtains an extremum and detects a time when the extremum appears. apparatus. 5. A pulse waveform pre-processing unit for performing a smoothing process on a pulse waveform stored in the storage unit, and calculating a second-order difference value at each time from a signal smoothed by the pulse waveform pre-processing unit. A pulse wave inflection point automatic detecting the time when the inflection point appears as an inflection point when the second-order difference value is equal to a sufficiently small predetermined value or smaller than the predetermined value. The electroencephalogram processing device according to claim 1, further comprising the pulse wave extremum detection unit having a detection unit. 6. An electroencephalogram sensor having an active electrode and a reference electrode for detecting an electroencephalogram waveform as a potential difference between two poles, a pulse wave sensor for detecting a pulse wave waveform from an earlobe or a fingertip, and the electroencephalogram sensor. An amplifier for amplifying a potential difference between an active electrode and the reference electrode and a potential of the pulse sensor, an analog-to-digital converter for converting an analog signal output from the amplifier to a digital signal, and an output signal of the analog-to-digital converter A pulse wave extremum detection unit that detects an extreme value of a periodic membranous waveform having a periodicity from the signal stored in the storage unit and a time point at which the extremum appears, and a pulse wave extremum detection A pulse wave amplitude value calculating unit that calculates a pulse wave amplitude value as a difference between adjacent extreme values detected by the unit, and the pulse wave amplitude from the extreme values detected by the pulse wave extreme value detecting unit. A determination unit having an addition segment determination unit that selects the extreme value and the present time when the pulse wave amplitude value calculated by the value calculation unit is within a predetermined range,
An electroencephalogram division starting point setting section that sets a point in time that is retroactive by a predetermined time by the output signal from the determination section as a starting point of the electroencephalogram waveform division, and individually calculates the electroencephalogram waveform from the starting point set by the electroencephalogram division starting point setting section. And a noise mixing estimating unit for obtaining an estimated waveform of an electrocardiographic waveform mixed as noise in the brain wave waveform by sequentially adding the segments to obtain an average value. An electroencephalogram signal processing unit, which extracts an electroencephalogram signal from which the noise is removed by subtracting the estimated waveform from the electroencephalogram waveform by synchronizing the starting point. 7. A waveform display for displaying a pulse waveform stored in the storage unit on a screen of a display device, and directly detecting an extreme value of the periodic pulse waveform and a time when the extreme value appears from the screen. The electroencephalogram processing device according to claim 6, further comprising the pulse wave extremum detection unit having a unit. 8. A pulse waveform preprocessing unit for performing a smoothing process on the pulse waveform stored in the storage unit, and calculating a first-order difference value at each time from the signal smoothed by the pulse waveform preprocessing unit. When the first-order difference value is equal to a sufficiently small predetermined value or smaller than the predetermined value, the difference type pulse wave pole detecting the time when the extreme value appears as the extreme value The electroencephalogram processing device according to claim 6, further comprising the pulse wave extremum detection unit having an automatic value detection unit. 9. A pulse waveform pre-processing unit for performing a smoothing process on a pulse waveform stored in the storage unit, and performing a wavelet process on the signal smoothed by the pulse waveform pre-processing unit. 7. The electroencephalogram processing according to claim 6, further comprising the pulse wave extremum detection unit having a wavelet transformation type pulse wave extremum automatic detection unit for obtaining an extremum and detecting a time when the extremum appears. apparatus. 10. A pulse waveform pre-processing unit for performing a smoothing process of a pulse waveform stored in the storage unit, and calculating a second-order difference value at each time from a signal smoothed by the pulse waveform pre-processing unit. A pulse wave inflection point automatic detecting the time when the inflection point appears as an inflection point when the second-order difference value is equal to a sufficiently small predetermined value or smaller than the predetermined value. The electroencephalogram processing device according to claim 6, further comprising the pulse wave extremum detection unit having a detection unit.