JP6638946B2 - Brain wave signal processing device, brain wave signal processing method, program, and recording medium - Google Patents

Description

  The present invention relates to an electroencephalogram signal processing device for processing electroencephalogram signals and an electroencephalogram signal processing method. Further, the present invention relates to a program for causing a computer to function as such an electroencephalogram signal processing device, and a recording medium on which such a program is recorded.

  The activity of each part of the brain can be regarded as a vibrator. At this time, higher-order functions of the brain can be regarded as a phase synchronization phenomenon between these vibrators. This means that, for example, the subject alternately displays the first image in which the balls are arranged in the upper left and lower right of the screen and the second image in which the balls are arranged at the upper right and lower left of the screen as an animation. It can be confirmed by presenting (Non-Patent Document 1).

  In fact, when observing the brain waves of a subject who is looking at these images, the recognition that "the two balls arranged vertically are moving horizontally in opposite phase" occurs, and at the same time, the brain waves (indicating the activity of the right brain) It can be confirmed that phase synchronization occurs between an electroencephalogram (hereinafter, also referred to as “left brain wave”) that indicates the activity of the left brain and a right brain wave (hereinafter also referred to as “left brain wave”). This means that the ball arranged at the upper left of the screen in the first image (cognitive information processing by the right brain) and the ball arranged at the upper right of the screen in the second image (cognitive information processing by the left brain) There is no other proof that the higher-order function of the brain to identify them can be regarded as the phase synchronization phenomenon described above.

  The same can be confirmed by observing the brain waves of a stroke patient. That is, when observing the brain waves of a stroke patient having a disorder in higher brain functions, it can be seen that phase synchronization is less likely to occur between the right brain wave and the left brain wave as compared with the brain waves of a healthy person. In addition, research results have been reported that there is a significant correlation between the likelihood of occurrence of phase synchronization immediately after illness and the effectiveness of rehabilitation for right and left brain waves of stroke patients (Non-Patent Document 2). .

Michael Rose, 1 other, "Neural Coupling Binds Visual Tokens to Moving Stimuli", The Journal of Neuroscience, November 2, 2005, 25 (44), p.10101-10104 Wenqing Wu, 6 others, "Impaired neuronal synchrony after focalischemic stroke in elderly patients", Clinical Neurophysiology, ELSEVIER, published online June 29, 2010, 122 (2011), p.21-26

  The research results reported in Non-Patent Document 2 suggest that the rehabilitation effect for stroke patients can be enhanced by achieving a state in which phase synchronization between right and left brain waves is likely to occur.

  In addition, according to recent studies by the inventors of the present application, the degree of phase synchronization between electroencephalogram signals detected at different positions of the subject's head is determined by the FIM (Functional Independence Measure) of a brain disease patient such as stroke or cerebral infarction. ) And NIHSS (NIH stroke scale). That is, it is considered that the rehabilitation effect can also be obtained by bringing the state in which the phase synchronization between the electroencephalogram signals detected at different positions of the subject's head easily occurs.

  Furthermore, recent studies by the inventors of the present application have revealed that in a healthy person, information transmission in the brain is generally symmetric between the left hemisphere and the right hemisphere, whereas it is asymmetric in patients with brain disease. . In other words, it is considered that the rehabilitation effect can be obtained by bringing the state in which information transmission in the brain tends to be symmetrical.

  However, there is no known method by which a brain disease patient trains his or her brain to achieve such a state. Also, there is no known method for a practitioner to train the brain of a brain disease patient in order to realize such a state.

  The present invention has been made in view of the above problems, and an object of the present invention is to realize an electroencephalogram signal processing device and the like that can be used for training on how to use the brain to improve the condition of a brain disease patient. .

  The use of the present invention is not limited to enhancing the rehabilitation effect for a brain disease patient. For example, the present invention can be applied to a game or the like in which a higher order function of the brain for a healthy person or a degree of phase synchronization or symmetry of information transmission in the brain is controlled.

  In order to achieve the above object, another electroencephalogram signal processing device according to the present invention sequentially acquires and processes each electroencephalogram signal detected at a plurality of locations in each of the right and left hemispheres of the subject's head. An electroencephalogram signal processing device, which is sequentially calculated from each of the electroencephalogram signals and, based on a phase synchronization index indicating the degree of phase synchronization between the electroencephalogram signals, the detection point of each of the electroencephalogram signals in the brain of the subject. An importance index calculating means for sequentially calculating an importance index indicating relative importance in information transmission of the subject, and information transmission in the right hemisphere and information transmission in the left hemisphere from the distribution of the importance index in the head of the subject. A symmetry index calculating means for sequentially calculating a symmetry index indicating the degree of symmetry with, a signal according to the symmetry index sequentially calculated by the symmetry index calculating means, or the symmetry index calculation A signal generation for generating a signal corresponding to the time average of the symmetry index sequentially calculated in the step and sequentially supplying the signal to the stimulating device, thereby giving a stimulus which changes in conjunction with the symmetry index to the subject. Means.

  Further, in order to achieve the above object, another electroencephalogram signal processing method according to the present invention provides an electroencephalogram signal processing device in which each electroencephalogram signal detected at a plurality of locations in each of the right hemisphere and the left hemisphere of the subject's head. A brain wave signal processing method of sequentially acquiring and processing the brain wave signal processing apparatus, wherein the brain wave signal processing device is sequentially calculated from the brain wave signals, from a phase synchronization index indicating the degree of phase synchronization between the brain wave signals, An importance calculation step of sequentially calculating an importance index indicating a relative importance in information transmission in the brain of the subject at each detection point of the brain wave signal, and an electroencephalogram signal processing device, From the distribution of the importance index, a symmetry index calculating step of sequentially calculating a symmetry index indicating a degree of symmetry between information transmission in the right hemisphere and information transmission in the left hemisphere, and the electroencephalogram signal processing device, A signal corresponding to the symmetry index sequentially calculated in the symmetry index calculation step, or a signal corresponding to a time average of the symmetry index sequentially calculated in the symmetry index calculation step is generated and stimulated. A signal generating step of sequentially providing the subject with the stimulus that changes in synchronization with the symmetry index by supplying the stimulus to the subject.

  ADVANTAGE OF THE INVENTION According to this invention, a subject can train his / her brain so that phase synchronization between electroencephalogram signals easily occurs or information transmission in the brain is performed symmetrically in both hemispheres.

It is a block diagram showing composition of an electroencephalogram signal processing device concerning one embodiment of the present invention. FIG. 4 is a block diagram illustrating a modification of the electroencephalogram signal processing device illustrated in FIG. 1. FIG. 2 is a diagram showing contents of a video displayed by the electroencephalogram signal processing device shown in FIG. 1. FIG. 2 is a block diagram illustrating a configuration of a computer that functions as the electroencephalogram signal processing device illustrated in FIG. 1. FIG. 9 is a diagram showing the correlation between PSI and FIMt by plotting experimental results with the vertical axis being LHPS and the horizontal axis being FIMt. FIG. 9 is a diagram showing the correlation between PSI and NIHSS by plotting experimental results with the vertical axis being LHPS and the horizontal axis being NIHSS. It is a figure which investigated about several cerebral infarction patients and shows the change of FIMt and LHPS. FIG. 7 is a diagram showing a list of p-values showing a correlation between various synchronization indices and various disease state indices in the α band for a certain population (a population of cerebral infarction patients). FIG. 10 is a diagram showing a list of p-values showing correlations between various synchronization indices and various disease state indices in the α band for another population (a population of cerebral infarction patients). (A) is a diagram showing a list of p-values indicating the correlation between various synchronization indexes and various pathological indexes in the α band for left hemisphere lesion patients. (B) is a view showing a list of p-values showing correlations between various synchronization indexes and various pathological indexes in the α band for right hemisphere lesion patients. FIG. 9 is a view showing a list of p-values indicating correlations between various synchronization indices and various disease state indices in the β1 band for another population (cerebral infarction patient population). (A) is a diagram showing a list of p-values showing correlations between various synchronization indexes and various pathological indexes in the β1 band for patients with left hemisphere lesions. (B) is a diagram showing a list of p-values showing correlations between various synchronization indexes and various pathological indexes in the β1 band for right hemisphere lesion patients. (A) is a diagram showing a list of p-values indicating the significance of the partial correlation between the synchronization index RHPS, the pathological index FIMm, and the lesion volume LV, and FIG. It is a figure which showed the p value which shows the significance of the partial correlation between the parameter | index IHPS, the pathological index FIMm, and the lesion volume LV, and (c) is a figure which shows the synchronous index LHPS, the pathological index FIMm, and the lesion volume LV. It is the figure which showed the p value which shows the significance of the partial correlation between in a list, and (d) is a schematic diagram showing the relationship of synchronous index RHPS, IHPS, and LHPS, pathological index FIMm, and lesion volume LV. . (A) is a diagram showing a list of p-values indicating the significance of the partial correlation between the synchronization index LHPS, the pathological index FIMm, and the lesion volume LV, and FIG. It is the figure which showed the p value which shows the significance of the partial correlation between the parameter | index LHPS, the pathological index FIMc, and the lesion volume LV at a list, and (c) is the synchronous index LHPS, the pathological index FIMm, and FIMc, and the lesion index. It is a schematic diagram showing the relationship of volume LV. It is a flowchart which shows the flow of the process which the said electroencephalogram signal processing apparatus performs. FIG. 9 is a block diagram illustrating a main configuration of an electroencephalogram signal processing device according to another embodiment of the present invention. It is a figure showing a measurement position of an electroencephalogram signal in a head. It is the figure which showed the spatial distribution of the importance of a healthy person and three patients from which a pathological index differs. It is a figure showing distribution of the average value of importance of each subject. It is a figure which shows the distribution of the average value of the importance of several test subjects with a damage in the left hemisphere, and the distribution of the average value of the importance of several test subjects with a damage in the right hemisphere. The distribution of the average value of the importance of a plurality of subjects whose FIMt is particularly low among the subjects having damage in the left hemisphere, and the average value of the importance of the plurality of subjects whose FIMt is particularly low among the subjects having damage of the right hemisphere. FIG. The vertical axis represents the value obtained by averaging the left and right differences in importance between the electrodes at positions symmetrical to the median plane, and the horizontal axis represents FIMt. The experimental results are plotted to show the correlation between the symmetry index and FIM. FIG. It is the figure which plotted the experimental result by making the difference of importance between F4-F3 into a vertical axis | shaft, and setting a horizontal axis to FIMt. It is the figure which plotted the experimental result by making the difference of importance between C4 and C3 into a vertical axis | shaft, and setting a horizontal axis to FIMt. It is the figure which plotted the experimental result by making the inner product of an importance vector into a vertical axis | shaft, and setting a horizontal axis to FIMt. It is a flowchart which shows an example of the electroencephalogram signal processing method which the said electroencephalogram signal processing apparatus performs.

[Embodiment 1]
One embodiment of an electroencephalogram signal processing device according to the present invention will be described below with reference to the drawings. In the following description, two quantities being “same” means that the order of the two quantities is the same, in other words, the larger quantity is 10 times the smaller quantity. Indicates that:

[Configuration of EEG signal processing device]
The configuration of the electroencephalogram signal processing device 1 according to the present embodiment will be described with reference to FIG.

  FIG. 1 is a block diagram illustrating a configuration of the electroencephalogram signal processing device 1. The electroencephalogram signal processing device 1 performs LHPS (Left Hemispheric Phase Synchronization), RHPS (Right Hemispheric Phase Synchronization), or CPS (Central Phase Synchronization) of the subject. This is a device for feeding back to the subject, and as shown in FIG. 1, an electroencephalogram signal acquisition unit 11, a first instantaneous phase identification unit (instantaneous phase identification unit) 12A, and a second instantaneous phase identification unit (instantaneous phase identification unit) Means) 12B, a phase synchronization index calculation unit (phase synchronization index calculation unit) 13, and a video signal generation unit (signal generation unit) 14.

  The electroencephalogram signal acquisition unit 11 includes an electroencephalogram signal X1 representing an electroencephalogram measured at a first position of the subject's head, and an electroencephalogram representing an electroencephalogram measured at a second position different from the first position of the subject's head. This is a configuration for acquiring the signal X2 from an external device (such as an electroencephalograph) connected to the electroencephalogram signal processing device 1. The electroencephalogram signal X1 and the electroencephalogram signal X2 acquired by the electroencephalogram signal acquisition unit 11 are passed to the first instantaneous phase identification unit 12A and the second instantaneous phase identification unit 12B, respectively.

  The source of the electroencephalogram signal X1 may be, for example, an electrode attached to the first position of the subject's head. In this case, the electroencephalogram signal acquisition unit 11 obtains the electroencephalogram signal X1 representing the electroencephalogram of the subject by sampling the potential of this electrode detected by the external device at a predetermined sampling period Δt. The sampling period Δt may be set to be, for example, 1 ms or more and 4 ms or less (Δt = 1 ms in the present embodiment). The value of the brain wave signal X1 sampled at the time t is hereinafter referred to as X1 (t).

  The source of the electroencephalogram signal X2 includes, for example, an electrode attached to the second position of the subject's head. In this case, the electroencephalogram signal acquisition unit 11 obtains the electroencephalogram signal X2 representing the electroencephalogram of the subject by sampling the potential of the electrode detected by the external device at the sampling period Δt. The value of the brain wave signal X2 sampled at the time t is hereinafter referred to as X2 (t).

  When feeding back the LHPS to the subject, the electroencephalogram signal acquisition unit 11 acquires electroencephalogram signals X1 and X2 representing brain waves measured at two points (first position and second position) on the right hemisphere of the subject's head. . When feeding back the RHPS to the subject, the electroencephalogram signal acquisition unit 11 outputs the electroencephalogram signals X1 and X2 representing the electroencephalogram measured at two points (first position and second position) on the left hemisphere of the subject's head. get. When feeding back the CPS to the subject, the electroencephalogram signal acquisition unit 11 acquires electroencephalogram signals X1 and X2 representing brain waves measured at two points (first position and second position) on the midline of the head of the subject. I do.

  The first instantaneous phase identification unit 12A is configured to sequentially identify (real-time identification) the instantaneous phase φ1 (f, t) from the electroencephalogram signal X1. Here, the instantaneous phase φ1 (f, t) refers to the phase at time t of the component wave X1f having the target frequency f among the component waves included in the brain wave signal X1. The target frequency f is a frequency specified by an operator or the like. For example, when 10 Hz is specified as the target frequency f, the instantaneous phase φ1 (f, t) specified by the first instantaneous phase specifying unit 12A indicates the phase of the α wave near the first position. Become. The instantaneous phase φ1 (f, t) specified by the first instantaneous phase specifying unit 12A is passed to the phase synchronization index calculating unit 13 as shown in FIG.

  The second instantaneous phase specifying unit 12B has a configuration for sequentially specifying the instantaneous phase φ2 (f, t) from the brain wave signal X2. Here, the instantaneous phase φ2 (f, t) refers to the phase at time t of the component wave X2f having the target frequency f among the component waves included in the brain wave signal X2. The instantaneous phase φ2 (f, t) specified by the second instantaneous phase specifying unit 12B is passed to the phase synchronization index calculating unit 13 as shown in FIG.

  Note that a known method may be used to specify the instantaneous phase φ1 (f, t). As a known method that can be used for specifying the instantaneous phase φ1 (f, t), for example, a method of performing convolution integration with a Morlet Wavelet is used. When performing convolution integration with Morlet Wavelet, the first instantaneous phase identification unit 12A determines the instantaneous phase φ1 (f, t) from time t1 = t−Δt × (M−1) / 2 to time tM. .., X1 (tM) of M brain wave signals X1 sampled at each sampling timing ti up to = t + Δt × (M−1) / 2 (for simplicity, M is referred to as Assumed odd). For this reason, the electroencephalogram signal processing device 1 includes a buffer (not shown) that holds at least the values of the M electroencephalogram signals X1, and converts the electroencephalogram signal X1 from the electroencephalogram signal acquisition unit 11 to the first instantaneous phase identification unit 12A. Feeding will take place via this buffer. The same applies to the specification of the instantaneous phase φ2 (f, t).

  The window width tM−t1 = Δt × (M−1) used to specify the instantaneous phase φ1 (f, t) is T, and the instantaneous phase φ1 (f, t) is obtained from X1 (t1),. ) Is assumed to be τ, at least a time T / 2 + τ from completion of the sampling of the electroencephalogram signal X1 (t) to completion of the specification of the instantaneous phase φ1 (f, t). It will cost. Note that the window width T is usually set to be three times or more and eight times or less the period (1 / f) corresponding to the target frequency f. Therefore, for example, when the target frequency f is 20 Hz, the window width T is 150 ms or more and 400 ms or less.

  In the present embodiment, the first instantaneous phase identification unit 12A identifies the instantaneous phase φ1 (f, t) with the minimum required time T / 2 + τ. That is, the delay δ from the completion of the sampling of the value X1 (t) of the electroencephalogram signal X1 to the completion of the specification of the instantaneous phase φ1 (f, t) is the minimum required time T / 2 + τ. However, if the delay δ from completion of the sampling of the value X1 (t) of the electroencephalogram signal X1 to completion of the specification of the instantaneous phase φ1 (f, t) is approximately equal to the minimum required time T / 2 + τ. Is sufficient to achieve the objectives of the present invention. In this specification, that the specification of the instantaneous phase φ1 (f, t) is a sequential specification (real-time specification) means that the delay δ is substantially equal to τ + T / 2. The same applies to the specification of the instantaneous phase φ2 (f, t) by the second instantaneous phase specifying unit 12B.

  The phase synchronization index calculating unit 13 is a configuration for sequentially calculating (real-time calculation) the phase synchronization index PSI (f, t) from the instantaneous phase φ1 (f, t) and the instantaneous phase φ2 (f, t). Here, the phase synchronization index PSI (f, t) is the phase synchronization of the brain wave signal X1 (more precisely, its component wave X1f) and the brain wave signal X2 (more precisely, its component wave X2f) at time t. Refers to a numerical value indicating the degree. The phase synchronization index PSI (f, t) calculated by the phase synchronization index calculation unit 13 is passed to the video signal generation unit 14, as shown in FIG.

  In the present embodiment, the phase synchronization index calculating unit 13 calculates the phase synchronization index PSI (f, t) defined by the following equation, at time t′1 = t−Δt × (N−1) / 2 to the instant t′N = t + Δt × (N−1) / 2, N pairs of instantaneous phases ΔφR (f, t′1) and φL (f, t′1) corresponding to each sampling timing t′i .., {ΦR (f, t′N), φL (f, t′N)} (assuming N is an odd number for simplicity). In the following equation, θ (f, t′i) is a phase difference defined by θ (f, t′i) = φR (f, t′i) −φL (f, t′i). is there.

  If the phase synchronization index PSI (f, t) is defined as above, when the phases of the brain wave signal X1 and the brain wave signal X2 are completely synchronized at time t, PSI (f, t) = 1, and At t, when the phases of the brain wave signal X1 and the brain wave signal X2 are not synchronized at all, PSI (f, t) = 0. Therefore, the phase synchronization index PSI (f, t) defined as described above is an appropriate index indicating the degree of phase synchronization between the brain wave signal X1 and the brain wave signal X2.

  When the electroencephalogram signal acquisition unit 11 adopts a configuration in which electroencephalogram signals X1 and X2 representing electroencephalograms measured at two points on the right hemisphere of the subject's head are adopted, the PSI (f) calculated by the phase synchronization index calculation unit 13 is used. , T) can be used as an index of LHPS. Note that LHPS is defined by an average value of phase synchronization indices between brain waves measured with each pair of electrodes for N pairs of electrodes arranged on the left hemisphere of the head. Therefore, the PSI (f, t) calculated by the phase synchronization index calculating unit 13 when the above configuration is adopted can be regarded as LHPS when N = 1, and LHPS when N ≧ 2. It can also be considered as a constituent quantity.

  When the electroencephalogram signal acquisition unit 11 adopts a configuration in which electroencephalogram signals X1 and X2 representing brain waves measured at two points on the left hemisphere of the subject's head are adopted, the PSI calculated by the phase synchronization index calculation unit 13 is used. (F, t) can be used as an index of RHPS. Note that RHPS is defined by an average value of phase synchronization indices between brain waves measured with N pairs of electrodes arranged on the right hemisphere of the head. Therefore, the PSI (f, t) calculated by the phase synchronization index calculating unit 13 when the above configuration is adopted can be regarded as the RHPS when N = 1, and the RHPS when N ≧ 2. It can also be considered as a constituent quantity.

  In addition, the electroencephalogram signal acquisition unit 11 displays an electroencephalogram representing an electroencephalogram measured at one point on the median line of the subject's head and a different point on the subject's head right hemisphere, left hemisphere, or median line. When the configuration for acquiring the signals X1 and X2 is adopted, the PSI (f, t) calculated by the phase synchronization index calculation unit 13 can be used as an index of CPS. The CPS is a phase synchronization between brain waves measured by N pairs of electrodes, one of which is arranged on the head midline and the other is arranged on the head left hemisphere, the right hemisphere, or the midline. Defined by the average value of the exponent. Therefore, the PSI (f, t) calculated by the phase synchronization index calculating unit 13 when the above configuration is adopted can be regarded as a CPS when N = 1, and a CPS when N ≧ 2. It can also be considered as a constituent quantity.

  Note that, as described above, the phase synchronization index calculating unit 13 refers to the N pairs of instantaneous phases to calculate the phase synchronization index PSI (f, t) at time t. For this reason, the electroencephalogram signal processing device 1 includes a buffer (not shown) that holds at least N instantaneous phases φ1 (f, t). The supply of the instantaneous phase φ1 (f, t) is performed via this buffer. For the same reason, the electroencephalogram signal processing device 1 includes a buffer (not shown) that holds at least N values of the instantaneous phase φ2 (f, t), and calculates the phase synchronization index from the second instantaneous phase identification unit 12B. The supply of the instantaneous phase φ2 (f, t) to the unit 13 is performed via this buffer.

  The window width t′N−t′1 = Δt × (N−1) used for calculating the phase synchronization index PSI (f, t) is T ′, and ｛φR (f, t′1), φL (f , T′1)},..., {ΦR (f, t′N), φL (f, t′N)}, the time required to calculate the phase synchronization index PSI (f, t) is τ ′. , It takes at least a time T ′ / 2 + τ ′ after the specification of the instantaneous phases φ1 (f, t) and φ2 (f, t) to complete the calculation of the phase synchronization index PSI (f, t). Will be. Note that the window width T 'is usually set to be three times or more and eight times or less the period (1 / f) corresponding to the target frequency f. Therefore, for example, when the target frequency f is 20 Hz, the window width T 'is 150 ms or more and 400 ms or less.

  In the present embodiment, the phase synchronization index calculation unit 13 calculates the phase synchronization index PSI (f, t) using the minimum required time T / 2 + τ. That is, the delay δ ′ from completion of the specification of the instantaneous phases φ1 (f, t) and φ2 (f, t) to completion of the calculation of the phase synchronization index PSI (f, t) is the minimum required time T '/ 2 + τ'. However, the delay δ ′ from the completion of the specification of the instantaneous phases φ1 (f, t) and φ2 (f, t) to the completion of the calculation of the phase synchronization index PSI (f, t) is the minimum required time T A similar degree to '/ 2 + τ' is sufficient to achieve the object of the present invention. In this specification, that the calculation of the phase synchronization index PSI (f, t) is a sequential calculation (real-time calculation) means that the above-mentioned delay δ ′ is substantially equal to the minimum required time T ′ / 2 + τ ′. I do.

  The calculation of the phase synchronization index PSI (f, t) by the phase synchronization index calculation unit 13 may be performed for each Δt (sampling cycle) or may be performed for each T ′ (window width). , T ′ / 2. By adopting a configuration in which the calculation of the phase synchronization index PSI (f, t) is performed every T 'or T' / 2, the calculation load can be reduced.

  The video signal generation unit 14 generates, from the phase synchronization index PSI (f, t), a video signal representing a video whose contents change in conjunction with the phase synchronization index PSI (f, t). The video signal generated by the video signal generation unit 14 is supplied to a display connected to or incorporated in the brain wave signal processing device 1. The content of the video represented by the video signal generated by the video signal generation unit 14 will be described later with reference to a different drawing.

  The display connected to or incorporated in the electroencephalogram signal processing device 1 changes the image represented by the image signal generated by the image signal generation unit 14, that is, the content thereof in conjunction with the phase synchronization index PSI (f, t). The video is presented to the subject. In other words, it functions as a stimulus giving device that gives a subject a visual stimulus that changes in conjunction with the phase synchronization index PSI (f, t). As a result, a real-time feedback loop that feeds back the subject's LHPS, RHPS, or CPS to the subject is formed.

  In the electroencephalogram signal processing device 1, after the sampling of the values X1 (t) and X2 (t) of the electroencephalogram signals X1 and X2 is completed, the instantaneous phases φ1 (f, t) and φ2 (f, t) The delay δ until the specification is completed is the minimum required time T / 2 + τ (when the specification of the instantaneous phase φ1 (f, t) and the specification of the instantaneous phase φ2 (f, t) are executed in parallel) or T / 2 + 2 × τ (when the instantaneous phase φ1 (f, t) and the instantaneous phase φ2 (f, t) are serially executed). Further, in the electroencephalogram signal processing device 1, a delay from completion of the specification of the instantaneous phases φ1 (f, t) and φ2 (f, t) to completion of calculation of the phase synchronization index PSI (f, t). δ ′ becomes the minimum required time T ′ / 2 + τ ′. Therefore, if the time required for the generation and output of the video signal is ignored, the degree of phase synchronization between the brain wave signal X1 and the brain wave signal X2 is changed, and the change is reflected on the image presented to the subject. (Hereinafter referred to as “response time of the electroencephalogram signal processing device 1”) is T / 2 + τ + T ′ / 2 + τ ′ or T / 2 + 2 × τ + T ′ / 2 + τ ′. Thereby, the response time of the electroencephalogram signal processing device 1 can be set to a time short enough to obtain the effect of the training. That is, the above-described feedback can be made more real-time, and the effect of training can be enhanced. For example, assuming that the window widths T and T ′ are T = T ′ = 400 ms, if the calculation times τ and τ ′ are sufficiently smaller than the window widths T and T ′, the response time of the electroencephalogram signal processing device 1 becomes It takes about 400 ms.

  If the response time of the electroencephalogram signal processing device 1 is 15 seconds or less, the subject can know the degree of the phase synchronization of his / her own brain wave substantially in real time, and thereby the degree of the phase synchronization of his / her own brain wave can be determined by his / her own. It becomes possible to control by will. In this specification, that the feedback is in real time means that the response time of the electroencephalogram signal processing device 1 is 15 seconds or less. It should be noted that (1) even if the stimulus according to the phase synchronization index PSI (f, t) at each time t is given to the subject at time t + Δ, (2) each time t0 from time t0 to time t0 + Δ Even if the stimulus corresponding to the average value of the phase synchronization index PSI (f, t) is given to the subject at time t0 + Δn, if Δ is 15 seconds or less, it is regarded as real-time feedback.

  Note that the shorter the response time of the electroencephalogram signal processing device 1 is, the higher the real-time feedback is, and the easier it is for the subject to control the degree of phase synchronization of his / her brain with his / her own will. Therefore, for example, a configuration in which the response time of the electroencephalogram signal processing device 1 is 5 seconds or less, 4 seconds or less, 3 seconds or less, 2 seconds or less, or 1 second or less is possible, and the latter configuration is more effective. Is expected.

(Modification)
The electroencephalogram signal processing device 1 shown in FIG. 1 employs a configuration in which the instantaneous phases φ1 (f, t) and φ2 (f, t) are specified by convolution integration with Morlet Wavelet. However, the present invention is not limited to this. That is, for example, a configuration may be adopted in which the instantaneous phases φ1 (f, t) and φ2 (f, t) are specified by performing analysis signal conversion using Hilbert transform after passing through a frequency filter.

  Further, in the electroencephalogram signal processing device 1 shown in FIG. 1, the phase based on the instantaneous phases φ1 (f, t) and φ2 (f, t) of the component waves X1f, X2f of the specific frequency included in the electroencephalogram signals X1, X2. Although the configuration for calculating the synchronization index PSI (f, t) is adopted, the present invention is not limited to this. That is, a configuration may be adopted in which the phase synchronization index PSI (t) is calculated based on the instantaneous phases φ1 (t) and φ2 (t) of the brain wave signals X1 and X2 including all frequency components. That is, the expression of the “instantaneous phase of the brain wave signal” is any of the instantaneous phases φ1 (f, t) and φ2 (f, t) of the specific frequency, and the instantaneous phases φ1 (t) and φ2 (t) whose frequency is not specified. Please note that it is not limited to the former.

  Further, in the electroencephalogram signal processing device 1 shown in FIG. 1, a video signal supplied to a stimulus giving device (for example, a display device) for giving a subject a visual stimulus that changes in accordance with the phase synchronization index PSI (f, t) is provided. Although it is configured to generate, the present invention is not limited to this. That is, any configuration can be used as long as it generates a signal to be supplied to the stimulus applying device that applies a stimulus that changes to the subject in synchronization with the phase synchronization index PSI (f, t).

  For example, the present invention can be replaced with a configuration that generates an audio signal to be supplied to a stimulus applying device that provides an auditory stimulus that changes in conjunction with the phase synchronization index PSI (f, t) to a subject. Examples of the stimulus applying device that applies an auditory stimulus to a subject include a speaker and the like. The volume of the auditory stimulus given to the subject via the stimulus imparting device may be linked to the phase synchronization index PSI (f, t), or the pitch of the auditory stimulus may be adjusted to the phase synchronization index PSI (f, t). The timbre may be interlocked, or the timbre may be interlocked with the phase synchronization index PSI (f, t). That is, the audio signal generated by the electroencephalogram signal processing device 1 may be such that its amplitude is interlocked with the phase synchronization index PSI (f, t), or its cycle is the phase synchronization index PSI (f, t). Or the waveform thereof may be linked to the phase synchronization index PSI (f, t).

  Further, the configuration may be replaced with a configuration that generates a signal to be supplied to a stimulus applying device that applies a somatosensory stimulus that changes in conjunction with the phase synchronization index PSI (f, t) to the subject. Examples of a stimulating device that applies a somatosensory stimulus to a subject include, for example, an electrical stimulator that stimulates the somatosensory of the subject by electrical means and a mechanical stimulator that stimulates the somatosensory of the subject by mechanical means And the like. A rehabilitation brace that bends the wrist is an example of such a mechanical stimulator. The intensity of the somatosensory stimulus given to the subject via the stimulus imparting device may be such that its intensity is linked to the phase synchronization index PSI (f, t), or its cycle is the phase synchronization index PSI (f, t). ), Or the pattern may be linked to the phase synchronization index PSI (f, t). That is, the signal generated by the electroencephalogram signal processing device 1 may be such that its amplitude is interlocked with the phase synchronization index PSI (f, t), or its cycle is equal to the phase synchronization index PSI (f, t). The waveform may be interlocked, or the waveform may be interlocked with the phase synchronization index PSI (f, t).

  More specifically, instead of providing a visual stimulus, an auditory stimulus, or a somatosensory stimulus according to the phase synchronization index PSI (f, t), an odor stimulus or a taste sensation according to the phase synchronization index PSI (f, t) is used. It is not impossible to adopt a configuration that provides stimulation. In short, the object of the present invention is achieved with any configuration as long as the subject can know the phase synchronization index PSI (f, t) as some sense.

  Further, in the electroencephalogram signal processing device 1 shown in FIG. 1, a configuration for calculating the phase synchronization index PSI (f, t) from a pair of electroencephalogram signals {X1, X2} is adopted. It is not limited. That is, a configuration for calculating the phase synchronization index PSI (f, t) from n brain wave signals {X1, X2,..., Xn} may be adopted (n is an even number of 2 m). By employing such a configuration, not only LHPS, RHPS, and CPS but also GPS (Global Phase Synchronization) can be fed back to the subject.

  FIG. 2 shows a block diagram of the electroencephalogram signal processing device 1 employing such a configuration. The electroencephalogram signal processing device 1 shown in FIG. 2 has n instantaneous phase identifying units (instantaneous phase identifying means) 12- having the same functions as the first instantaneous phase identifying unit 12A and the second instantaneous phase identifying unit 12B in FIG. 1 to 12-n is different from the electroencephalogram signal processing device 1 in FIG. Further, in the electroencephalogram signal processing device 1 shown in FIG. 2, the phase synchronization index calculating unit 13 calculates (1) the instantaneous phases φj (f, t) and φj + 1 (f, t) for each j of 1 ≦ j ≦ m. PSIj (f, t) calculation, and (2) average value of PSIj (f, t)） PSI1 (f, t) +... + PSIm (f, t) as phase synchronization index PSI (f, t) This is different from the electroencephalogram signal processing device 1 in FIG. 1 in that the processing for calculating｝ / m is performed. In other respects, the configuration is the same as that of the electroencephalogram signal processing device 1 in FIG.

  In FIG. 2, a phase synchronization index PSI1 (f, t) is obtained by pairing the brain wave signal X1 and the brain wave signal X2, and a phase synchronization index PSI2 (f, t) is formed by pairing the brain wave signal X3 and the brain wave signal X4. .., And the phase synchronization index PSIm (f, t) is obtained by pairing the brain wave signal Xn−1 and the brain wave signal Xn, but the present invention is not limited to this. That is, the method of assembling pairs of brain wave signals for obtaining the phase synchronization index is arbitrary, and n × (n−1) / 2 ways of assembling pairs are allowed.

  However, when the LHPS is fed back to the subject, it is assumed that all pairs are pairs of electroencephalogram signals indicating potentials of electrodes attached to two points on the left hemisphere of the subject's head. When the RHPS is fed back to the subject, it is assumed that all pairs are pairs of brain wave signals indicating the potentials of the electrodes attached to two points on the right hemisphere of the subject's head. Also, when feeding back the CPS to the subject, all pairs have an electroencephalogram signal indicating the potential of the electrode attached to one point on the midline of the subject's head, and the subject's head left hemisphere different from this point, It is assumed that the pair is a pair with an electroencephalogram signal indicating the voltage of the electrode attached to one point on the right hemisphere or the midline. When the GPS is fed back to the subject, at least (1) a pair of electroencephalogram signals indicating potentials of electrodes attached to two points on the left hemisphere of the subject's head, (2) a right hemisphere of the subject's head An electroencephalogram signal pair indicating the potential of the electrode attached to the above two points, and (3) an electroencephalogram signal indicating the potential of the electrode attached to the point on the left hemisphere of the subject's head and on the right hemisphere of the subject's head. And a pair of electroencephalogram signals indicating the potentials of the electrodes attached to the points.

  Note that a configuration may be adopted in which a part or all of LHPS, RHPS, IHPS, and CPS are multiplied by a predetermined coefficient and added, and a linear sum is fed back to the subject. At this time, the average value of a plurality of PSIs calculated from a pair of electroencephalogram signals indicating potentials of mutually different electrode pairs on the left hemisphere of the head (the above-described “LHPS when N ≧ 2”) is subjected to linear summing. Alternatively, each of the plurality of PSIs (the above-described “LHPS when N = 1”) may be individually subjected to the linear sum. The same applies to RHPS, IHPS, and CPS. Further, PSI calculated from a pair of brain wave signals indicating the potential of the electrode pair arranged on the midline of the head may be added to the target of the linear sum.

  As described above, a configuration is employed in which a phase synchronization index is sequentially calculated according to the degree of phase synchronization between brain wave signals using brain wave signals acquired at three or more different positions on the subject's head. As a result, it is possible to calculate a phase synchronization index that more accurately reflects the degree of phase synchronization in the brain of the subject as compared with the case where brain wave signals acquired at two locations are used. This makes it possible to provide more appropriate feedback according to the state in the brain of the subject.

[Video presented to the subject]
Next, an example of an image presented to the subject by the electroencephalogram signal processing device 1 will be described with reference to FIG. 3A to 3E are diagrams each showing an example of a video presented to the subject by the electroencephalogram signal processing device 1.

  The video shown in FIG. 3A includes a rectangular object 21 whose long side is parallel to the horizontal axis of the screen at the center of the screen. In this video, the length L of the long side of the object 21 changes in conjunction with the phase synchronization PSI (f, t). The linking is arbitrary, but here, a positive correlation is provided between the length L of the long side of the object 21 and the phase synchronization index PSI (f, t). More specifically, the length L of the long side of the object 21 is made proportional to the phase synchronization index PSI (f, t).

  When presenting this video to the subject, the subject is required to be aware that the length L of the long side of the object 21 is long. The subject controls his / her brain so as to promote the increase of LHPS, RHPS, CPS, or GPS by consciously increasing the length L of the long side of the object 21.

  The video shown in FIG. 3B includes a rectangular object 22 whose long side is parallel to the vertical axis of the screen at the center of the screen. In this video, the length L of the long side of the object 22 changes in conjunction with the phase synchronization PSI (f, t). The method of linking is arbitrary, but here, a positive correlation is provided between the length L of the long side of the object 22 and the phase synchronization index PSI (f, t). More specifically, the length L of the long side of the object 22 is made proportional to the phase synchronization index PSI (f, t).

  When presenting this video to the subject, the subject is required to be aware that the length L of the long side of the object 22 is long. The subject controls his / her brain so as to promote the increase of LHPS, RHPS, CPS, or GPS by consciously increasing the length L of the long side of the object 22.

  The image shown in FIG. 3C includes a colored disk-shaped object 23 with a variable color temperature at the center of the screen. In this video, the color temperature of the object 23 changes in conjunction with the phase synchronization PSI (f, t). The method of linking is arbitrary, but here, a negative correlation is provided between the color temperature of the object 23 and the phase synchronization index PSI (f, t). More specifically, the color temperature of the object 23 and the phase synchronization index PSI (f, t) are made in inverse proportion.

  When presenting this video to the subject, the subject is required to be aware that the color of the object 23 is warmed (cooled). The test subject will control his brain to promote an increase in LHPS, RHPS, CPS, or GPS by conscious of warming the color of the object 23.

  The image shown in FIG. 3D includes a grayscale photograph 24 having a variable number of gradations at the center of the screen (in FIG. 3D, a grayscale photograph having two gradations, that is, a black and white photograph is exemplified. ing). In this video, the number of gradations of the grayscale photograph 24 changes from 2 to 256 in conjunction with the phase synchronization PSI (f, t). The linking is arbitrary, but a positive correlation is provided between the number of tones of the photograph 24 and the phase synchronization index PSI (f, t). More specifically, the number of tones of the photograph 24 and the phase synchronization index PSI (f, t) are made proportional.

  When presenting the video to the subject, for example, a task of specifying the subject is imposed on the subject. The test subject controls his or her brain to promote the increase of LHPS, RHPS, CPS, or GPS by increasing the number of gradations of the photograph 24 and conscious of identifying the subject.

  The image shown in FIG. 3 (e) is shown at the upper part of the screen where two balls are arranged at the upper left and lower right of the screen, and at the lower part of the figure where two balls are arranged at the upper right and lower left of the screen. The screen is an animation displayed alternately. In this animation, the distance D between the two balls in the horizontal direction of the screen changes in conjunction with the phase synchronization index PSI (f, t). The method of linking is arbitrary, but here, a negative correlation is provided between the phase synchronization index PSI (f, t) and the distance D between the two balls.

  In this animation, it may be recognized that two balls arranged side by side vibrate up and down in opposite phases, and it is recognized that two balls arranged up and down vibrate left and right in opposite phases Sometimes. When the subject has the latter recognition, the subject's right brain and left brain phases are likely to be synchronized, and when the subject's right brain and left brain phases are synchronized, the subject recognizes the latter recognition. It is likely to have. That the subject has the latter recognition means that the ball arranged at the upper left of the screen shown in the upper part of the figure and the ball arranged at the upper right of the screen shown in the lower part of the figure are identified with the cooperation of the right brain and the left brain. It is nothing but doing.

  When presenting this animation to the subject, the subject is asked, for example, a question such as "Does two vertically arranged balls seem to be oscillating right and left in opposite phases?" The subject will control his brain to promote IHPS (Inter-Hemispheric Phase Synchronization) by consciously trying to recognize this animation as such. And this would also control to promote LHPS, RHPS, CPS or GPS increase.

[Configuration example of brain wave signal processing device]
The electroencephalogram signal processing device 1 can be configured using, for example, a computer (electronic computer). FIG. 4 is a block diagram illustrating a configuration of a computer 100 that can be used as the electroencephalogram signal processing device 1.

  As shown in FIG. 4, the computer 100 includes an arithmetic unit 120, a main storage device 130, an auxiliary storage device 140, and an input / output interface 150 connected to each other via a bus 110. Examples of a device that can be used as the arithmetic device 120 include a CPU (Central Processing Unit). Examples of a device that can be used as the main storage device 130 include a semiconductor RAM (random access memory). A device that can be used as the auxiliary storage device 140 is, for example, a hard disk drive.

  The input device 200 and the output device 300 are connected to the input / output interface 150 as shown in FIG. An electroencephalograph for detecting the electroencephalogram signals X1 and X2, a keyboard for specifying a target frequency f, and the like are examples of the input device 200 connected to the input / output interface 150. A display for displaying an image whose contents change in conjunction with the phase synchronization index PSI (f, t) is an example of the output device 300 connected to the input / output interface 150. Of course, the input device 200 and the output device 300 are not limited to this example as long as they can receive necessary input and output necessary information. For example, the input device 200 and the output device 300 may be configured by a touch panel.

  Various programs for operating the computer 100 as the electroencephalogram signal processing device 1 are stored in the auxiliary storage device 140. Specifically, a brain wave signal acquisition program, a first instantaneous phase identification program, a second instantaneous phase identification program, a phase synchronization index calculation program, and a video signal generation program are stored. These programs may be modules included in a numerical calculation library such as MATLAB (registered trademark).

  The arithmetic unit 120 expands the programs stored in the auxiliary storage device 140 on the main storage device 130 and executes the instructions included in the programs expanded on the main storage device 130, thereby executing the computer 100 Function as an electroencephalogram signal acquisition unit 11, a first instantaneous phase identification unit 12A, a second instantaneous phase identification unit 12B, a phase synchronization index calculation unit 13, and a video signal generation unit 14. The main storage device 130 also functions as a buffer that holds the values of the brain wave signals X1 and X2 and a buffer that holds the instantaneous phases φ1 (f, t) and φ2 (f, t).

  Here, the configuration has been described in which the computer 100 functions as the electroencephalogram signal processing device 1 using the programs stored in the auxiliary storage device 140 which is an internal recording medium, but the present invention is not limited to this. Not something. That is, a configuration may be adopted in which the computer 100 functions as the brain wave signal processing device 1 using a program recorded in an external recording medium. As the external recording medium, a computer-readable “temporary tangible medium” such as a tape, a disk, a card, a semiconductor memory, and a programmable logic circuit can be used.

  Further, the computer 100 may be configured to be connectable to a communication network, and the above programs may be supplied to the computer 100 via the communication network. This communication network is not particularly limited as long as it can transmit a program. Note that the present invention can also be realized in the form of a data signal embedded in a carrier wave, in which the program is embodied by electronic transmission.

[Application of EEG signal processing device]
One of the uses of the electroencephalogram signal processing device 1 is rehabilitation for a brain disease patient such as a stroke.

  The PSI of stroke patients tends to be lower than the PSI of healthy individuals. This is considered to be a reflection of the fact that cooperation in each part of the brain is not performed smoothly in stroke patients. The tendency of PSI to decrease in stroke patients is observed regardless of the injury site due to stroke. For example, it is known that IHPS decreases even if the injury site is a part of the right brain.

  In addition, recent studies have revealed that there is a correlation between PSI measured before starting rehabilitation and the effect of rehabilitation on rehabilitation for stroke patients. That is, it has become clear that rehabilitation for stroke patients with high PSI is more effective than rehabilitation for stroke patients with low PSI.

  This trend suggests that rehabilitation for stroke patients may be enhanced by including training to increase PSI. The brain wave signal processing device 1 can be used as a training device for such training. Of course, it is expected that similar effects can be obtained even when applied to rehabilitation for brain disease patients other than stroke.

[Correlation between PSI and disease state index]
Next, the correlation between the LHPS and the disease state index will be described with reference to FIG. FIG. 5 is a diagram showing the correlation between PSI and FIMt by plotting the experimental results with LHPS on the vertical axis and FIMt on the horizontal axis. In addition, FIM (Functional Independence Measure: Functional Independence Index) is a score of the degree of independence in predetermined evaluation items such as self-care such as diet, changing clothes, excretion, and communication. Is FIMt (FIM total). The greater the FIMt, the higher the degree of independence, and the smaller the FIMt, the lower the degree of independence (requires extra care). In addition, LHPS in the figure is a value calculated based on the measurement result of the electroencephalogram in the α band (8 to 13 Hz).

As shown in FIG. 5, the result was that FIMt increased as LHPS increased. In other words, the smaller the LHPS (the lower the degree of synchronization in the left brain), the lower the FIMt (the lower the degree of independence). The p value indicating the degree of correlation was on the order of 10 ^-3 , indicating that LHPS and FIMt have a high positive correlation. For this reason, it is considered that it is possible to promote improvement of FIMt by making the subject recognize how to use the brain to increase LHPS.

  Next, the correlation between PSI and NIHS (National Institute of Health (NIH) Stroke Scale) will be described with reference to FIG. FIG. 6 is a diagram showing the correlation between PSI and NIHSS by plotting experimental results with the vertical axis being LHPS and the horizontal axis being NIHSS. The NIHSS is one of the scales for evaluating the severity of stroke, and is obtained by converting the state of the subject in a predetermined evaluation item into a score and summing up the scores. NIHSS, as opposed to FIMt, indicates that the higher the score, the higher the severity.

As shown in FIG. 6, the result is that NIHSS decreases as LHPS increases. In other words, subjects with larger LHPS (higher degree of synchronization in the left brain) had lower NIHSS (lower severity). The p value indicating the degree of correlation was on the order of 10 ^-3 , indicating that LHPS and NIHSS had a high negative correlation. For this reason, it is considered that NIHSS can be improved by making the subject recognize how to use the brain to increase LHPS.

[Relationship between functional recovery and synchronization index recovery]
FIG. 7 is a diagram showing changes in FIMt and LHPS, which were examined for a plurality of cerebral infarction patients. For each patient, two measurements before and after rehabilitation are performed, and the measurement results are plotted on a coordinate plane of the vertical axis LHPS and the horizontal axis FIMt. And the arrow from the plot before the rehabilitation to the plot after the rehabilitation is described to indicate the change of FIMt and LHPS before and after the rehabilitation.

  In the example shown, of the 12 patients, 10 have increased both FIMt and LHPS after rehabilitation. Thus, it is understood that the function is recovered when the value of LHPS increases, that is, the function recovery is correlated with the synchronization index.

[Statistical significance of correlation between various synchronization indicators and various disease state indicators]
In the previous and previous sections, it was explained that various pathological indices show significant correlation with LHPS, but the synchronous index showing significant correlation with various pathological indices is not limited to LHPS. That is, the various disease state indexes show a significant correlation with IHPS, RHPS, CPS, and GPS.

  Hereinafter, the statistical significance of the correlation between the various synchronization indexes and the various disease state indexes will be described with reference to FIGS.

  FIG. 8 is a view showing a list of p-values indicating the statistical significance of the correlation between various synchronization indices and various disease state indices in a certain population (a population of cerebral infarction patients). Values indicating a significant correlation (p ≦ 0.05) are underlined, and values indicating a correlation tendency (p ≦ 0.09) are shown in bold.

  The synchronization index is a value calculated based on the measurement result of the electroencephalogram in the α band (8 to 13 Hz). FIMm is an exercise item of the above-mentioned FIM, and FIMc is a recognition item. FMA (Fugl-Meyer Assessment) is one of the methods for evaluating physical function, FMAul is one of the methods for evaluating motor function of the upper limb, and FMAll is one of the methods for evaluating motor function of the lower limb. is there.

  From FIG. 8, it is confirmed that LHPS, GPS, and CPS in the α band are significantly correlated with the majority of pathological indicators. This suggests that the feedback of any one of LHPS, GPS, and CPS in the α band to a cerebral infarction patient may improve the condition.

  FIG. 9 to FIG. 12 are lists showing p-values indicating the statistical significance of the correlation between various synchronization indices and various disease state indices in a population (a cerebral infarction patient population) different from the above population. .

  In particular, FIG. 9 shows the results of calculating the p-value indicating the statistical significance of the correlation between various synchronization indices and various disease state indices in the α band (8 to 13 Hz) without limiting the population. FIG. 10 (a) shows the result of calculating the p-value indicating the statistical significance of the correlation between various synchronization indices and various pathological indices in the α band by limiting the population to left hemisphere lesion patients, FIG. 10 (b) shows the result of calculating the p-value indicating the statistical significance of the correlation between various synchronization indices and various pathological indices in the α band by limiting the population to right hemisphere lesion patients. Again, values showing significant correlations (p ≦ 0.05) are underlined.

  From FIG. 9, it is confirmed that RHPS in the α band is significantly correlated with the majority of pathological indicators. This suggests that the feedback of RHPS in the α band to a cerebral infarction patient may improve the condition. Also, from FIG. 10 (b), it can be confirmed that IHPS, LHPS, RHPS, GPS, and CPS in the α-band significantly correlate with the pathological index of the majority in the right hemisphere lesion patient. This suggests that for right hemisphere lesion patients, it is possible to improve the disease state by feeding back any of IHPS, LHPS, RHPS, CPS, and GPS in the α band.

  FIG. 11 shows the result of calculating the p-value indicating the statistical significance of the correlation between various synchronization indexes and various disease state indexes in the β1 band (13 to 18 Hz) without limiting the population. FIG. 12 (a) shows the result of calculating the p-value indicating the statistical significance of the correlation between various synchronization indices and various pathological indices in the β1 band, with the population limited to left hemisphere lesion patients, FIG. 12 (b) shows the result of calculating the p-value indicating the statistical significance of the correlation between various synchronization indices and various pathological indices in the β1 band, with the population limited to right hemisphere lesion patients. Again, values showing significant correlations (p ≦ 0.05) are underlined.

  From FIG. 11, it is confirmed that the IHPS in the β1 band is significantly correlated with the majority of pathological indicators. This suggests that the feedback of IHPS in the β1 band to a cerebral infarction patient may improve the condition. Also, from FIG. 12 (b), it can be confirmed that for the right hemisphere lesion patient, both IHPS and CPS in the β1 band are significantly correlated with the majority pathological index. This suggests that for patients with right hemisphere lesions, it is possible to improve the disease state by feeding back IHPS or CPS in the β1 band.

  From these, the following conclusions are drawn. That is, conventionally, only IHPS was considered to show a significant correlation with various pathological indices. However, if the target patients are limited, or if the target frequencies are limited, LHPS, RHPS, CPS, GPS Also show significant correlation with various pathological indicators. That is, the feedback of LHPS, RHPS, CPS, or GPS to the subject may possibly improve the condition.

In addition, various pathological indices of a cerebral infarction patient may correlate with not only various synchronous indices but also a lesion volume (LV). Therefore, a p-value indicating the statistical significance of the following partial correlation was obtained by partial correlation analysis. In addition, the value calculated from the MRI structural image of the cerebral infarction patient was used as the lesion volume.

(1) partial correlation between various synchronization indices (LHPS, IHPS, RHPS) and various pathological indices (FIMm, FIMc) excluding the influence of lesion volume (LV),
(2) partial correlation between various pathological indices (FIMm, FIMc) excluding the influence of various synchronous indices (LHPS, IHPS, RHPS) and lesion volume (LV),
(3) Partial correlation between the lesion volume (LV) excluding the influence of various pathological indices (FIMm, FIMc) and various synchronous indices (LHPS, IHPS, RHPS).

  FIGS. 13A to 13C are diagrams showing a list of p-values representing the statistical significance of the above three partial correlations for right hemispheric lesion patients. (A) shows a synchronization index RHPS, a disease state index FIMm, and a lesion volume LV, (1) partial correlation between RHPS and FIMm, (2) partial correlation between FIMm and LV, (3) LV and RHPS. Shows the p-value of the partial correlation with (B) shows the synchronous index IHPS, the pathological index FIMm, and the lesion volume LV, (1) partial correlation between IHPS and FIMm, (2) partial correlation between FIMm and LV, (3) LV and IHPS. Shows the p-value of the partial correlation with (C) shows the synchronization index LHPS, the pathological index FIMm, and the lesion volume LV, (1) partial correlation between LHPS and FIMm, (2) partial correlation between FIMm and LV, (3) LV and LHPS. Shows the p-value of the partial correlation with

  FIG. 13D is a schematic diagram illustrating the relationship among the synchronization indexes RHPS, IHPS, and LHPS, the pathological index FIMm, and the lesion volume LV.

  According to FIG. 13A, the synchronization index RHPS shows a significant partial correlation (p ≦ 0.05) with the disease state index FIMm, whereas the lesion volume LV shows a significant partial correlation with the disease state index FIMm. It can be seen that there is no significant partial correlation between them. That is, it is understood that the synchronization index RHPS indicating the network soundness in the right hemisphere is strongly related to the disease state of the right hemisphere lesion patient, not the lesion volume LV. This suggests that the RHPS in the α-wave band (8 Hz to 13 Hz) may be fed back to a cerebral infarction patient to improve the condition.

  13B and 13C that a significant partial correlation (p ≦ 0.05) exists between the disease state index FIMm and the synchronization indexes LHPS and IHPS. Therefore, it can be understood that the synchronization indices LHPS and IHPS, like the synchronization index RHPS, are indices that better explain the pathology of the right hemisphere lesion patient than the lesion volume LV.

  FIGS. 14A and 14B are diagrams showing a list of p-values representing the statistical significance of the above three partial correlations for patients with left hemisphere lesions. (A) shows a synchronization index LHPS, a pathological index FIMm, and a lesion volume LV, (1) partial correlation between LHPS and FIMm, (2) partial correlation between FIMm and LV, (3) LV and LHPS. Shows the p-value of the partial correlation with (B) shows the synchronization index LHPS, the disease state index FIMc, and the lesion volume LV, (1) partial correlation between LHPS and FIMc, (2) partial correlation between FIMc and LV, (3) LV and LHPS. Shows the p-value of the partial correlation with

  FIG. 14C is a schematic diagram illustrating a relationship among the synchronization index LHPS, the disease state index FIMm, and FIMc, and the lesion volume LV.

  According to FIG. 14, the synchronization index LHPS shows a significant partial correlation (p ≦ 0.05) between both the disease state index FIMm and the disease state index FIMc, whereas the lesion volume LV shows the disease state index FIMc. Shows a significant partial correlation with the pathological condition index FIMm, but does not show a significant partial correlation with the pathological condition index FIMm. In other words, it is understood that the synchronization index LHPS indicating the network soundness in the left hemisphere, not the lesion volume LV, is strongly related to the pathological condition of the left hemisphere lesion patient. This suggests that the feedback of the synchronization index LHPS of the β band (13 to 21 Hz) to the left hemisphere lesion patient may improve the condition.

[Processing flow]
Next, a flow of processing executed by the electroencephalogram signal processing device 1 shown in FIG. 1 will be described with reference to FIG. FIG. 15 is a flowchart illustrating an example of an electroencephalogram signal processing method executed by the electroencephalogram signal processing device 1.

  First, the electroencephalogram signal acquisition unit 11 acquires an electroencephalogram signal of a subject (S1). Here, two electroencephalogram signals (X1, X2) measured at two locations (first position and second position) are acquired. The electroencephalogram signal acquisition unit 11 passes X1 of the acquired electroencephalogram signal to the first instantaneous phase identification unit 12A, and passes X2 to the second instantaneous phase identification unit 12B. In the case of the electroencephalogram signal processing device 1 of FIG. 2, electroencephalogram signals measured at at least three places on the subject's head are acquired, and each electroencephalogram signal is passed to a different instantaneous phase identification unit 12-n.

  Next, the first instantaneous phase identification unit 12A and the second instantaneous phase identification unit 12B identify the instantaneous phase at the target frequency f from the brain wave signal passed from the brain wave signal acquisition unit 11 (S2). Note that the target frequency f is specified in advance by an operator or the like. Then, the first instantaneous phase identifying unit 12A and the second instantaneous phase identifying unit 12B pass the identified instantaneous phases φ1 (f, t) and φ2 (f, t) to the phase synchronization index calculating unit 13.

  Next, the phase synchronization index calculation unit 13 calculates a phase synchronization index from the instantaneous phases received from the first instantaneous phase identification unit 12A and the second instantaneous phase identification unit 12B (S3), and converts the calculated phase synchronization index into a video signal. The information is passed to the generation unit 14.

  Then, the video signal generation unit 14 generates a video signal according to the received phase synchronization index (S4, signal generation step), and supplies the generated video signal to a display connected to or incorporated in the brain wave signal processing device 1. . Thus, an image corresponding to the image signal is displayed on the display (S5). For example, when the image of FIG. 3A is displayed, the length L of the long side of the object 21 to be displayed may be determined in advance with respect to the value of the reference phase synchronization index. Thus, the length L of the object 21 can be determined from the ratio of the received phase synchronization index to the reference value, and a video signal for performing such display can be generated.

  The processes in S1 to S5 are repeated until the electroencephalogram signal processing ends (YES in S6). By this repetition, a phase synchronization index is sequentially calculated from the sequentially acquired brain wave signals, a video signal corresponding to the phase synchronization index is sequentially generated, and sequentially supplied to the display. Therefore, it is possible to transmit the result of the subject who has seen the image displayed on the display, which is conscious of changing the image, to the subject almost in real time by the change of the image. As a result, the subject can recognize how to use the brain to increase LHPS, RHPS, CPS, or GPS, and can efficiently improve the degree of synchronization.

[Embodiment 2]
Next, another embodiment of the present invention will be described with reference to FIGS. In this embodiment, an example will be described in which a symmetry index indicating the degree of symmetry between hemispheres of information transmission is used instead of the LHPS in the above embodiment. Note that the same components as those in the above embodiment are denoted by the same reference numerals, and description thereof will be omitted.

[Configuration of EEG signal processing device]
The configuration of the electroencephalogram signal processing device 2 of the present embodiment will be described with reference to FIG. FIG. 16 is a block diagram illustrating a main configuration of the electroencephalogram signal processing device 2 according to the present embodiment. The electroencephalogram signal processing device 2 functions as a point where the detection position of the electroencephalogram acquired by the electroencephalogram signal acquisition unit 11 exists in both the right brain and the left brain, and a symmetry index calculation unit (importance index calculation unit and symmetry index calculation unit). 2) and the EEG signal processing apparatus 1 shown in FIG. 2 in that a video signal corresponding to the symmetry index is displayed instead of the phase synchronization index.

  Although details will be described later, the symmetry index and the disease state index are correlated. For this reason, in the electroencephalogram signal processing device 2, similarly to the electroencephalogram signal processing device 1 of the above-described embodiment, the result of the subject who sees the video displayed on the display conscious of changing the video is referred to as the result of the video. The change can be transmitted to the subject almost in real time. This allows the subject to recognize how to use the brain to increase the value of the symmetry index, and to efficiently increase the value of the symmetry index. Further, as described above, since the symmetry index and the disease state index are correlated, improvement of the disease state can be expected by improving the symmetry index.

  The electroencephalogram signal acquisition unit 11 acquires an electroencephalogram signal in at least two places of the right hemisphere of the head and an electroencephalogram signal in at least two places of the left hemisphere of the head. 12-n specifies the instantaneous phase of each of the acquired brain wave signals. From the viewpoint of calculating a symmetry index accurately reflecting the symmetry between the hemispheres of information transmission, the acquired electroencephalogram signal may be an electroencephalogram signal detected at a position symmetrical with respect to the median plane. preferable.

  Then, the phase synchronization index calculation unit 13 (phase synchronization index calculation means) performs a process of calculating a phase synchronization index by pairing the instantaneous phases identified by the instantaneous phase identification units 12-1 to 12-n with all possible phases. This is performed for a combination of instantaneous phases. For example, when four EEG signals of A1 to A4 are acquired, 16 combinations of instantaneous phases (a combination of the same locations such as A1-A1 are counted as one set, and A1-A2 and A2-A1 are counted. And the like, but the order of the combination is different and counted as an individual pair), the phase synchronization index is calculated for all of these combinations. Note that the PSI is 1 for a combination of the same portions, and the PSI value is the same for the same pair even if the order of the combination is different. Therefore, the PSI may be calculated for substantially six combinations (A1-A2, A1-A3, A1-A4, A2-A3, A2-A4, A3-A4).

  The symmetry index calculation unit 21 calculates a symmetry index indicating the degree of symmetry between information transmission hemispheres, that is, the degree of symmetry between information transmission in the right hemisphere and information transmission in the left hemisphere. The symmetry index calculation unit 21 calculates a symmetry index from the phase synchronization index output from the phase synchronization index calculation unit 13, and outputs the calculated symmetry index to the video signal generation unit 14. Then, the video signal generation unit (signal generation means) 14 generates a video signal representing a video whose contents change in conjunction with the symmetry index, and supplies the video signal to a display (not shown) to display the video.

[Method of calculating symmetry index]
Subsequently, a method of calculating the symmetry index will be described based on a specific example. As described above, the symmetry index is calculated using the phase synchronization index calculated by the phase synchronization index calculation unit 13, for example, the PSI defined by [Equation 1]. Here, an example will be described in which PSI is calculated from brain wave signals measured at five locations Fp1, Fp2, C3, C4, and Cz.

  In addition, Fp1, Fp2, C3, C4, and Cz indicate the measurement positions of the brain wave signal in the head as shown in FIG. As shown, Fp1 and Fp2, and C3 and C4 are symmetrical positions with respect to the median plane. Fp1 and Fp2 are near the frontal region, C3 and C4 are near the crown, and Cz is the midpoint (on the median plane) between C3 and C4.

  First, a matrix P having the PSI calculated by the phase synchronization index calculation unit 13 as a component is generated. When the measurement positions of the electroencephalogram signal are six as in this example, the matrix P is a matrix of 5 rows and 5 columns. The (1,1) component of the matrix P ′ is the PSI between the brain wave signal Fp1 and the brain wave signal Fp1, and the (1,2) component of the matrix P ′ is the difference between the brain wave signal Fp1 and the brain wave signal Fp2. PSI. Other components of the matrix P 'are defined similarly. An example of the matrix P thus defined is as follows.

  Next, a matrix Q is obtained by normalizing the matrix P such that the sum of the PSI values of each column or each row becomes 1. In the present embodiment, the matrix Q is obtained by normalizing so that the sum of the PSI values of each column becomes one. When the matrix P is as described above, the matrix Q is as follows.

  Subsequently, the matrix Q is subjected to eigenvalue decomposition.

Q = VDV ^-1
In the above formula, the matrix D is a matrix in which the eigenvalues of the matrix Q are arranged diagonally, and the matrix V is a matrix in which the eigenvectors of the matrix Q are arranged in each column.

  Next, with reference to the result of the eigenvalue decomposition, an eigenvector v of the matrix Q corresponding to the eigenvalue 1, that is, v that satisfies Qv = 1v is obtained. In terms of actual numerical calculation, the eigenvector v of the matrix Q corresponding to the eigenvalue closest to 1 is obtained. For example, when the diagonal component of the first column of the matrix Q is closest to 1, the column vector of the first column of the matrix V is the eigenvector v to be obtained, and the diagonal component of the second column of the matrix Q is closest to 1. , The column vector of the second column of the matrix V becomes the eigenvector v to be obtained.

  Furthermore, the importance vector v 'is obtained by normalizing the eigenvector v so that the sum of the components becomes 1. When the matrix Q is as described above, the importance vector v 'is as follows.

  Each component of the importance vector obtained in this way can be used as an importance index indicating the relative importance of information detection in the brain at each detection point of the corresponding brain wave signal. In particular, when each component of the importance vector obtained in this manner is used as an importance index, a high importance is assigned to a detection part having a high importance (a value of the importance index is large) and a detection part forming a network. (Importance index with a large value) can be assigned. In the above importance vector, the importance of Cz is the highest (0.215), and the importance of Fp2 is the lowest (0.188). In other words, in the information processing network in the subject's brain, the processing in which the Cz brain region cooperates with other brain regions occupies a relatively large weight, and the FP2 brain region cooperates with other brain regions. The numerical value indicates that the weight of the activated process is relatively small. Therefore, by using each component of the importance vector obtained as described above as an importance index, a symmetry that accurately indicates the degree of symmetry between information transmission in the right hemisphere and information transmission in the left hemisphere of the subject is obtained. An index can be calculated and a stimulus linked thereto can be given to the subject.

  Note that the method of calculating the eigenvector is not limited to eigenvalue decomposition, and it goes without saying that the eigenvector may be calculated by another calculation method.

  Here, the correlation between the spatial distribution of importance and the disease state index will be described with reference to FIGS. FIG. 18 is a diagram showing the spatial distribution of importance of healthy subjects and three patients (high FIMt, medium FIMt, low FIMt) having different pathological indices. Note that the PSI used for calculating the importance was calculated using the frequency f as the α band.

  As shown in the figure, the importance of a healthy person is such that a region of high importance spreads from the crown to the frontal region (around Fz in FIG. 17), and the spatial distribution of the importance is approximately relative to the median plane. It is symmetrical. The spatial distribution of the degree of importance of a patient with high FIMt is substantially symmetric with respect to the median plane, similarly to a healthy person. On the other hand, such a symmetry is broken in the spatial distribution of the importance of patients with middle FIMt and low FIMt.

  In order to confirm such a tendency, the number of subjects was increased, the importance of each subject was averaged, and the distribution was examined. The result is shown in FIG. There are 4 healthy subjects, 13 subjects with high FIM, 17 subjects with middle FIM, and 10 subjects with low FIM. As shown in the figure, the importance of the subject with low FIM is asymmetric with respect to the median plane, and the region of high importance is biased toward the right hemisphere.

  Such a tendency was commonly observed in subjects with damaged right hemisphere and subjects with damaged left hemisphere. That is, as shown in FIG. 20, when 13 subjects with a damaged left hemisphere and 26 subjects with a damaged right hemisphere are averaged and examined for their distributions, all of them have a relative median plane. The result was that the regions of high importance were skewed toward the right hemisphere.

  Further, as shown in FIG. 21, four subjects with particularly low FIMt (FIMt is 38, 42, 63, and 64, respectively) among subjects with damage to the left hemisphere, and FIMt among subjects with damage to the right hemisphere, The distribution was examined by averaging the importance of the four subjects who were particularly low (FIMt was 31, 19, 24, and 40, respectively). Also in this result, the spatial distribution of the importance of each subject group was asymmetric with respect to the median plane, and the region of high importance was biased toward the right hemisphere. It is a new finding that the right hemisphere becomes more important and the left hemisphere becomes less important even in a subject with an injury to the right hemisphere.

  The symmetry index used in the present embodiment is an index that numerically indicates the symmetry between the left and right hemispheres of the importance distribution as described above. Specifically, a difference in importance (left-right difference) at a position symmetric with respect to the median plane is calculated as a symmetry index. The difference in the importance index between the detection points symmetric with respect to the median plane is zero if the information transmission in the right hemisphere and the information transmission in the left hemisphere are completely symmetric, and becomes larger as the degree of asymmetry increases to some extent. Therefore, it can be used as a symmetry index indicating the degree of symmetry. In addition, if the difference in the importance index between the detection points symmetric with respect to the median plane is used as the symmetry index, the arithmetic processing for calculating the symmetry index becomes extremely simple.

[Correlation between symmetry index and disease state index]
Next, the correlation between the symmetry index and the disease state index will be described with reference to FIGS. In FIG. 22, electrodes (Fp1-Fp2, F3-F4, C3-C4) located symmetrically with respect to the median plane are paired, and the importance of each pair (in FIG. 22, denoted as "importance"). ) (In the figure, referred to as “left-right difference”; the same applies hereinafter) divided by the sum of the pairs, and the average value (symmetry index) for all pairs is used as the vertical axis, and the horizontal axis is used. FIG. 9 is a diagram plotting experimental results as FIMt and showing a correlation between a symmetry index and FIM. FIG. 23 is a diagram in which a vertical axis indicates a value (a symmetry index) obtained by dividing a difference in importance between the pairs F4 and F3 by a sum of the importance of the pair, and FIMt indicates a horizontal axis. FIG. 24 is a diagram in which the vertical axis indicates a value (symmetry index) obtained by dividing the difference in importance between the pairs C4 and C3 by the sum of the importance of the pair, and the horizontal axis indicates FIMt.

22 to 24, the smaller the symmetry index, the larger the FIM. In other words, the lower the FIM (the lower the degree of functional independence), the larger the symmetry index (the information transmission in the brain is biased). Also, the p-value indicating the degree of correlation was on the order of 10 ^-2 , indicating that there is a correlation between the symmetry index and the FIM. Therefore, as in the case of the phase synchronization index in the above embodiment, it is considered that it is possible to promote the improvement of the pathological index by making the subject recognize how to use the brain to reduce the symmetry index (reduce the asymmetry). . In particular, the value of p is extremely low between F4 and F3 shown in FIG. 23, and it can be seen that the difference in importance obtained between F4 and F3 is extremely suitable as a symmetry index.

  The increase or decrease of the symmetry index can be recognized by the subject by presenting a video as shown in FIG. However, the phase synchronization index has a positive correlation with the pathological index, while the symmetry index has a negative correlation with the pathological index. For this reason, the method of linking with the video may be reversed. For example, as shown in FIG. 3A, when presenting to a subject by changing the length of the object 21, when the symmetry index is reduced, the length of the object 21 is changed according to the degree of the reduction. May be longer. In this case, the subject is made to be conscious so that the length of the object 21 becomes long as in the above embodiment. Of course, the video to be displayed only needs to be such that the subject can recognize the increase or decrease in the symmetry index. For example, the object 21 may be lengthened when the symmetry index increases. In this case, the subject is made aware that the length of the object 21 is shortened. Alternatively, the subject may be made to recognize the increase or decrease in the symmetry index by a sensory stimulus other than visual.

  Note that the distribution of the importance of healthy persons is usually symmetric. Therefore, the similarity between the spatial distribution of the importance of the subject and the spatial distribution of the importance of the healthy person may be used as the symmetry index. For example, FIG. 25 is a diagram in which the vertical axis is the amount obtained by dividing the inner product of the importance vector of the healthy person and the importance vector of the subject by the product of the norms, that is, the plot of the experimental results is plotted with the horizontal axis being FIMt. . As shown, there is a positive correlation between the value of cos θ and FIMt, and it was confirmed that cos θ can be used as a relative index. Note that the value of cos θ approaches 1 as the angle θ between the importance vector of the healthy subject and the importance vector of the subject approaches zero. The fact that the angle θ formed by the importance vectors is close to zero means that the importance of both hemispheres is highly symmetric.

  Also, the similarity between the spatial distribution of the importance of the right hemisphere electrode and the spatial distribution of the importance of the left hemisphere electrode, for example, norm is the inner product of the importance vector of the right hemisphere electrode and the importance vector of the left hemisphere electrode. May be used as the symmetry index. The value of cos θ approaches 1 as the angle θ between the vector of the right hemisphere electrode and the importance vector of the left hemisphere electrode approaches zero. The fact that the angle θ formed by the importance vectors is close to zero means that the importance of both hemispheres is highly symmetric.

  Here, the method of evaluating the similarity between two importance vectors by the inner product of these two importance vectors has been described, but the present invention is not limited to this. That is, the similarity between the two importance vectors may be evaluated using another amount such as a correlation coefficient.

[Processing flow]
Subsequently, a flow of processing executed by the electroencephalogram signal processing device 2 will be described with reference to FIG. FIG. 26 is a flowchart illustrating an example of an electroencephalogram signal processing method executed by the electroencephalogram signal processing device 2.

  First, the electroencephalogram signal acquisition unit 11 acquires an electroencephalogram signal of the subject (S11). The acquired electroencephalogram signal may include an electroencephalogram signal measured at each of a plurality of locations of each hemisphere, but here, for simplicity, two locations of each hemisphere (F3, F4, C3, C4 ) Will be described in which four electroencephalogram signals (XR1, XR2, XL1, XL2) measured in step (1) are acquired. The electroencephalogram signal acquisition section 11 passes XR1 of the acquired electroencephalogram signals to the instantaneous phase identification section 12-1, and similarly passes the electroencephalogram signals XR2, XL1, XL2 to the instantaneous phase identification sections 12-2 to 12-4.

  Next, the instantaneous phase identification units 12-1 to 12-4 identify the instantaneous phase at the target frequency f from the brain wave signal passed from the brain wave signal acquisition unit 11 (S12). Note that the target frequency f is specified in advance by an operator or the like. Then, the instantaneous phase identification units 12-1 to 12-4 synchronize the instantaneous phases φR1 (f, t), φR2 (f, t), φL1 (f, t), and φL2 (f, t) which are respectively identified. The index is passed to the index calculation unit 13.

  Next, the phase synchronization index calculation unit 13 calculates a phase synchronization index from the instantaneous phases received from the instantaneous phase identification units 12-1 to 12-4 (S13). Specifically, 16 combinations are created from the four instantaneous phases received from the instantaneous phase identification units 12-1 to 12-4, and a phase synchronization index is calculated for each combination. Then, the phase synchronization index calculation unit 13 passes the calculated phase synchronization index to the symmetry index calculation unit 21 as a 4-by-4 matrix G '. Note that the process of setting the phase synchronization index to the matrix G 'may be performed by the symmetry index calculation unit 21.

  Next, the symmetry index calculation unit 21 calculates a symmetry index from the phase synchronization index received from the phase synchronization index calculation unit 13 (S14). Although not shown, the symmetry index is calculated by calculating the importance from the phase synchronization index (importance calculation step) and calculating the symmetry index from the calculated importance (symmetry index). Calculation step). That is, the symmetry index calculation unit 21 has a function as an importance index calculation unit and a symmetry index calculation unit. Of course, these means may be configured as individual functional blocks.

  In the importance calculation step, the symmetry index calculation unit 21 calculates a matrix G obtained by normalizing the phase synchronization index (here, a matrix G ′ of 4 rows and 4 columns), performs eigenvalue decomposition on the matrix G, and obtains a matrix V Get. Then, this is normalized to obtain a matrix V '. The components of the matrix V 'are eigenvectors (importance).

  Subsequently, in a symmetry index calculation step, the symmetry index calculation unit 21 calculates a symmetry index from the matrix V '. Note that the symmetry index may be a value obtained by averaging the difference in importance between the electrodes, which is a component of the matrix V ', in both hemispheres as shown in FIG. Alternatively, the difference in importance between two electrodes at positions symmetric with respect to the median plane as shown in FIGS. 23 and 24 may be used. Then, the symmetry index calculation unit 21 passes the symmetry index calculated as described above to the video signal generation unit 14.

  Thereafter, the video signal generation unit 14 generates a video signal according to the received phase synchronization index (S15, signal generation step), and supplies the generated video signal to a display connected to or incorporated in the brain wave signal processing device 1. I do. Thereby, an image based on the image signal is displayed on the display (S16).

  Then, the processing of S11 to S16 is repeatedly performed until the brain wave signal processing ends (YES is determined in S17). By this repetition, it is possible to transmit the result of the subject who has watched the video displayed on the display conscious of changing the video to the subject in substantially real time by the change of the video. Thus, the subject can recognize how to use the brain to increase the degree of synchronization, and can efficiently improve the degree of synchronization. Further, as described above, the degree of synchronization is correlated with the disease state index. Therefore, improvement of the degree of synchronization can be expected to improve the disease state.

  In the above embodiment, an example has been described in which a symmetry index indicating the degree of symmetry between information transmission in the right hemisphere and information transmission in the left hemisphere is calculated from the spatial distribution of importance in the subject's head. However, the symmetry index only needs to indicate the degree of symmetry between information transmission in the right hemisphere and information transmission in the left hemisphere, and is not limited to this example.

[Summary]
An electroencephalogram signal processing device according to one aspect of the present invention is an electroencephalogram signal processing device that sequentially obtains and processes electroencephalogram signals representing electroencephalograms of a subject, wherein the electroencephalogram signal at a first position of the subject's head, Phase synchronization index calculating means for successively calculating a phase synchronization index indicating the degree of phase synchronization with the electroencephalogram signal at the second position of the subject's head, and a phase synchronization index sequentially calculated by the phase synchronization index calculating means Or a signal corresponding to the time average of the phase synchronization index sequentially calculated by the phase synchronization index calculation means and sequentially supplied to the stimulating device, thereby interlocking with the phase synchronization index. Signal generating means for applying a stimulus that changes to the subject, wherein the first position and the second position are: (1) two different positions on the left hemisphere of the subject's head; (2) The subject's head Two different locations on the hemisphere, or, wherein the (3) position of the head midline of the subject, and a position on the head of a different said subject with the position it.

  Further, the electroencephalogram signal processing method according to one aspect of the present invention is an electroencephalogram signal processing method in which the electroencephalogram signal processing device sequentially acquires and processes electroencephalogram signals representing the electroencephalogram of the subject. A phase synchronization index calculating step of successively calculating a phase synchronization index indicating a degree of phase synchronization between the brain wave signal at the first position of the subject's head and the brain wave signal at the second position of the subject's head; A processing unit that outputs a signal corresponding to the phase synchronization index sequentially calculated in the phase synchronization index calculation step, or a signal corresponding to a time average of the phase synchronization index sequentially calculated in the phase synchronization index calculation step; Generating the stimulus and sequentially supplying the stimulus to the stimulus applying device to provide the subject with a stimulus that changes in conjunction with the phase synchronization index, wherein the first position and the second position Are (1) two different positions on the left hemisphere of the subject's head, (2) two different positions on the right hemisphere of the subject's head, or (3) on the midline of the subject's head. And a position on the head of the subject different from the position.

  According to the above configuration, the degree of phase synchronization between sequentially acquired electroencephalogram signals, in particular, LHPS (Left Hemispheric Phase Synchronization), RHPS (Left Hemispheric Phase Synchronization), in which a significant correlation is recognized with various pathological indicators, The subject can be fed back to recognize Right Hemispheric Phase Synchronization (Central Phase Synchronization) or CPS (Central Phase Synchronization). This allows the subject to control his / her brain with his / her own intention so that the degree of phase synchronization between brain wave signals is increased. By repeating such control, the subject can train his / her brain so that phase synchronization between brain wave signals is likely to occur.

  Further, in the electroencephalogram signal processing device according to one aspect of the present invention, from the change in the degree of phase synchronization between the electroencephalogram signals until the change is reflected on the stimulus given to the subject by the stimulus applying device. Preferably, the delay is 15 seconds or less.

  Also, in the electroencephalogram signal processing device according to one aspect of the present invention, the instantaneous phase is sequentially identified from the electroencephalogram signal at the first position of the subject's head, and the electroencephalogram at the second position of the subject's head. An instantaneous phase identifying means for sequentially identifying an instantaneous phase from the signal, wherein the phase synchronization index calculating means calculates the phase synchronization index from the instantaneous phase identified by the instantaneous phase identifying means, and the instantaneous phase identifying means Assuming that a window width used for specifying the instantaneous phase is T and each calculation time required for the instantaneous phase specifying means to calculate the instantaneous phase is τ, the instantaneous phase at the time t after the sampling of the value of the electroencephalogram signal at the time t is completed. Is preferably about the same as T / 2 + τ or T / 2 + 2 × τ.

  Further, in the electroencephalogram signal processing device according to one aspect of the present invention, the window width used by the phase synchronization index calculation means for calculating the phase synchronization index is T ′, and the calculation required by the phase synchronization index calculation means for calculation of the phase synchronization index is performed. It is preferable that the delay from completion of the specification of the instantaneous phase at time t to completion of the calculation of the phase synchronization index at time t is approximately equal to T ′ / 2 + τ ′, where τ ′ is the time.

  Further, in the electroencephalogram signal processing device according to one aspect of the present invention, the stimulus applying device is a display, and the signal generation unit generates a video signal and sequentially supplies the video signal to the display, thereby obtaining the phase synchronization index. It is preferable to provide the subject with a visual stimulus that changes in conjunction with the subject.

  Further, in the electroencephalogram signal processing device according to one aspect of the present invention, the stimulus applying device is a speaker, and the signal generation unit generates an audio signal and supplies the audio signal to the speaker, so that the phase synchronization index is increased. Preferably, the subject is provided with an auditory stimulus that changes in conjunction with the subject.

  Further, in the electroencephalogram signal processing device according to one aspect of the present invention, the stimulus applying device is an electrical stimulator or a mechanical stimulator, and the signal generating unit generates a signal to generate the signal and the electrical stimulator or the mechanical stimulus. It is preferable that the subject is given a somatosensory stimulus that changes in conjunction with the phase synchronization index by supplying the subject with the somatic sensory stimulus.

  In the electroencephalogram signal processing device according to one aspect of the present invention, the subject is a brain disease patient, and the subject is trained to train his / her brain so that phase synchronization in the brain of the subject is likely to occur. It preferably functions as a device.

  In the electroencephalogram signal processing device according to one aspect of the present invention, it is preferable that the first position and the second position are two different positions on the right hemisphere of the subject's head.

  In the electroencephalogram signal processing device according to one aspect of the present invention, it is preferable that the first position and the second position are two different positions on the left hemisphere of the subject's head.

  Further, in the electroencephalogram signal processing device according to one aspect of the present invention, the phase synchronization index calculation means uses each of the electroencephalogram signals obtained at least at three or more different positions on the subject's head by using each of the electroencephalogram signals. It is preferable to sequentially calculate a phase synchronization index according to the degree of phase synchronization between the two.

  An electroencephalogram signal processing device according to another aspect of the present invention is an electroencephalogram signal processing device that sequentially acquires and processes each electroencephalogram signal detected at a plurality of locations in each of a right hemisphere and a left hemisphere of a subject's head, From the phase synchronization index indicating the degree of phase synchronization between the respective brain wave signals, which is successively calculated from the respective brain wave signals, the relative position of each detection point of the brain wave signal in information transmission in the brain of the subject is calculated. Importance index calculating means for sequentially calculating an importance index indicating importance, and a distribution of the importance index in the subject's head, the degree of symmetry between information transmission in the right hemisphere and information transmission in the left hemisphere. The symmetry index calculating means for sequentially calculating the symmetry index shown, and the signal according to the symmetry index sequentially calculated by the symmetry index calculating means, or the signal is sequentially calculated by the symmetry index calculating means Signal generation means for generating a signal according to the time average of the symmetry index and sequentially supplying the signal to the stimulus applying device, thereby giving a stimulus which changes in conjunction with the symmetry index to the subject. It is characterized by the following.

  In the electroencephalogram signal processing method according to another aspect of the present invention, the electroencephalogram signal processing device sequentially acquires and processes each electroencephalogram signal detected at a plurality of locations in each of the right and left hemispheres of the subject's head. An electroencephalogram signal processing method, wherein the electroencephalogram signal processing device sequentially calculates, from the respective electroencephalogram signals, a phase synchronization index indicating a degree of phase synchronization between the electroencephalogram signals, to detect each of the detection points of the electroencephalogram signal. An importance calculation step of sequentially calculating an importance index indicating a relative importance in information transmission in the brain of the subject, and an electroencephalogram signal processing device, from the distribution of the importance index in the head of the subject, A symmetry index calculating step of sequentially calculating a symmetry index indicating a degree of symmetry between information transmission in the right hemisphere and information transmission in the left hemisphere, and the electroencephalogram signal processing device performs the symmetry index calculation step A signal corresponding to the symmetry index sequentially calculated in the above manner, or a signal corresponding to a time average of the symmetry index sequentially calculated in the symmetry index calculating step is generated and sequentially supplied to the stimulating device. A signal generating step of providing the subject with a stimulus that changes in conjunction with the symmetry index.

  According to the above configuration, each brain wave signal detected at a plurality of locations in the right and left hemispheres of the subject's head is sequentially calculated from each brain wave signal, and indicates the degree of phase synchronization between the brain wave signals. From the phase synchronization index, the importance index of each detection point is sequentially calculated. Note that a combination of detection positions with a high degree of phase synchronization can be said to have relatively high importance in information transmission in the brain, and thus the above-mentioned importance index can be calculated from the phase synchronization index.

  According to the above configuration, a symmetry index indicating the symmetry between information transmission in the right hemisphere and information transmission in the left hemisphere is sequentially calculated from the distribution of the importance index in the head of the subject. The “distribution” is a distribution of positions on the head.

  Here, as described above, information transmission in the brain of a healthy person is symmetric between the left and right hemispheres, but is asymmetric in a brain disease patient. For this reason, the above-mentioned symmetry index can be used for the training of the brain of the subject, similarly to the phase synchronization index in the left hemisphere.

  That is, by giving the subject a stimulus that changes in conjunction with the symmetry index calculated as described above, the subject is increased so that the symmetry index becomes larger (so that information transmission in the brain approaches symmetry). Can control the brain on their own. By repeating such control, the subject can train his / her brain so that information transmission in the brain is performed symmetrically in both hemispheres.

  Further, in the electroencephalogram signal processing device according to another aspect of the present invention, the importance index calculating means includes an eigenvector of a matrix obtained by normalizing a matrix having a component of a phase synchronization index indicating a degree of phase synchronization between the electroencephalogram signals. It is preferable to calculate an eigenvector whose eigenvalue is closest to 1, and use each component of the eigenvector as the importance index.

  Further, in the electroencephalogram signal processing device according to another aspect of the present invention, the symmetry index calculating means uses a difference of the importance index between detection points symmetric with respect to the median plane of the subject as the symmetry index. Preferably, it is calculated.

  The electroencephalogram signal processing device according to each aspect of the present invention may be realized by a computer, and in this case, the computer is operated as each unit (software element) included in the electroencephalogram signal processing device, whereby the electroencephalogram signal processing device is operated. The control program of the electroencephalogram signal processing device that realizes the above by a computer, and a computer-readable recording medium recording the control program also fall within the scope of the present invention.

[Appendix]
The present invention is not limited to the embodiments described above, and various modifications are possible within the scope shown in the claims, and embodiments obtained by appropriately combining technical means disclosed in different embodiments. Is also included in the technical scope of the present invention. Furthermore, new technical features can be formed by combining the technical means disclosed in each embodiment.

  The electroencephalogram signal processing device according to the present invention can be suitably used, for example, as a training device for enhancing a rehabilitation effect on a brain disease patient. Further, it can be used for general purposes for enhancing higher-order functions of the brain, and can also be used as a game machine or the like.

1, 2 EEG signal processing devices 12A, 12B, 12-n Instantaneous phase specifying unit (instantaneous phase specifying means)
13 phase synchronization index calculation unit (phase synchronization index calculation means)
14. Video signal generation unit (signal generation means)
21 Symmetry index calculation unit (importance index calculation means, symmetry index calculation means)
100 computer 140 auxiliary storage device (recording medium)
X1, X2 EEG signal Xn EEG signal PSI (f, t) Phase synchronization index

Claims (6)

An electroencephalogram signal processing device that sequentially acquires and processes each electroencephalogram signal detected at a plurality of locations in each of the subject's head right hemisphere and left hemisphere,
From the phase synchronization index indicating the degree of phase synchronization between the brain wave signals, which has been successively calculated from the brain wave signals, the relative importance of each detection point of the brain wave signal in information transmission in the brain of the subject is calculated. Importance index calculating means for sequentially calculating the importance index indicating the degree,
From the distribution of the importance index in the head of the subject, a symmetry index calculating means for sequentially calculating a symmetry index indicating a degree of symmetry between information transmission in the right hemisphere and information transmission in the left hemisphere,
A signal corresponding to the symmetry index sequentially calculated by the symmetry index calculation means, or a signal corresponding to a time average of the symmetry index sequentially calculated by the symmetry index calculation means is generated. By sequentially supplying to the stimulus imparting device, signal generation means for giving the subject a stimulus that changes in conjunction with the symmetry index,
An electroencephalogram signal processing device comprising: The importance index calculating means calculates an eigenvector of a matrix obtained by normalizing a matrix having a component of a phase synchronization index indicating a degree of phase synchronization between brain wave signals, and an eigenvector closest to 1 is calculated. Using each component of the eigenvector as the importance index,
The electroencephalogram signal processing device according to claim 1, wherein: The symmetry index calculating means calculates a difference of the importance index between detection points symmetric with respect to the median plane of the subject as the symmetry index,
The electroencephalogram signal processing device according to claim 1 or 2, wherein: An electroencephalogram signal processing method in which the electroencephalogram signal processing device sequentially acquires and processes each electroencephalogram signal detected at a plurality of locations in each of the subject's head right hemisphere and left hemisphere,
From the phase synchronization index indicating the degree of phase synchronization between the brain wave signals, the brain wave signal processing device sequentially calculates the brain wave signals from the brain wave signals. An importance calculating step of sequentially calculating an importance index indicating relative importance in transmission,
Symmetry index calculation in which the electroencephalogram signal processing device sequentially calculates a symmetry index indicating a degree of symmetry between information transmission in the right hemisphere and information transmission in the left hemisphere from the distribution of the importance index in the head of the subject. Steps and
The electroencephalogram signal processing device, according to the signal according to the symmetry index sequentially calculated in the symmetry index calculation step, or according to the time average of the symmetry index sequentially calculated in the symmetry index calculation step A signal generation step of controlling the stimulus providing device to provide the subject with a stimulus that changes in conjunction with the symmetry index by sequentially generating and supplying a signal to the stimulus providing device ,
An electroencephalogram signal processing method comprising:   A program for causing a computer to function as the electroencephalogram signal processing device according to any one of claims 1 to 3, wherein the program causes the computer to function as each of the units.   A computer-readable recording medium on which the program according to claim 5 is recorded.

Images