Brain activity signal measurement system

Abstract

This research aims to establish and apply analytical techniques for measuring brain activity using resting-state functional magnetic resonance imaging (rs-fMRI), which is widely used in research on brain diseases and cognition. [Solution] The brain activity signal measurement system is characterized by comprising: a signal acquisition unit 10 that acquires a plurality of time-series signals with different amplitudes using a sensor or receiving coil; a frequency conversion unit 20 that converts the time-series signals acquired by the signal acquisition unit 10 into the frequency domain; and a signal identification unit 30 that identifies frequency components corresponding to the time change of signal amplitude in the frequency domain converted by the frequency conversion unit 20.

Description

Technical Field

[0001] The present invention relates to a brain activity signal measurement system for measuring brain activity signals.

Background Art

[0002] Conventionally, an MRI apparatus used for measuring brain function by magnetic resonance imaging (MRI) consists of a static magnetic field generator, a gradient magnetic field generator, and a signal collection device (transceiver coil) as shown in the following Document 1 and the like.

[0003] In brain function measurement, since the discovery of fMRI (BOLD method) around 1990, stimulation presentation has been used to activate the brain in order to measure brain activity (tb-fMRI). This is because the activity of the brain regions activated in response to the presented stimuli is detected by fMRI.

[0004] The method of presenting a task and measuring brain activity is widely used, but mainly research results such as comparisons between groups, for example, comparisons between patients with brain mental diseases and healthy subjects, are numerous. At the individual level, it is limited to some limited clinical applications such as measuring motor function and identifying the central sulcus. There are several reasons for this, but the two major problems are (1) One of them is that there is variation in the brain activity of the measurement target (subject) with respect to the stimulus (or task). (2) The other is the signal-to-noise ratio (SNR) required for analyzing brain activity at the individual level. The brain activity data at the individual level has a much lower SNR compared to the brain activity data at the group level. Regarding the above (1), resting-state fMRI (rs-fMRI) that emerged around 2004 can measure brain activity without using a task, so it is expected to be applied to clinical settings and the like. In the case of comparison with healthy subjects at the group subject level, mental diseases such as dementia and ASD can be detected with high accuracy, and application to the detection and diagnosis of mental diseases in individuals is expected. (2) Regarding this, several analysis methods have been proposed, but a more effective method is desired.

Prior Art Documents

[0005] [Patent Document 1] Japanese Patent Publication No. 2007-190352 [Overview of the project] [Problems that the invention aims to solve]

[0006] (1) Conventional rsfMRI could not guarantee that the time-series signals measured by the MRI device were bold signals. Signals in the bandwidth of approximately 0.01 Hz to approximately 0.1 Hz are mainly used as rsfMRI signals, and indicators such as brain functional networks obtained by further analysis of these signals are used for clinical applications.

[0007] In view of the above circumstances, the present invention aims to establish a method for identifying bold signals and to apply it to the analysis of brain activity measurements using resting-state functional magnetic resonance imaging (rs-fMRI), which is widely used in the study of brain diseases and cognition. (2) The objective is to establish a method for measuring dynamic brain activity signals with a high signal-to-noise ratio using various measurement techniques that analyze brain activity, including MRI. [Means for solving the problem]

[0008] The brain activity signal measurement system of the first invention is a brain activity signal measurement system that measures brain activity signals, sensor or multiple Using a receiving coil, From multiple detection systems with different sensitivities, the same object being measured is detected. A signal acquisition unit that acquires multiple time-series signals with different amplitudes, A frequency conversion unit that converts the plurality of time-series signals acquired by the signal acquisition unit into the frequency domain, In the frequency domain converted by the frequency conversion unit, The relative relationship of signal amplitudes originating from multiple detection systems is evaluated to determine whether it is consistent with the relationship that arises depending on the sensitivity of the detection system, and the frequency components that are consistent are identified as brain-derived components. Signal identification unit and It is characterized by having the following features.

[0009] According to the brain activity signal measurement system of the first invention, for example, By acquiring various signals, including resting-state functional magnetic resonance imaging (rs-fMRI) signals, as multiple time-series signals with different amplitudes, converting the acquired time-series signals into the frequency domain, and identifying each frequency component corresponding to the time change in signal amplitude within the converted frequency domain, it is possible to realize an analysis method for brain activity measurements, including rs-fMRI.

[0010] The brain activity signal measurement system of the second invention is, in the first invention, The aforementioned sensitivity is, The shape, arrangement, or shape of the receiving coil. Distance to the object being measured It is a sensitivity derived from, The signal identification unit identifies brain-derived components by utilizing the fact that signals originating from the same measurement target exhibit a relative amplitude relationship corresponding to the sensitivity in each receiving coil. Characterized by

[0012] 2nd The brain activity signal measurement system of this invention can extract signals originating from brain regions (not necessarily BOLD). This method is a novel approach that can reduce noise contained in MRI signals. In brain research, in addition to the commonly used fMRI (BOLD method), numerous dynamically changing signals related to brain activity are measured (ASL: arterial spin labeling, VASO: vascular space-occupancy, functional diffusion, neuronal current imaging, etc.). The method of this invention can be applied to all of these dynamic signal measurements.

[0013] Specifically, in the case of magnetic resonance imaging (MRI), most current MRI machines use multi-channel receiving coils, and it is possible to utilize the differences in sensitivity that arise from the positional relationship between the brain region being measured and the receiving coil. By utilizing this, multiple time-series signals with different amplitudes can be acquired, the acquired time-series signals can be converted into frequency components, the frequency components originating from the brain can be identified based on the information from the receiving coil, and then converted back into the time domain to extract signals originating from the brain.

[0014] For example, in the case of a multi-channel receiving coil, signals with different amplitudes can be obtained by using the distance information (sensitivity) from the target brain region to each coil. It is characterized by using this distance information for extracting signals derived from the brain. If the amplitude of the signals from each coil at each frequency point discretized in the frequency domain is semi-proportional to the distance from the coil, it is identified as a signal derived from the brain. By converting the identified signals in the frequency domain back to the time domain, a dynamic signal derived from the brain can be obtained.

[0015] The brain activity signal measurement system according to the third invention, in the second invention, The signal identification unit is characterized by identifying brain-derived components by evaluating whether the signal amplitude of each receiving coil satisfies a predetermined sequential relationship based on the distance from the measurement target to each receiving coil or the coil sensitivity corresponding to that distance at each individual frequency point on the frequency axis. The fourth invention The brain activity signal measurement system, in any one of the first to third inventions, The signal identification unit is characterized by identifying and removing components corresponding to frequency points that do not satisfy the predetermined relationship as noise components. The 5 brain activity signal measurement system according to the invention, in the first invention, in the case of electroencephalogram measurement, the signal acquisition unit is characterized in that a plurality of electrodes with different cross-sectional areas are arranged, and each electrode has a different resistance value, thereby acquiring a plurality of time-series signals with different amplitudes.

[0016] The 5 According to the brain activity signal measurement system of the invention, as signal acquisition, a plurality of time-series signals with different amplitudes are acquired from a plurality of electrodes (each electrode having a different resistance value) with different cross-sectional areas of the electrodes, frequency conversion is performed, and for each point in each frequency domain, those in which the amplitude of the signal is semi-proportional to the resistance are identified as components derived from the brain, and those obtained by converting them back to the time domain again can be extracted as components derived from the brain.

[0017] The 6 brain activity signal measurement system according to the invention, in the first invention, the signal acquisition unit is characterized in that in near-infrared spectroscopy (NIRS), a plurality of time-series signals with different amplitudes are acquired by adjusting the arrangement of the light source or the receiving sensor and the optical path length.

[0018] The 6According to the brain activity signal measurement system of the invention, specifically in the case of near-infrared spectroscopy (NIRS), by adjusting the arrangement of the light source or receiving sensor and the optical path length, multiple time-series signals with different amplitudes can be acquired, frequency components can be identified from the acquired time-series signals, and brain-derived components can be extracted based on the identified frequency components.

[0019] The 7 The brain activity signal measurement system of the invention, in the first invention, The signal acquisition unit is characterized by acquiring multiple time-series signals with different amplitudes in magnetoencephalography (MEG) by adjusting the sensitivity or arrangement of magnetic field sensors.

[0020] The 7 According to the brain activity signal measurement system of the invention, by adjusting the sensitivity or arrangement of magnetic field sensors in relation to a magnetoencephalogram, multiple time-series signals with different amplitudes can be acquired, frequency components can be identified from the acquired time-series signals, and brain-derived components can be extracted based on the identified frequency components. [Brief explanation of the drawing]

[0021] [Figure 1] A system configuration diagram showing the configuration of the brain activity signal measurement system according to the present invention. [Figure 2] Figure 1 is an explanatory diagram showing the processing steps of the brain activity signal measurement system. [Figure 3] Figure 1 is an explanatory diagram showing other processing examples of the brain activity signal measurement system. [Modes for carrying out the invention]

[0022] Hereinafter, an embodiment of the brain activity signal measurement system according to the present invention will be described with reference to the drawings. Note that a detailed explanation of the signal acquisition principle of the MRI device and related diagrams are disclosed in Patent Document 1 and other documents, so a detailed explanation will be omitted here.

[0023] As shown in Figure 1, the brain activity signal measurement system of this embodiment is a brain activity signal measurement system for measuring brain activity signals, and comprises a signal acquisition unit 10, a frequency conversion unit 20, and a signal identification unit 30.

[0024] The signal acquisition unit 10 acquires the sensing value output via the sensing unit X of the sensor or receiving coil as multiple time-series signals with different amplitudes.

[0025] The frequency conversion unit 20 converts the time-series signal acquired by the signal acquisition unit 10 into the frequency domain.

[0026] The signal identification unit 30 identifies the frequency components corresponding to the time variation of the signal amplitude in the frequency domain converted by the frequency conversion unit 20.

[0027] The brain activity indicators, such as frequency components, identified in this way are output to a display unit Y, such as a screen.

[0028] Next, referring to Figure 2, we will describe signal acquisition and analysis using a multi-echo sequence in rs-fMRI as a first embodiment.

[0029] As shown in Figure 2(a), the signal acquisition unit 10 acquires signals using an MRI device at different echo times (TE1, TE2, TE3). Here, the echo times are TE1 (short echo time): For example, 9ms. It has the characteristic of having a small signal amplitude (after normalization). TE2 (Intermediate Echo Time): For example, 20ms. The signal amplitude (after normalization) exhibits intermediate characteristics. TE3 (Long Echo Time): For example, 49ms. This characteristic is associated with a large signal amplitude (after normalization).

[0030] The signal acquisition unit 10 utilizes the multi-echo sequence function of the MRI device to acquire signals corresponding to different echo times at the same time point. For example, it acquires blood oxygen concentration-dependent (BOLD) signals in specific areas of the brain (e.g., frontal lobe, occipital lobe).

[0031] Next, the frequency conversion unit 20 performs a Fourier transform on the time-series signals TE1, TE2, and TE3 acquired by the signal acquisition unit 10, converting them to the frequency domain. Specifically, as shown in Figure 2(b), in the frequency domain, a characteristic appears in which the signal amplitude increases monotonically with increasing echo time.

[0032] Next, the frequency conversion unit 20 analyzes the frequency components by comparing the signal amplitudes of each echo time and identifies frequency components that satisfy the TE-dependent criterion (TE1 < TE2 < TE3).

[0033] Next, the signal identification unit 30 identifies frequency components that satisfy the TE dependency criterion as bold components. Components that do not satisfy the TE dependency criterion are automatically removed as noise components.

[0034] In the first embodiment described above, signal acquisition and analysis using multi-echo sequences in functional MRI (tb-fMRI) during task performance and functional MRI (rs-fMRI) at rest improves the accuracy of brain activity measurement by precisely extracting bold components. Since noise components are reduced, the signal-to-noise ratio (SNR) is significantly improved. Clinically, this is useful for the early diagnosis of brain diseases such as Alzheimer's disease and schizophrenia.

[0035] Next, as a second embodiment, in the first embodiment, the signal acquisition unit 10 may adopt a configuration in which it acquires multiple time-series signals with different amplitudes by adjusting the sensitivity of the receiving coil of the magnetic resonance imaging (MRI) apparatus by the shape, arrangement, or other method of the receiving coil, or by utilizing the sensitivity derived from the inherent properties of the receiving coil.

[0036] In measurements using a standard single echo sequence (not a multi-echo sequence), signals from multiple receiving coils are utilized. Most modern MRI systems used in clinical practice and research employ multi-channel receiving coils consisting of multiple coils. Spatial resolution in fMRI measurements is approximately isotropically 3 mm for 3T MRI and approximately isotropically 1 mm for 7T MRI. These finely divided regions are called voxels. Considering the example of a 64x64 matrix with 30 slices covering the entire brain in a 3T system, signals can be obtained from 122,800 voxels. Furthermore, signals can be obtained from multiple receiving coils within each individual voxel.

[0037] Specifically, Figure 3 shows the case with three receiving coils. A signal from a single voxel located in the posterior part of the brain is obtained from three coils. In the usual method, these three signals are integrated in an appropriate way to create a single signal (this integration can also reduce noise). In contrast, the proposed method uses the signal strength of the DC component of the three signals (s1_dc < s2_dc < s3_dc). This is because the signal becomes stronger the closer it is to the voxel being measured. The strength of the AC component should be the same as that of the DC component, so after frequency conversion, a noise-free AC signal can be obtained by picking up AC(f) that satisfies the DC strength condition at each frequency point. For example, in the case of white noise originating from the coil, it has the same strength in all frequency bands regardless of the distance from the coil, so it is removed as noise. When using such multi-channel receiving coils, it can be applied without special hardware or ingenuity in the measurement sequence.

[0038] For example, it can be used by incorporating it into the image reconstruction algorithm attached to the MRI device. Alternatively, it can be used for post-processing by using raw data (data in its sampled state: data before imaging) after image acquisition. The advantage of the method in this invention is that it is not affected by deviations arising from differences between MRI manufacturers or differences between measurement facilities. Of course, the development of receiving coils to make the proposed method more effective can also be considered. Thus, according to the second embodiment, it is also effective for rs-fMRI and tb-fMRI in removing noise. That is, it is possible to extract signals originating from the brain (which may also include physiological signals that are not necessarily bold), and remove noise. Furthermore, it has a high potential for practical application.

[0039] Next, as a third embodiment, we will describe an example of acquiring signals with different amplitudes through sensor design in EEG (electroencephalography). In EEG, signals with different amplitudes are acquired by designing the sensor as follows.

[0040] Specifically, three electrodes are placed on the scalp. Each electrode is designed to have a different cross-sectional area and different resistance values, for example, 1Ω, 2Ω, and 3Ω. Electrode 1 (1Ω): Acquires signals with small amplitude. Electrode 2 (2Ω): Acquires the signal at the midpoint of the amplitude. Electrode 3 (3Ω): Acquires signals with large amplitude.

[0041] Here, the time-series signals acquired by the signal acquisition unit 10 are obtained from each electrode as follows. S1 = S0 + N (small amplitude) S2 = S02 + N (mid amplitude) S3 = S03 + N(large amplitude) Here, S0 represents the brain activity signal, and N represents the noise component.

[0042] Next, the frequency conversion unit 20 performs a Fourier transform on the time-series signals acquired by each electrode, converting them to the frequency domain. As a result, in the frequency domain, the components related to S0 appear in order of amplitude.

[0043] Next, the frequency conversion unit 20 removes noise components and extracts frequency components corresponding to the brain activity signal (S0).

[0044] Next, the signal identification unit 30 identifies frequency components that satisfy the TE-dependent criterion as BOLD components, as part of the extraction of components originating from the brain.

[0045] In the third embodiment described above, it is possible to reduce noise specific to EEG signals (e.g., signals originating from electromyography and environmental noise). Furthermore, because highly accurate brain activity signals can be obtained, it is effective in diagnosing brain diseases such as epilepsy.

[0046] Next, as a fourth embodiment, we will describe the cases of NIRS (Near-Infrared Spectroscopy) and MEG (Magnetoencephalography).

[0047] In the case of NIRS, multiple detectors are placed in a narrow area to be measured, and signals with different amplitudes are acquired by changing the gain of the detectors.

[0048] On the other hand, in the case of MEG, multiple magnetic field sensors are placed in a narrow measurement area, and signals with different amplitudes are acquired by adjusting the sensitivity. Sensor 1: Low sensitivity (small amplitude) Sensor 2: Medium sensitivity (medium amplitude) Sensor 3: High sensitivity (large amplitude) The signal acquisition unit 10 then acquires the magnetic field signals measured by each sensor as time-series data.

[0049] Next, the frequency conversion unit 20 converts the acquired signal into the frequency domain and performs analysis of the frequency components corresponding to the time change of the signal amplitude.

[0050] Next, the signal identification unit 30 identifies the frequency components that appear in order of amplitude, and extracts the components related to brain activity signals.

[0051] In the fourth embodiment described above, NIRS enables high-precision measurement of changes in blood oxygenation and deoxygenation, and MEG enables noise-free measurement of weak magnetic field changes in the brain.

[0052] As explained in detail above, the brain activity measurement system of this embodiment makes it possible to establish an analysis method for measuring brain activity using resting-state functional magnetic resonance imaging (rs-fMRI), which is widely used in the study of brain diseases and cognition, and can be applied to EEG, NIRS, and MEG.

[0053] Furthermore, in the first to fourth embodiments described above, the system may be configured to present a stimulus and measure the brain activity in response to it. For example, in a method of presenting a light stimulus, since the brain activity signal increases with the brightness of the light, the presentation of three different types of stimuli with varying light brightness may be repeated periodically to obtain signals with different amplitudes from the brain activity signal itself, which is detected by the sensing unit X of the first to fourth embodiments. In this case, multiple signals with different amplitudes can be measured using a normal measurement method, and only the brain activity can be picked up by frequency conversion. [Explanation of symbols]

[0054] 10...Signal acquisition unit, 20...Frequency conversion unit, 30...Signal identification unit, X...Sensing unit, Y...Display unit.

Claims

1. A brain activity signal measurement system that measures brain activity signals, A signal acquisition unit that acquires multiple time-series signals with different amplitudes originating from the same object to be measured from multiple detection systems with different sensitivities using sensors or multiple receiving coils, A frequency conversion unit that converts the plurality of time-series signals acquired by the signal acquisition unit into the frequency domain, A signal identification unit evaluates whether the relative relationship of signal amplitudes originating from multiple detection systems in the frequency domain converted by the frequency conversion unit matches the relationship that arises according to the sensitivity of the detection system, and identifies the matching frequency component as a brain-derived component. A brain activity signal measurement system characterized by comprising the following features.

2. In the brain activity signal measurement system according to claim 1, The aforementioned sensitivity is a sensitivity that originates from the shape, arrangement, or distance of the receiving coil to the object being measured. A brain activity signal measurement system characterized in that the signal identification unit identifies brain-derived components by utilizing the fact that signals originating from the same object of measurement show a relative amplitude relationship in each receiving coil according to the sensitivity.

3. In the brain activity signal measurement system according to claim 2, The brain activity signal measurement system is characterized in that the signal identification unit identifies brain-derived components by evaluating whether the signal amplitude of each receiving coil satisfies a predetermined sequential relationship based on the distance from the measurement target to each receiving coil or the coil sensitivity corresponding to that distance at each individual frequency point on the frequency axis.

4. In the brain activity signal measurement system according to any one of claims 1 to 3, The brain activity signal measurement system is characterized in that the signal identification unit identifies and removes components corresponding to frequency points that do not satisfy the predetermined relationship as noise components.

5. In the brain activity signal measurement system according to claim 1, in the case of electroencephalography, The brain activity signal measurement system is characterized by the arrangement of multiple electrodes with different cross-sectional areas, each electrode having a different resistance value, thereby acquiring multiple time-series signals with different amplitudes.

6. In the brain activity signal measurement system according to claim 1, The brain activity signal measurement system is characterized in that the signal acquisition unit acquires multiple time-series signals with different amplitudes by adjusting the arrangement of the light source or receiving sensor and the optical path length in near-infrared spectroscopy (NIRS).

7. In the brain activity signal measurement system according to claim 1, The brain activity signal measurement system is characterized in that the signal acquisition unit acquires multiple time-series signals with different amplitudes by adjusting the sensitivity or arrangement of magnetic field sensors in magnetoencephalography (MEG).

Images