JP2690678B2 - Approximate model display device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a biomagnetism measuring device or the like for measuring a magnetic field generated from a living body or the like and estimating a signal source of the magnetic field such as an active site of the living body.

[0002]

2. Description of the Related Art Conventionally, SQUI has been used as a highly sensitive magnetic sensor.
D sensor is known (SQUID is a superconducting quantum interference magnetometer
That is). A biomagnetism measuring device is a device that measures weak magnetism generated from a human body using this SQUID sensor. Measuring the magnetic field from the heart and brain, the active site (signal source)
Is estimated and applied to clinical medicine such as arrhythmia and epilepsy.

Here, the principle of the biomagnetism measuring device is shown in FIG.
This will be briefly described with reference to FIG.

That is, in FIG. 8, 101 is a brain activity current, 102 is a brain magnetic field, 103 is a SQUID sensor, and 104.
Is liquid helium.

When a person thinks about something or receives an external stimulus, nerve cells in the cerebrum are excited and a current flows in the brain.
When such a brain activity current 101 flows, a brain magnetic field 102 based on the "right-handed screw law" is generated. The above SQUID sensor 103 measures this brain magnetic field. Since the brain magnetic field is extremely weak and is several hundred millionth of the earth's magnetism, and can be detected only by the super-sensitive magnetic sensor called SQUID that utilizes the superconducting phenomenon at the liquid helium temperature, the liquid helium 104 is filled. The SQUID sensor 103 is arranged in the container.

In the above-mentioned configuration, inverse problem analysis is used as a general method to estimate the signal source corresponding to the electrophysiologically active site in the brain from the measured brain magnetic field. As a result, the position of the current source can be specified in the coordinate system of the SQUID sensor 103.
It is also possible to estimate multiple signal sources simultaneously.
In this way, if the active site of the brain is known, it can be diagnosed whether the function of the brain is working normally, where it is abnormal, or the like.

That is, according to the inverse problem analysis method, in order to measure a magnetic field generated from a selected part of a living body and estimate a signal source by using a biomagnetism measuring device, for example, a conductive sphere model having a uniform conductivity is used. It is necessary to perform the analysis using the Grynspan-Geselowitz equation, etc., which considers the influence of volume current in. Therefore, it is necessary to create an approximate model that approximates the selected part of the living body.

Therefore, an approximate model is created by a method of directly measuring the shape of the head surface from the outside using a digitizer or the like, and the center position of the approximate model in the coordinate system of the SQUID sensor 103 is determined by estimation.

Using the approximate model determined in this way, the theoretical magnetic field value from the current dipole assumed as a signal source existing in the approximate model and the actual magnetic field measured by the ultrasensitive magnetic sensor The signal source can be estimated by calculating the position, size, and direction of the current dipole that minimizes the sum of the squared errors of the values.

By the way, in the magnetic field measurement of the brain, there is a method of estimating a signal source by assuming that the head is a sphere or a model close to an actual skull shape (Jan WH Meijis and Maria).
J. Peters, "Various models of the head and their in
fluence on MEG's "Biomagnetism '87, p102-105).

In measuring the magnetic field of the heart, there is a method of estimating the signal source by assuming that the chest is a flat plate model or a torso model close to the actual shape (Adriaan van Ooster).
om, et al., "Contribution maps in magnetcardiograhy,
Advances in Biomagnetism, Plenum publishing Co.,
p343-348 ").

[0012]

However, in the above configuration, the estimation accuracy of the current dipole by the inverse problem analysis is greatly affected by the position, size and shape of the created approximate model. However, because the positional relationship between the selected part of the living body and the created approximate model is not accurate, how closely the created approximate model matches the actual living body, that is, the relative position, size, shape, etc. of each There was a problem that it was impossible to grasp the relationship.

The present invention has been made in view of the above problems of the conventional biomagnetism measuring device, and an object thereof is to provide an approximate model display device capable of accurately grasping the relative positional relationship between the approximate model and the living body. To do.

[0014]

Means for Solving the Problems The present invention of claim 1, raw
For use in estimating at least the position of a signal source that is present in the measurement target of the body and generates a magnetic field, an approximate model creation means for creating an approximate model of the measurement target based on the tomographic image of the measurement target, A combining unit that combines the created approximate model and the tomographic shape, a display unit that can display the combined image, and the combined image.
Change to change the approximation model for image data
To confirm the indicating means and its modified approximate model
And a confirmation instructing means, and the tomographic images are orthogonal to each other.
It is an image viewed from three directions.
Images that combine the layer image and the approximate model can be displayed simultaneously.
When the synthesizing means uses the change instruction means,
The approximate model of the synthesized image is changed.
And, in conjunction with it, the approximate model of other combined images
It is an approximate model display device that can be changed .

The present invention according to claim 2 is within a measurement object of a living body.
Estimate at least the location of a signal source that is present and that produces a magnetic field
Selecting means for selecting an initial approximation model to be used when
And the disconnection of the selected initial approximation model and the measurement target.
Displays a composite method for combining the layer image and the combined image
Comprising a display unit configured to, a change instruction means for changing the size and position of the initial approximation model for the combined image, and a settlement instruction means for determining the changed approximate model, the tomographic Images are three directions orthogonal to each other
The image is viewed from above, and the display means displays each tomographic image and the front image.
It is possible to display an image that is composed of the approximate model and
The synthesizing means may be any of the above-mentioned ones according to the change instruction means.
When the approximate model of the combined image is changed, it
Change the approximate model of other combined images in conjunction with
It is an approximate model display device capable of performing.

[0016]

According to the first aspect of the present invention, the approximate model creating means uses the approximate model of the measurement target for use in estimating at least the position of the signal source existing in the measurement target of the living body and generating the magnetic field. It is created based on the tomographic image of the measurement target. The synthesizing unit synthesizes the created approximate model and the tomographic shape, the display unit displays the synthesized image, and the change instructing unit compares the synthesized image data.
Then, the approximation model is changed, and the confirmation instruction means,
The modified approximation model is determined, and the tomographic images are
The image is viewed from three directions orthogonal to the
Is an image obtained by synthesizing each tomographic image and the approximate model.
It can be displayed at the same time, and the synthesizing means can display the change instructing means.
Therefore, the approximation model of the composited image
If it is changed, it will be linked with other composite images.
The approximation model can also be changed .

[0017]Claim 2In the present invention,Within the measurement target of the living body
At least the position of a signal source present in
Means to select the initial approximation model to be used whenChoice
AndThe synthesizing means includes the selected initial approximation model and the
Synthesize with tomographic image of measurement targetAndThe display means was synthesized
Display the image,The change instructing means adds to the combined image
In contrast to the initial approximation modelSize and positionChanged
The confirmation instruction means confirms the changed approximate model.
However, the tomographic image is an image viewed from three directions orthogonal to each other.
Yes, the display means displays each tomographic image and the approximate model.
It is possible to simultaneously display an image in which and are combined, and the combining means,
Any one of the above by the change instruction means Composite image
When the approximate model of the image is changed, other
You can also change the approximate model of the composite image of
You.

[0018]

An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram for explaining the structure of the approximate model display device in this embodiment, and the structure will be described with reference to FIG.

That is, in the figure, 11 is a SQUID sensor for measuring the magnetic field from the signal source, 12 is a data collecting section for collecting the data measured by the SQUID sensor 11, and 13 is an approximate model, Data for combining with data from a tomographic imaging device 18 of an external device described later and for estimating a signal source (estimation of position, direction, size, etc. of the signal source) based on the measurement result using the SQUID sensor 11. A processing / control unit, 14 is a display unit as a display unit of the present invention for displaying an approximate model, a tomographic image and the like, and 15 is a change instruction unit of the present invention for instructing a change of the approximate model. Is a change instruction unit, 16 is a confirmation instruction unit as the confirmation instruction means of the present invention for finally confirming the approximate model, and 17 is a recording unit for recording various data by a printer or the like. A. Reference numeral 18 is a tomographic imaging device as an external device for taking a tomographic image of the head of the subject. Further, the data processing / control unit 13 includes the approximate model creating means and the synthesizing means in the present invention.

Next, referring to the flow chart shown in FIG. 2, the operation of the approximate model display device of this embodiment when a tomographic image of the head of a subject is used to create a biological approximate model and the result is displayed. And explain.

As "step 21", the tomographic imaging apparatus 1
8 (MRI device) captures a tomographic image of the head.

That is, a large number of tomographic images of the subject's head are picked up by the tomographic image pickup device 18, and these tomographic image data are taken into the data processing / control section 13. The more the number of tomographic images is, the better in order to create an approximate model and improve the display accuracy in image creation. However, since the measurement time is long, about 128 images are normally taken. The tomographic image pickup device 18 is preferably an MRI device, but it is also possible to use another tomographic image pickup device such as a CT device (note that a tomographic image pickup device as an in-brain image analysis device has a MRI device as a typical device). There is a CT device, the former is suitable for imaging brain tissue,
The latter is suitable for skeleton photography).

As "step 22", the data processing / control section 13 creates an image representing the shape of the head from the tomographic image obtained by the tomographic imaging device 18.

Here, the image creation result will be described with reference to FIGS. However, each figure shows a state in which the images of the approximation model (sphere model) described later have already been combined, but at the stage of step 22, the images of these approximation models have not been created yet.

That is, with respect to a tomographic image viewed from a predetermined direction (eg, sagittal direction), tomographic images in other directions (eg, coronal direction and axial direction) orthogonal thereto are created. 3 to 5 explain the image creation result. Reference numeral 51 is a head tomographic image in the sagittal direction (see FIG. 3), 52 is a head tomographic image in the coronal direction (see FIG. 4), and 53 is a head tomographic image in the axial direction (see FIG. 5). , 56 are approximate models.

Further, two sectional directions (eg, sagittal direction and coronal direction) arbitrarily selected by the operator in each tomographic image viewed from three directions are combined to create a composite image in a three-dimensional space. . FIG. 6 shows a composite image in the three-dimensional space created in this way. Reference numeral 54 is a composite tomographic image of a cross-sectional view viewed from two directions.

Further, the edge of the scalp is extracted from the tomographic image,
Create a 3D surface model. FIG. 7 shows a head epidermis image thus created. 55 is a scalp image.

As "step 23", the data processing / control section 13 determines, based on these many tomographic image data,
An approximate model 56 of the head is created.

That is, as the approximate model 56 of the head, there are generally a sphere model, a multi-layer sphere model, a triangular element model and the like.

The sphere model approximates a complicated head to one sphere, and is approximated by estimating the scalp, skull or brain position. The multi-layered sphere model estimates the scalp, skull, or brain position and approximates them with a plurality of spheres having different radii. On the other hand, the triangular element model expresses the head epidermis, the skull, the brain position, and the like with triangular elements. Although the triangular element model is the most accurate model close to the actual head, the calculation time of model creation and signal source estimation becomes long. If a sphere model inscribed in the scalp or cerebral cortex is used, it fits well with the clinical purpose.
A case will be described where a sphere model 56 (hereinafter, the sphere model is also given the same reference numeral as the approximate model 56). And
A method for obtaining an approximate model (particularly, a sphere model) is known as follows.

That is, first, a surface model of the scalp or cerebral cortex is created from a tomographic image.

Next, of this surface model, a portion used for approximation is selected (that is, there is a portion of the surface model not used for approximation).

Finally, the maximum sphere inscribed in the surface model is searched. Specifically, the center position (x,
y, z) and radius r as parameters, the distance d between the representative discrete point (E _i ) on the surface model and the surface of the sphere
_i (However, this distance d _i is the center position (x, y, z) of the sphere.
And a discrete point (E _i ) on a straight line), d _i ≧
Under the condition of 0, the sphere with the maximum r is obtained. still,
Instead of inscribed, circumscribed or the least squares method may be used for approximation.

As will be described later, the approximate model is
The center, radius, etc. can be changed interactively to allow changes and modifications.

In step 24, the data processing / control unit 13 causes the image representing the shape of the head and the sphere model 56.
Are combined and the combined image is displayed on the display unit 14 such as a CTR.

Here, display examples of the created composite image are shown in FIGS.

That is, FIGS. 3 to 5 are examples of a composite image display of each cross-sectional view viewed from three directions and the sphere model 56 image, and FIG. 4 shows two cross-sectional views viewed from three directions. It is an example in which a combined image of cross-sectional views (sagittal direction, coronal direction) and a spherical model 56 image are three-dimensionally combined and displayed.
Here, each sectional view seen from three directions is the sagittal direction,
Each slice image in the coronal direction and the axial direction is set as an arbitrary slice image, and the sphere model 56 is combined and displayed as a surface model. FIG. 7 shows a head skin image 55 and a sphere model 56.
3 is an example of a composite image display in which each of the surfaces is displayed and combined.

As "step 25", the operator
It is determined whether or not the created sphere model 56 is changed, and an instruction is given.

That is, as a result, the operator can visually and three-dimensionally understand the positional relationship between the actual head shape and the sphere model 56 based on the tomographic image. And the like (FIGS. 3 to 5, 6, and 7 are all displayed at the same time on the display unit 14), for example, when it is determined that the deviation between the actual head shape and the spherical model 56 is large. In step 25, the change instruction section 15 such as a mouse is used to instruct to change the sphere model 56, and the process proceeds to step 26. On the other hand, if there is no change, the confirmation instruction section 16 such as a keyboard is used to change the sphere model 5
An instruction to confirm 6 is issued, and the process proceeds to step 28.

As "step 26", the center position, radius, etc. of the sphere model 56 are changed.

That is, in order to change the center position and radius of the sphere model 56 to the contents desired by the operator by looking at the composite image displayed on the display unit 14, the change instruction unit 15 such as a mouse is used. Based on any one of the above composite images, the change content is specifically instructed.

As "step 27", the change instruction section 15
Transmits the instruction content to the data processing / control unit 13, and the data processing / control unit 13 changes the sphere model 56 and the like in accordance with the instruction content and compositely displays the new sphere model 56 on the display unit 14 (in this case, (Along with a change based on any of the images, the tomographic image, the stereoscopic image, or the like displayed on the display unit 14 changes in whole or in part in conjunction). If the operator determines from the composite image that the change is necessary again, the change instructing section 15 gives an instruction to that effect, and the process proceeds to step 26 again.

As "step 28", the sphere model 56 is confirmed, and the process ends.

That is, when the operator looks at the displayed composite image of the sphere model 56 and determines that there is no need to change, and instructs the confirmation instructing section 16 to that effect, it is currently displayed. An instruction to confirm the sphere model 56 is transmitted from the confirmation instruction unit 16 to the data processing / control unit 13, and the process of creating the sphere model 56 ends. Various kinds of data are recorded by the recording unit 17 such as a printer as necessary.

When the center and the radius of the sphere model 56 are determined as described above, the information about them is obtained by the SQUID sensor 1.
It is used for the inverse problem analysis for estimating the signal source from the brain magnetic field data measured in 1. However, since the center position of the sphere model 56 is represented by the coordinate system of the tomographic imaging apparatus 18 (MRI apparatus), the SQUID sensor 11 is moved from the coordinate system of the tomographic imaging apparatus 18 (MRI apparatus) by another alignment means. It needs to be converted to the coordinate system. Conventionally, as a method of obtaining the relative positional relationship between the coordinate systems of the SQUID sensor and the tomographic imaging apparatus (MRI apparatus), when measuring the brain magnetic field, the above three points on the head of the subject during magnetic measurement are used. A magnetic field generating coil is attached, and the position of the magnetic field generating coil is measured by a predetermined magnetic field detecting device. On the other hand, at the time of capturing a tomographic image, markers that can be detected by the tomographic imaging device at the same three points on the head of the subject (for example,
Liver oil marker) is attached and the position of the marker showing high accuracy value is detected from the tomographic image to obtain the positional relationship between the SQUID sensor and the tomographic imaging device (Magneteic localiz
ation of neuroral activityin the human brain, T.Ya
mamoto et al., New York Univ., Proc. Natl. Acad.Sc
i. USA), the magnetic field of the magnetism generating coil attached to the above three points is multi-channel without using a predetermined magnetic detection device.
The method of estimating the position of the magnetism generating coil by measuring with a SQUID sensor (The positioning problem in biomagnetic mea
surements: a solution for arrays of superconducting
sensors, SNErne et al., IEEETransaction on Magn
etics, vol. MAG-23, No. 20, 1987) is known.

In the above embodiment, the case where the data processing / control unit 13 automatically creates an approximate model of the head based on a large number of tomographic image data has been described, but the present invention is not limited to this. The data processing / control unit 13 does not automatically create an approximate model of the head based on the tomographic image data, but, for example, on a display unit such as a CRT, a plurality of types other than the shape of the measurement target are displayed. The initial approximation model (for example, a model that can be used as an approximation model such as a sphere model, a multi-layer sphere model, a triangular element model, a flat plate model, a torso model, etc.) may be displayed. That is, by providing the selecting unit (not shown) as the selecting means of the present invention of claim 2, which allows the operator to select an approximate model desired from the plurality of initial approximate models, Display contents of the display (for example,
The operator performs all or part of the processing such as selecting the initial approximation model while looking at the image of the initial approximation model, tomographic image, etc., changing the position and size of the approximation model, and finalizing the final approximation model. Of course, you may do it yourself.

Further, in the above embodiment, as "step 22", the data processing / control unit 13 creates an image representing the shape of the head from the tomographic image by the tomographic imaging device 18, and based on this, the head Although the case of creating the approximate model has been described, the present invention is not limited to this, and, for example, the approximate model of the measurement target may be created by directly using the tomographic image captured by the tomographic imaging apparatus.

In the above embodiment, the case where the center, radius, etc. can be interactively changed by using the change instructing section 15 or the like in order to change or modify the created approximate model will be described. However, not limited to this, if an approximate model can be created automatically from the tomographic image with high accuracy,
The change instruction unit 15 and the like are unnecessary.

Further, in the above embodiment, the position and size of the approximate head model that is automatically created are automatically created, but the present invention is not limited to this. It may be simply created, or its size may be specified manually.

Further, in the above-described embodiment, regarding the display method, the case where the sphere model is created and displayed as a three-dimensional surface model as shown in FIG. 6 has been described. Stereoscopic display may be used.

In addition to the display method in the above embodiment,
In order to make it easier to see the positional relationship between the three-dimensional cross-sectional view and the sphere model, the mesh size, color, transparency, etc. may be appropriately changed.

Further, in the above-mentioned embodiment, the three-direction cross-sectional views have been described as being created and displayed as the cross-sectional views in the sagittal, coronal and axial directions as shown in FIGS. 3 to 5. However, the present invention is not limited to this. Also, any oblique sectional view may be used.

In the above embodiment, the composite image shown in FIG.
The explanation has been given on the case where the operator arbitrarily combines two cross-sectional directions (for example, the sagittal direction and the coronal direction) to create a composite image in a three-dimensional space. Of course, it is also possible to change the direction and so on, and change the so-called "line of sight" of the operator regarding the display image on the display unit.

In the above embodiment, the display section shown in FIG.
Although the case where all the images in FIG. 7 are displayed at the same time has been described, the present invention is not limited to this, and in short, as long as the approximate model and the tomographic image are combined and displayed, the types of images displayed at the same time and The number etc. does not matter.

Further, in the above embodiment, the synthetic display of the head epidermis image and the sphere model shown in FIG. 7 has been described in the case where the head epidermis and the sphere model are created and displayed as a three-dimensional surface model, but the present invention is not limited to this. , Other display methods such as wire frame display may be used, and the mesh size, color or transparency of both can be changed appropriately to make it easier to see the positional relationship between the head epidermis and the sphere model. Of course it's okay.

In addition to the display described in the above embodiment, for example, an arbitrary tomographic image can be superimposed so that the positional relationship between the anatomical information and the sphere model can be grasped. Good.

Further, although the measurement target of the present invention is described as a head as a living body in the above embodiment, the measurement target is not limited to this, and may be, for example, the heart, or the subject may have brain death. It may be in a state, or it may be inanimate in nature, or may be anything as long as it has a signal source for generating a magnetic field.

Further, the present invention may be realized in terms of hardware using each component, or not limited to this, and may be realized by software.

[0060]

As is apparent from the above description,
According to the present invention of claim 1, the created approximation model is an actual raw material.
It's even easier to say how close you are to your body
In addition, it can be grasped accurately and depending on the degree of agreement,
It has the advantage that a similar model can be changed .

The present invention according to claim 2 is obtained by selection.
To what degree the initial approximation model
It is even easier and more accurate to say
Depending on the degree of matching, the approximation model may be changed.
It has the advantage of coming .

[Brief description of the drawings]

FIG. 1 is a configuration diagram for explaining a configuration of an approximate model display device according to an embodiment of the present invention.

FIG. 2 is a flowchart of a process / operation including the approximate model display device according to the embodiment.

FIG. 3 is a diagram in which a halftone image displayed on a display is output from a printer as an example of a composite image display of a cross-sectional view of the head viewed from the sagittal direction and a spherical model image by the display unit according to the embodiment. Yes, it is a photograph instead of a drawing.

FIG. 4 is a diagram in which a halftone image displayed on a display is output from a printer as an example of a composite image display of a cross-sectional view of the head viewed from the coronal direction and a spherical model image by the display unit according to the embodiment. Yes, it is a photograph instead of a drawing.

FIG. 5 is a diagram in which a halftone image displayed on a display is output from a printer as an example of a composite image display of a cross-sectional view of the head viewed from the axial direction and a spherical model image by the display unit according to the embodiment. Yes, it is a photograph instead of a drawing.

FIG. 6 is a three-dimensional composite display of a combined image of two cross-sectional views (sagittal direction and coronal direction) of each cross-sectional view seen from three directions and a spherical model image by the display unit according to the embodiment. As an example, a halftone image displayed on the display is output from a printer and is a photograph instead of a drawing.

FIG. 7 is a diagram in which a halftone image displayed on a display is output from a printer as a composite image display example in which a head epidermis image and a sphere model are surface-displayed and composited by the display unit according to the embodiment. Yes, it is a photograph instead of a drawing.

FIG. 8 is an explanatory diagram showing the principle of a conventional biomagnetism measuring device.

[Explanation of symbols]

 11 SQUID sensor 12 Data collection unit 13 Data processing / control unit 14 Display unit 15 Change instruction unit 16 Confirmation instruction unit 17 Recording unit 18 Tomographic imaging device 56 Approximate model (sphere model) 101 Brain activity current 102 Brain magnetic field 103 SQUID sensor 104 Liquid helium

Front Page Continuation (72) Inventor Yoshifumi Sadahiro 3-2-95 Chiyosaki, Nishi-ku, Osaka City OGIS Research Institute (72) Inventor Hideyuki Tsutsui 3-29 Chiyosaki Nishi-ku, Osaka City OGIS Research Institute (56) ) References JP-A-3-1839 (JP, A) JP-A-3-118040 (JP, A)

Claims (2)

(57) [Claims] 1. A method for estimating at least a position of a signal source existing in a measurement target of a living body and generating a magnetic field,
An approximate model creating means for creating the approximate model of the measurement object based on the tomographic image of the measurement object; a combining means for combining the created approximate model and the tomographic shape; and displaying the combined image. and display means that you can, before
The above approximation model is applied to the synthesized image data.
Change instruction means to be changed and the changed approximate model
And a determination instructing means for determining the tomographic image , wherein the tomographic images are images viewed from three directions orthogonal to each other.
The display means displays each tomographic image and the approximate model.
The images combined can be displayed at the same time.
Any one of the images synthesized by the change instruction means
When the approximation model of is changed, other
An approximate model display device characterized in that an approximate model of a combined image can also be changed . 2. A magnetic field that exists in a measurement target of a living body and generates a magnetic field.
First for use in estimating at least the location of the signal source
Selection means for selecting the period approximation model, and the selected first
Synthesis for synthesizing the approximate model and the tomographic image of the measurement target
Means, display means for displaying the combined image, and size of the initial approximation model for the combined image.
And a change instruction means for changing the position and a confirmation instruction means for confirming the changed approximate model, and the tomographic image is an image viewed from three directions orthogonal to each other.
The display means displays each tomographic image and the approximate model.
The images combined can be displayed at the same time.
Any one of the images synthesized by the change instruction means
When the approximation model of is changed, other
An approximate model display device characterized in that an approximate model of a combined image can also be changed .