JP3538726B2 - Magnetic resonance device displaying diffusion coefficient of diffusion tensor - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a magnetic resonance apparatus and, more particularly, to a method for determining a running state of a fibrous biological tissue such as a nerve fiber or a fibrous tissue, and a method of determining the running state of the determined fibrous biological tissue. The present invention relates to a magnetic resonance apparatus having a display device suitable for displaying.

[0002]

2. Description of the Related Art Diffusion measurement by a magnetic resonance apparatus is a method for measuring a diffusion coefficient of a molecule contained in an object to be measured.
The measured diffusion coefficient takes into account various factors such as restrictions on cell membranes and intracellular structures in a living body, and flows and movements such as intracellular flow. For this reason, it is not possible to obtain a pure diffusion coefficient, but on the contrary, clinical application is being studied to have information reflecting the state of a living body.

[0003] In particular, the measurement of nerve fiber running is one of the applied techniques of diffusion measurement. The diffusion of water molecules in a nerve fiber and the surrounding myelin sheath is hardly restricted in a direction parallel to the running direction of the fiber, but is strongly limited in a direction perpendicular to the running direction of the fiber by the influence of the cell membrane. For this reason, when the diffusion coefficient is measured, it is high in the direction parallel to the running of the nerve fiber and low in the direction perpendicular thereto. This is called diffusion anisotropy, and a magnetic resonance apparatus measures a diffusion tensor composed of a 3-by-3 symmetric matrix. By calculating the principal value (maximum eigenvalue) and principal vector (eigenvector corresponding to the maximum eigenvalue) of the diffusion tensor, it becomes possible to determine the traveling direction of the nerve fiber. For this, see Journal of Magnetic Resonance (Journal
of Magnetic Resonance), B 103, pp. 247-254, 199
Reported for four years.

[0004] As a method of displaying the running of the obtained nerve fiber, a method using an ellipsoidal sphere (Biophysical Journal, 66, 259-267,
1994), a method using Maximum Intensity Plot (MIP) (JP-A-7-371), a method using arrows and triangles (Magnetic Resonance in Medicine).
ance in Medicine), No. 34, pages 786-791, published in 1995).

The method using an ellipsoidal sphere expresses the diffusion tensor obtained by each voxel with an ellipsoidal sphere. Since the major axis of the ellipsoid indicates the traveling direction of the nerve fiber and the flatness of the ellipsoid indicates the degree of diffusion anisotropy, it is suitable for confirming the traveling direction of the nerve fiber in each voxel. However, when a two-dimensional image is represented by this display method, the relationship between voxels is not clear, and it is difficult to grasp how the nerve fibers travel as a whole. Furthermore, in the case of three dimensions, display itself is also difficult.

The method using MIP is a method of acquiring an image composed of a plurality of slices subjected to diffusion emphasis in a certain direction, and displaying only a portion having a high luminance, that is, a nerve fiber portion. This method makes it easy to find abnormalities such as demyelinating diseases of nerve fibers, but has the drawback that the running direction of nerve fibers in each voxel cannot be observed.

The method using an arrow or the like is a method in which the main component of the diffusion tensor is represented by an arrow or the like and is superimposed on a morphological image and displayed. In this display method, running of nerve fibers is easy to grasp in a two-dimensional image, but it becomes difficult in the case of three-dimensional images, and there is a problem that a spatial resolution must be reduced or a morphological image must be enlarged and displayed. is there.

As described above, it is difficult to easily grasp the running of nerve fibers by the above three methods.
In particular, it is necessary to simultaneously grasp which areas of the gray matter are connected to the nerve fibers of the white matter and whether there is no abnormality in the running of the nerve fibers.It is necessary to obtain information on the running direction of each voxel and the connection information between the voxels. Was difficult because it was essential.

Further, a measurement method applying the above method has been proposed for running not only nerve fibers but also muscle fibers such as myocardium. This is discussed, for example, in Magnetic Resonance in Medicine, No. 34, pages 786-791, 1995.
Reported in the year issuance. This has a similar problem.

[0010]

In the above-mentioned prior art, it is difficult to display the running of nerve fibers so that they can be easily grasped. In particular, it is difficult to grasp how nerve fibers are connected between voxels, and it is therefore difficult to grasp the running of all nerve fibers. For example, it has been difficult to determine which areas of the gray matter are connected by nerve fibers and to determine whether there is any abnormality in the running. An object of the present invention is to provide a display device suitable for grasping the running of nerve fibers.

[0011]

According to the present invention, a point located in the direction with the highest diffusion coefficient is derived using a diffusion tensor at each point on an image, and data comprising a series of derived points is created. Provided is a display device including a processing unit and a display unit for displaying data including a series of generated points. The specific processing means calculates the direction in which the diffusion coefficient is the highest at an arbitrary start point, derives the next point separated by a predetermined distance from the start point in that direction, and also at this next point, the diffusion coefficient is the highest. The high direction is calculated, the next point separated by a predetermined distance is derived, and iteratively repeats, and data consisting of a series of points is created by this repetition. Since the direction with the highest diffusion coefficient matches the running direction of the fibrous biological tissue such as nerve fibers, this process creates points along the running of the biological tissue as a series of data, and displays this series of data. Thus, the running of the living tissue can be easily grasped.

According to the present invention, a series of data may be created from one point serving as a starting point or a plurality of starting points in a certain area. As a result, it is possible to select a fibrous biological tissue of only a point or a region of interest.

When setting the distance to the next point,
The distance may be changed according to the magnitude of the calculated diffusion coefficient. In a portion of the voxel where the curvature of the fibrous biological tissue is small, the tissue has a uniform shape in the same direction, and the diffusion coefficient in this direction is high. For this reason, by increasing the distance at a location where the diffusion coefficient is high, the error does not increase so much, but the number of repetition calculations can be reduced, that is, the processing time can be reduced.

If the magnitude of the signal strength at the derived point is smaller than a predetermined value, or if at least one of the conditions when the diffusion anisotropy of the diffusion coefficient is small is satisfied, the processing is repeated. May be terminated. As a result, points where the target object does not exist, points where the signal intensity is low and sufficient accuracy is not obtained, or points where fibrous biological tissue is not properly run and diffusion anisotropy is reduced are excluded from the series of data. It is possible to do.

In the means for displaying data consisting of a series of points, the data may be connected by a broken line or a curve, and displayed on an image obtained by the magnetic resonance apparatus. Accordingly, since both the morphological image and the running of the fibrous biological tissue can be observed at the same time, the position of the fibrous biological tissue can be grasped.

Further, a stripe pattern may be drawn on a line connecting a series of data, and a moving image may be displayed so as to move temporally in a direction in which the stripe pattern is connected. This makes it easier to determine each data when data consisting of a plurality of series of points is displayed at the same time. Further, when observing the running of the fibrous biological tissue included in a certain predetermined area,
Since the moving image is displayed so that the stripe pattern moves in the running direction of the tissue in the region, the running of the tissue in the entire region can be easily grasped.

[0017]

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

FIG. 1 is a schematic configuration diagram of a magnetic resonance apparatus according to the present invention. In FIG. 1, reference numeral 1 denotes a magnet that generates a static magnetic field, 2 denotes an object to be measured, and 3 denotes an object to generate and measure a high-frequency magnetic field.
Coils for the detection of magnetic resonance signals arising from the
Reference numerals 5 and 6 denote gradient magnetic field generating coils for generating gradient magnetic fields in the X, Y and Z directions, respectively. Reference numeral 7 denotes a coil driving device for supplying a current to the gradient magnetic field generating coils 4, 5, and 6. Reference numeral 8 denotes a computer for controlling the operation of each device, and 9 denotes a display device for processing and displaying measured data.

Next, an outline of the operation of the present apparatus will be described. The high-frequency magnetic field for exciting the nuclear spins of the measurement target 2 is generated by shaping the waveform of the high frequency generated by the synthesizer 10 with the modulator 11, amplifying the power, and supplying the coil 3 with current. The gradient magnetic field generating coils 4, 5, and 6 supplied with current from the coil driving device 7 generate gradient magnetic fields and modulate magnetic resonance signals from the measurement target 2. The modulated signal is received by the coil 3, amplified by the amplifier 12, and detected by the detector 13
, And is input to the computer 8. The computer 8 transfers the input data to the display device 9 and displays the data processing and the processing result on the display device 9. In some cases, the computer 8 performs data processing in order to speed up the processing. The computer 8 controls each device to operate at a timing and intensity programmed in advance. Among the programs, those describing the high-frequency magnetic field, the gradient magnetic field, and the timing and intensity of signal reception are called a sequence.

FIG. 2 shows an example of the sequence. An excitation high-frequency magnetic field pulse 21 is applied to induce a magnetic resonance phenomenon in a measurement target. The slice gradient magnetic field 24 is applied simultaneously with the application of the high-frequency magnetic field pulse, and a slice in the Z direction is selected. The magnetization is inverted by applying the inversion high frequency magnetic field pulse 22,
An echo 23 is generated. The generated echo 23 is stored as data by AD sampling. Also, AD
Readout gradient magnetic field 2 in X direction at the same time as sampling
5 is applied to add the position information in the X direction to the data. In addition, a phase encoding gradient magnetic field 26 is applied to give position information in the Y and Z directions to the data, and the intensity is changed every time measurement is repeated. Two diffusion gradient magnetic fields 27 are applied between the excitation high-frequency magnetic field pulse 21 and the inverted high-frequency magnetic field pulse 22 and between the inverted high-frequency magnetic field pulse 22 and the echo 23 to provide diffusion information. The diffusion gradient magnetic field is a combination of a gradient magnetic field in the X direction generated from the gradient magnetic field generating coil 4, a gradient magnetic field in the Y direction generated from the gradient magnetic field generating coil 5, and a gradient magnetic field in the Z direction generated from the gradient magnetic field generating coil 6. Or given alone. The two diffusion gradient magnetic fields are adjusted so that the time integral of the intensity becomes equal. At this time, if there is no diffusion motion, the phase of the magnetization dephased by the first diffusion gradient magnetic field is completely rephased by the second diffusion gradient magnetic field and compared with the case where no diffusion gradient magnetic field is applied. The signal strength does not attenuate. However, if the spread occurs, the phase cannot be completely rephased, so that the signal intensity attenuates at a rate corresponding to the intensity. In the ideal case, the signal strength attenuation is given by:

[0021]

(Equation 1)

Here, D represents a diffusion tensor, 3 rows 3
It is a symmetric matrix of columns. b is called a gradient magnetic field factor, and is calculated by the following equation from the application time and the applied intensity of the diffusion gradient magnetic field.

Bij = ∫γ ² {∫Gi (τ) dτ∫Gj (τ) dτ} dt In order to calculate the diffusion tensor, measurement is performed at least seven times by changing the application direction of the gradient magnetic field factor. It is calculated by Equation 1). Principal values and principal vectors have a significant meaning in obtaining the fiber running from the diffusion tensor. The main vector indicates the direction of the highest diffusion coefficient, which coincides with the running direction of the fiber. The method of calculating eigenvalues and eigenvectors is described in, for example, Numerical Computation Handbook, edited by Ohno, Isoda, and supervised by Ohmsha, 1990, and will not be described here.

Next, the operation of the display device of the present invention will be described. The display device of the present invention includes processing means for processing diffusion tensor data and means for displaying data obtained by the processing means. Specific processing means and display means are based on the algorithm shown in FIG. First, the direction in which the diffusion coefficient is the highest at the predetermined point a serving as the starting point, that is, the direction of the main vector is derived. Next, a point b that is separated from the point a by a predetermined distance b in the above direction is derived. Next, the derived points
This process is repeated starting from b. Create data consisting of a series of points derived in the iterative process and display it. FIG. 4 shows a schematic diagram of this processing on an image. FIG.
, The arrow on the grid corresponding to the pixel of the image indicates the main vector calculated by the above method. First,
Set the start point p0. Next, the principal value at this point p0,
Compute the principal vector. At this time, when the point p0 is on the grid, the calculated main value and main vector are used, and when it is not on the grid, it is calculated from the surrounding grid points by linear interpolation or the like. Thereby, the running direction of the nerve fiber at the point p0 is derived. A point p1 separated by a predetermined distance in the direction of the calculated main vector is derived. Next, this process is repeated with the point p1 as a starting point, and a series of data p0, p1,... Pn is created.
This results in a series of data consisting of points along the nerve fibers. Finally, this series of data is displayed.

Here, as the distance from the starting point to the next point becomes smaller, the error becomes smaller, but there is a problem that the calculation time becomes longer. In practice, it is sufficient to set this distance to a value smaller than the width of the grid. Also, by changing this distance according to the magnitude of the main value, it is possible to reduce the calculation time. This is because, when the curvature of the nerve fiber is small in the voxel, the nerve fiber has a uniform shape in the same direction, and the main value increases. For this reason, by increasing the distance to the next point at a point where the main value is large, the error does not increase so much, but the number of repetitive calculations, that is, the processing time can be reduced.

If either the signal intensity at the point p1 is smaller than a preset threshold value or the degree of the diffusion anisotropy at the point p0 is small, or both are satisfied, the repetitive operation is performed. May be provided. This makes it possible to exclude points where there is no measurement target, points where the signal intensity is small and sufficient accuracy is not obtained, or points where nerve fibers are not in a well-formed shape. Although points are derived only in the direction of the main vector in FIG. 4 for simplicity, the points are also derived in the direction opposite to the main vector. As a result, the points are derived in both directions even when the intermediate point of the fiber is set as the starting point, so that the entire fiber can be drawn.
In addition, when points are derived in one direction, errors accumulate, and points far from the original fiber may be derived. In order to suppress this, the data may be corrected by starting derivation of a point from an end point or an intermediate point.

FIG. 5 shows an example of a display result by the display means of the present invention. In FIG. 5, on the morphological image measured by the magnetic resonance apparatus, data consisting of a series of points is displayed by being connected by a polygonal line or a curve. This makes it easier to grasp the position of the nerve fiber on the morphological image. In addition, the morphological image is expressed in black and white shades, and the line representing the nerve fiber is expressed in another color, and the shade of the color is changed according to the magnitude of the main value, so that the running of the nerve fiber can be grasped more. Easy to do. Although FIG. 5 shows a two-dimensional representation, a similar display is possible with a three-dimensional image. At this time, data consisting of a series of points is drawn as a polygonal line or a curve on a three-dimensional image displayed by volume rendering or the like. In addition, depth information can be provided by displaying the image using a stereoscopic display.

In FIG. 5, a white cross indicates a set starting point. As a starting point setting means, a pointing device such as a mouse, or a coordinate value input means using a keyboard or the like may be provided.

Further, a drawing may be made such that the luminance value of a line segment connecting a series of data forms a stripe pattern, and a moving image may be displayed so as to move in the direction in which the stripe pattern is linked. Thereby, even when a plurality of start points are set, it is easy to distinguish a plurality of polygonal lines or curves. Note that instead of luminance, colors such as lightness and saturation may be changed to draw a stripe pattern.
As the stripe pattern, a method using a function for binarizing the setting of luminance, such as a dotted line or a broken line, or a method using a function for taking a continuous value such as a sine function can be used.

It is to be noted that not only a starting point setting means but also a starting point area setting means may be provided. For example, a pointing device or a means for inputting coordinate values for setting an area to be observed on an image, or a means for setting a threshold value for signal strength and extracting a region having a signal strength equal to or greater than the threshold value, and a means for setting a threshold value for diffusion anisotropy For extracting a region having a diffusion anisotropy greater than or equal to a threshold. In addition, when an area serving as a start point is set, data processing is performed as follows. Starting from a certain point in the area, data consisting of a series of points is created by the above-described method, and this series of data is removed from the area.
This process is repeated until there are no more points in the area to create a plurality of series of data. Display data is created from this data as follows. First, a luminance value is assigned so that a stripe pattern is drawn on a polygonal line or a curve representing each data with reference to an end point of each data. When there are a plurality of luminance values included in each voxel, the average or representative value is calculated, and one image having the luminance values is created. Next, a luminance value is reassigned so that the stripe pattern moves temporally in the connected direction, and a second image in which the luminance value of each voxel is calculated in the same manner as in the previous period is created. By repeating this, a series of obtained image data is displayed as a moving image. FIG. 6 shows a schematic diagram of one created image. The area surrounded by the dotted line represents the set observation area, and the striped pattern on the image is calculated from the striped pattern assigned to each data consisting of a series of points. This striped pattern is displayed as a moving image so as to temporally move in the running direction of the nerve fiber, and the running of the nerve fiber can be easily grasped.

In the present invention, the running of nerve fibers has been described. However, the present invention can be applied to other fibrous structures, for example, muscle fibers such as myocardium.

[0032]

As is apparent from the above description, according to the present invention, each of the main vectors of a series of points corresponding to the direction having the highest diffusion coefficient is derived, and the data obtained from the main vector is represented by a broken line or a curve. Alternatively, by displaying them in a striped pattern, the state of a fibrous biological tissue such as a nerve fiber can be displayed favorably.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating an example of a device configuration of a magnetic resonance apparatus which is an application apparatus of the present invention.

FIG. 2 is a diagram showing an example of a basic sequence used for drawing a nerve fiber.

FIG. 3 is a diagram showing an algorithm of processing means and display means provided in the display device of the present invention.

FIG. 4 is a diagram schematically showing an algorithm of a processing unit and a display unit of the present invention on an image.

FIG. 5 is a schematic diagram showing a display result by a display unit of the present invention.

FIG. 6 is a schematic diagram showing another example of a display result by the display unit of the present invention.

[Explanation of symbols]

1 Magnet for generating static magnetic field 2 Measurement target 3 High frequency magnetic field generation and signal detection coil 4,5,6 gradient coil 7 Coil drive 8 Calculator 9 Display device 10 Synthesizer 11 Modulator 12 Amplifier 13 Detector 21 Excitation RF pulse 22 Inverted high frequency magnetic field pulse 23 echo 24 slice gradient magnetic field 25 Readout gradient magnetic field 26 Phase Encoding Gradient Magnetic Field 27 Diffusion gradient magnetic field

Continuation of the front page (56) References JP-A 7-371 (JP, A) JP-A 7-313487 (JP, A) JP 9-501330 (JP, A) JP 8-500021 (JP , A) R. Nakata, Magnetic Resonance Axon Image, German-Japanese Medical Journal, Japan, 1997, Vol. 42, No. 2, p. 255-263 (58) Fields investigated (Int. Cl. ⁷ , DB name) A61B 5/055 JICST file (JOIS)

Claims (1)

(57) [Claims] 1. A method for determining a running state of a fibrous living tissue in a living body by processing diffusion tensor data at each point in the living body obtained by measuring a diffusion tensor of a molecule in the living body by magnetic resonance. Processing means, and display means for displaying the running state obtained by the processing means, the processing means calculates the direction in which the diffusion coefficient at the start point is the highest from an arbitrary point, and Deriving the next point separated by a predetermined distance from the start point, calculating the direction in which the diffusion coefficient is the highest also at this next point, deriving the next point separated by a predetermined distance, and sequentially repeating the same. A magnetic resonance apparatus, wherein data consisting of a series of points are generated by the processing means, and the display means connects and displays the data consisting of a series of points obtained by the processing means.