JPH07194573A - Magnetic resonance diagnostic system and maximum intensity projection for magnetic resonance blood vessel image - Google Patents

Abstract

PURPOSE:To shorten processing time by adding adjacent two or more voxel values on projection lines, and finding a maximum value from an added result in a system to obtain a two-dimensional blood stream image by projecting an image on a plane, by photographing a three-dimensional blood stream image of a specimen. CONSTITUTION:Image data such as MRA is obtained by performing image reconstituting processing according to a magnetic resonance signal sent from a data gathering part 13 according to output of a probe 9, and an obtained image is displayed on a display 16. In this case, a see-through direction setting part to set the see-through direction inputted at a console 15 to three-dimensional image data by the data gathering part 13, is arranged in an electronic computer 14, and adjacent two voxel values on respective see-through directional projection lines are added together, and respective added results and a maximum value being a reference are compared with each other. A value judged as large is stored as a new maximum value, and a two-dimensional image is created according to the maximum value. According to this constitution, since the number of comparing-judging times can be reduced, processing time can be shortened.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of displaying a blood vessel image obtained by a magnetic resonance diagnostic apparatus.

[0002]

2. Description of the Related Art In recent years, with the development of a magnetic resonance diagnostic apparatus, a magnetic resonance angiography method (also referred to as MRA) for taking a blood vessel image in a living body has been put into practical use.

As one method of such MRA, a method of obtaining blood flow information based on the time-of-flight effect has been conventionally known. This is because when imaging at a short TR (repetition time), the stationary tissue in the region of interest (ROI) does not have enough time for the in-plane magnetization to return to equilibrium completely and the signal is small. On the other hand, since the blood flow flowing into the imaging region is not affected by the RF pulse, the property of obtaining a high signal is used. By such a method, only the blood flow information can be selectively collected, and three-dimensional data of the blood flow information of the subject can be obtained.

However, since the amount of data in a three-dimensional image is enormous, there is a method of cutting out a two-dimensional image from the three-dimensional image and viewing each image one by one. In this method, useful information is obtained from the data. Becomes harder. Therefore, a method of projecting a three-dimensional image on a plane in a predetermined direction to create a two-dimensional blood vessel image is often adopted. As a typical example of this creation method, the maximum intensity projection method (MIP: Maximum Inte
nsity Projection). This is a method in which a predetermined direction is set on a three-dimensional image, the maximum value is found on the projection lines of all voxels orthogonal to this direction, and a two-dimensional image is created based on this.

FIG. 6 is an explanatory diagram of the maximum intensity projection method.
3D image 101 defined in 3D space of X, Y, Z
Is an example in which the maximum value is projected in the Z-axis direction. As shown, in this example, blood vessels 102,
103 exists, and a two-dimensional image 105 is obtained by performing maximum intensity projection in the line-of-sight direction 104.

Next, the principle of maximum intensity projection will be described with reference to FIG. B ₁ , B ₂ , ..., B _m shown in the figure indicate voxels on one projection line along the line-of-sight direction 104. In the maximum intensity projection method, first the Z coordinate is 1 on the projection line.
The voxel of is temporarily set as the maximum value MAXI. Also, the X coordinate is i, the Y coordinate is j, the Z coordinate is k, and the voxel value of each coordinate is defined as A (i, j, k). That is, the maximum value MA
XI is MAXI = A (i, j, 1).

Then, the maximum value MAXI is compared with the size of the voxel value of each voxel B _{1 to} B _m , and the larger one is set as a new maximum value MAXI (step ST31 to FIG. 5).
ST33). Then, the maximum value MAXI (step ST34) at the time when the comparison with the last voxel B _m is completed is set as the maximum value on this projection line. This method is easy to carry out and good two-dimensional images can be obtained.

[0008]

However, in such a conventional maximum intensity projection method, a long processing time is required since all voxel values on each projection line are compared with the maximum value MAXI. Further, when actually making a diagnosis, it is necessary to perform projection from a large number of line-of-sight directions, and the problem of processing time becomes even more prominent.

The present invention has been made to solve such a conventional problem, and an object of the present invention is to provide a maximum value projection method of a magnetic resonance blood vessel image capable of shortening the processing time and the maximum value projection method. It is to provide a magnetic resonance diagnostic apparatus to which the value projection method can be applied.

[0010]

In order to achieve the above object, the magnetic resonance diagnostic apparatus of the present invention projects a three-dimensional blood flow image of a subject by using a magnetic resonance angiography method, and In a magnetic resonance diagnostic apparatus to which a maximum intensity projection method for obtaining a two-dimensional blood flow image by projecting an image on a plane from a predetermined direction can be applied, a predetermined number of voxels adjacent to each other on each projection line in the projection direction. It is characterized by having means for adding values and means for making the maximum value of the addition results the maximum value of this projection line.

Further, the maximum intensity projection method of the present invention is a method for projecting a three-dimensional blood flow image projected by a magnetic resonance diagnostic apparatus onto a plane from a predetermined direction to obtain a two-dimensional blood vessel image. In the value projection method, first adding a predetermined number of two or more adjacent voxel values on each projection line in the projection direction,
And the second step of setting an arbitrary value of the addition result to the maximum value, and the other addition result and the maximum value are sequentially compared and determined, and the larger one is set to the new maximum value. And a fourth step of determining the maximum value obtained after the comparison judgment is performed for all addition results as the maximum value of this projection line.

[0012]

[Function] Blood (hereinafter referred to as "foreground") signal strength and stationary tissue (hereinafter referred to as "background")
The ratio of the signal intensity to the signal intensity determines the contrast-to-noise ratio (CNR) of the MIP image. Using current MRA imaging techniques, the resulting CNR is usually good enough that the foreground signal strength is several times greater than the background signal strength. Utilizing this property, the processing is not performed while comparing the signal intensities of individual voxels on the projection line (hereinafter referred to as voxel values), but the total value of two or more adjacent voxels on the projection line and the processing It is possible to reduce the number of comparisons by comparing the maximum values MAXI obtained up to that point. Generally, in a computer, the comparison processing takes a longer processing time than the addition processing. Therefore, by reducing the comparison calculation, the MIP processing time becomes shorter.

For high CNRs, the sum of only background values is expected to have a relatively small value for the strength of the foreground values. Further, in the blood vessel image, the number of voxels forming the background is generally larger than the number of voxels forming the normal foreground, and the processing time is significantly shortened.

In the method proposed by the present invention, when searching for the maximum value, the conventional MIP method compares voxel values one by one, whereas two or more data are collectively processed. There is. When the total number of voxels handled collectively is larger than MAXI, the individual voxel values of the voxels treated collectively are compared to obtain the maximum value.
However, when the foreground is far from the starting point on the projection line, the maximum value is close to the noise level until the foreground is reached, and the number of comparisons is large. The following measures can be considered to avoid such problems.

It is to initialize MAXI at the median partial value on the projection line. That is, blood vessels are RO
This makes it more likely that MAXI will have a foreground value at the beginning of the comparison process, since it is more likely to be in the middle of I. On the other hand, when the data size is large and the projection line is long, MAXI is used in order to find the foreground value at an early stage of processing.
It is effective to initialize by comparing the values of several points in the projection line at regular intervals and using the maximum value.

That is, in the conventional MIP algorithm, in order to obtain the brightness at a specific position of the projected image, each voxel brightness on the projection line is compared with the initial value, and the larger value is sequentially set as a new maximum value. How to set. The processing that occupies most of the total processing time is the time of the comparison operation for each pixel. For a 256x256x32 data array, a comparison of 2031616 (256x256x31) is required if the maximum was initialized with the first value of the projection line.

According to the present invention, the three-dimensional blood vessel image data photographed by the magnetic resonance diagnostic apparatus is used in the conventional MIP from a predetermined direction.
Since the maximum value can be found with a smaller number of comparison operations than the modulo, the processing time can be shortened.

[0018]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of a magnetic resonance diagnostic apparatus according to the present invention. In the figure, the static magnetic field magnet 1,
The magnetic field homogeneity adjustment coil 3 and the gradient magnetic field generation coil 5 are respectively an excitation power source 2 and a magnetic field homogeneity adjustment coil power source 4.
It is driven by the power source 6 for the gradient magnetic field generating coil. As a result, a uniform static magnetic field and a gradient magnetic field having a linear gradient magnetic field distribution in three directions which are orthogonal to each other in the same direction are applied to the subject 7. Further, a high frequency signal is sent from the transmitter 10 to the probe 9, and a high frequency magnetic field is applied to the subject 7. Here, the probe 9 may be used for both transmission and reception, or may be provided separately for transmission and reception.

The magnetic resonance signal received by the probe 9 is subjected to quadrature phase detection by the receiving section 11 and, after A / D conversion transferred to the data collecting section 13, is sent to the electronic computer 14. As described above, the excitation power supply 2, the magnetic field uniformity adjustment coil power supply 4, the gradient magnetic field generation coil power supply 6, the transmission unit 10, the reception unit 11,
The data collection unit 13 is controlled by the system controller 12.

The system controller 12 is the electronic computer 1.
Controlled by console 15 via 4. The electronic computer 14 performs image reconstruction processing based on the magnetic resonance signal sent from the data collection unit 13 to obtain image data such as MRA. The obtained image is displayed on the image display 16. The computer 14 and the bed 8 are the console 1
Controlled by 5.

FIG. 2 is a block diagram showing the internal structure of the electronic computer 14 which is the main part of this embodiment. As shown in the figure, the electronic computer 14 includes a perspective direction setting unit 21 that sets a perspective direction input to the console 15 for three-dimensional image data provided from the data collection unit 13, and projections of the perspective directions. An adder 22 that adds two adjacent voxel values on the line, a comparator 23 that compares each addition result with a predetermined maximum reference value, and a comparator 23.
As a result of the comparison, the memory 24 that stores the value determined to be larger as a new maximum value, and finally the memory 2
The image processing unit 25 creates a two-dimensional image based on the maximum value stored in No. 4.

Next, the operation of this embodiment will be described with reference to the flow chart shown in FIG.

In the computer 14, first, the three-dimensional MRA is
The data A (i, j, k) is read (step ST1).
After that, A (i, j,
The perspective direction of k) is set by the perspective direction setting unit 21 (step ST2), and the projection line is set in this direction (step ST3).

Next, the initial maximum value M is stored in the memory 24.
Stores AXI. Usually, since there are many blood vessels near the center of the region of interest, the voxel value at the midpoint of the projection line is set to the maximum value MAXI in this embodiment. That is, MA
XI = A (i, j, K / 2) (step ST
4). However, K is the maximum value of k.

After that, the values of two adjacent voxels on the projection line are added. That is, S = A (i, j, k) + A (i, j, k + 1) (step ST5).

Then, the comparator 23 compares the addition result S with the maximum value MAXI stored in the memory 24 (step ST6), and when MAXI> S (YES in step ST6). The maximum value is not rewritten and the process proceeds to the comparison with the addition result of the next voxel value (NO in step ST8).

On the other hand, when MAXI <S (NO in step ST6), MAXI, A (i, j, k),
The voxel values of A (i, j, k + 1) are compared, and the maximum value is stored in the memory 24 as a new maximum value MAXI (step ST7).

When the processing is completed for all voxels on the projection line (YES in step ST8), the maximum value on the projection line becomes MAXI stored in the memory 24 at the end. That is, the pixel value I (i, i of the projected image
j) is I (i, j) = MAXI (step ST
9). Thereafter, similar processing is performed for each projection line, and when the processing is completed for all projection lines (YES in step ST10), the obtained pixel value I (i, i,
A projection image is created based on j) (step ST1).
1).

Next, the operation for obtaining the maximum value on one projection line will be described in more detail with reference to FIG. B _{1 to} B _m shown in the figure represent voxels existing on the projection line, and S 1
_{1 to} S _m-1 indicate the addition result of adjacent voxel values. Then, in step ST21, the addition result S _k and the maximum value MA
XI is compared, and if S _k <MAXI then M
The AXI is not rewritten and the next addition result is compared.

On the other hand, when S _k > MAXI, MAX
The largest one of I, A (i, j, B _k ) and A (i, j, B _{k +1} ) is _defined as MAXI (step ST22). Then, this processing is continued until the comparison is completed up to the final addition result S _m-1 (step ST24).

In this way, the maximum value on the projection line is obtained,
The maximum intensity projection image in the perspective direction can be obtained.

In this way, in this embodiment, the values of adjacent voxels on one projection line are added, and the maximum value is obtained using the addition result. Therefore, as in the past,
Since the maximum value is not compared for all voxel values, the number of comparison judgments when executing the process can be reduced,
The processing time can be significantly shortened. For example, 25
Tested on an MRA data set containing 6 * 256 * 32 cat basilar vessels, when using the sum of two pixels and when MAXI was initialized at the midpoint of the projection line, The processing speed could be reduced by 20% compared to MIP.

In this embodiment, an example in which two adjacent voxel values are added has been shown, but the present invention is not limited to this, and three or more voxel values may be added.

In this embodiment, the voxel value at the center of the projection line is set to the initial maximum value, but other values may be set to the maximum value.

[0035]

As described above, according to the present invention,
Two or more voxel values adjacent to each other on the projection line are added, and the maximum value is obtained using the addition result. Therefore, it is possible to reduce the number of comparison determinations when executing the processing, and it is possible to obtain an effect of significantly reducing the processing time.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a magnetic resonance diagnostic apparatus according to the present invention.

FIG. 2 is a block diagram showing an internal configuration of a computer.

FIG. 3 is a flowchart showing the operation of the embodiment of the present invention.

FIG. 4 is a flowchart showing an operation for obtaining a maximum value on a projection line.

FIG. 5 is a flowchart showing the operation of the conventional maximum intensity projection method.

FIG. 6 is an explanatory diagram showing the principle of the maximum intensity projection method.

[Explanation of symbols]

 12 system controller 14 computer 21 perspective direction setting unit 22 adder 23 comparator 24 memory 25 imaging processing unit

Claims (2)

[Claims] 1. A three-dimensional blood flow image of a subject is photographed and
A magnetic resonance diagnostic apparatus utilizing a maximum intensity projection method for obtaining a two-dimensional blood flow image by projecting a two-dimensional blood flow image onto a plane from a predetermined direction, wherein a predetermined number of two or more adjacent to each projection line in the predetermined direction. An addition unit that sequentially adds each voxel value, and a unit that sets the maximum value of the addition results obtained by sequentially adding the voxel values as the maximum value of this projection line. A characteristic magnetic resonance diagnostic apparatus. 2. A maximum value projection method of a magnetic resonance blood vessel image for obtaining a two-dimensional blood vessel image by projecting a three-dimensional blood flow image photographed by a magnetic resonance diagnostic apparatus onto a plane from a predetermined direction, wherein: A first step of sequentially adding two or more voxel values adjacent to each other on each projection line for a predetermined number, a second step of maximizing an arbitrary value of the addition results, and another addition The third step of sequentially comparing and judging the result and the maximum value and setting the larger one as a new maximum value, and the maximum value obtained after the comparison judgment is performed for all addition results Fourth to determine the maximum value
And a maximum intensity projection method of a magnetic resonance blood vessel image, comprising: