JPH0721351A - Image processing method and image processor - Google Patents

Abstract

PURPOSE:To realize the image processing method and the image processor which can execute a desired three-dimensional image display by an MIP or 3D, etc., processing by performing an efficient and exact mask processing to tomographic data obtained by a CT helical scan or MR, etc. CONSTITUTION:This image processor is provided with an image accumulating means 1 for accumulating image data to be processed, mask generating means 3, 4 for executing a processing determined in advance for generating a mask to the image data accumulated in the image accumulating means 1, and generating the mask to all image data, a mask processing part 5 for executing a mask processing to all image data by using the mask generated by the mask generating means 3, 4, and generating the image data to which the mask processing is finished, and an image processing part 6 for executing a desired image processing by using the mask-processed image data subjected to mask processing by the mask processing part 5.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing method and an improvement in stereoscopic image display of a tomographic image in an image processing apparatus.

[0002]

2. Description of the Related Art With the spread of high-speed CT and the practical use of helical scanning in recent years, attention has been paid to three-dimensional image display. Here, the helical scan means rotating the gantry including the X-ray tube and the detector a plurality of times, while moving the table on which the subject is placed forward or backward to spiral the X-ray around the subject. It is a method of scanning. Then, as the simplest three-dimensional image display method, MIP
(Maximum Intensity Projection) processing is known. The MIP processing is a projection image obtained by extracting the largest value for all pixels in each observation direction with respect to all original images to be processed. Moreover, since it is possible to extract the minimum value, the average value, or the specific value without limiting to the maximum value,
Hereinafter, this is referred to as an image projection method or IP (Image Projection)
Call it the law. For example, by contrast radiography, the lesion is low de
Sometimes it is rendered as nsity. At that time, Min_IP (Mi_IP (Mi
nimum Intensity Projection) processing is effective.

In general high-speed CT, tens to hundreds of images are usually obtained per patient. In particular, recently, a lung cancer mass screening using a helical scan has been proposed, but the image processing capability (interpretation efficiency) becomes more important than the CT processing capability at that time (setting of table moving speed suitable for diagnosis). There is. The IP method is expected as a new display method capable of solving such a problem.

In addition, by using a series of IP images obtained by finely changing the observation direction at various angles, it is possible to stereoscopically observe the running of blood vessels and the relationship between lesions and blood vessels in the cine mode. Such three-dimensional observation is considered to be useful for determining the tumor size and malignancy / benignity. Therefore, the IP method can be considered as one basic display function when performing continuous scanning in the axial direction in high-speed CT.

In addition to the above-mentioned CT, it is conceivable to use MR to acquire a tomographic image. However,
MR angio basically does not show bones, but CT images show bones quite strongly. Normally, the bone interferes with the CT MIP process. Therefore, CT MIP
In the process, the removal of unnecessary areas including bone becomes extremely important in the pre-process.

The following two methods can be considered for extracting a region of interest including such bone removal. The first method for extracting a region of interest is a method of setting unnecessary thresholds to eliminate unnecessary regions (threshold method). The second method is a method of extracting a region of interest with a mask (mask method).

[0007]

However, depending on the threshold value method, the region of interest cannot be extracted well when the contrast between the unnecessary organ and the organ of interest is difficult (CT value difference is small). There is. For example, by imaging, the CT values of bone and blood vessel may be considerably close to each other, and in such a case, it may be difficult to extract a region of interest.

Further, a specific pattern such as a square or a circle is mainly used as a mask in the case of processing by the mask method. In this way, it is efficient to perform mask processing using a specific pattern such as a square or circle, but it is not possible to completely remove unnecessary areas, or conversely the area of interest to be observed is deleted significantly. May occur. On the other hand, it is also possible to individually create a suitable mask by plotting on the image. According to such a method of plotting on each image, it is possible to completely remove only the unnecessary area, but it is necessary to work for each individual image, and the efficiency is greatly deteriorated. Therefore, in the mask method, efficient and accurate mask generation has been desired.

Further, as a three-dimensional image display method other than the above-mentioned MIP processing, there is 3D processing for generating a stereoscopic image using a plurality of tomographic data. For this 3D processing, unless a plurality of tomographic data from which the region of interest has been extracted is prepared in advance, a desired site may become invisible due to the outer bone or the like. For example, when the lungs and the heart are displayed in 3D using the tomographic data of the chest, there are cases where the ribs interfere and the inside cannot be seen.

For this reason, it is necessary to remove the unnecessary area by performing mask processing or threshold processing on all of the plurality of tomographic data in advance. However, as described above, removing unnecessary areas has problems in work efficiency and accuracy of removing unnecessary areas.

In addition, there has been a demand for displaying an arbitrary image at an arbitrary angle in both three-dimensional image display of the MIP process and the 3D process. The present invention has been made in view of the above points, and an object thereof is to efficiently perform mask processing on tomographic data obtained by CT helical scan, MR, etc., and to obtain a desired three-dimensional shape by MIP processing, 3D, etc. An object is to realize an image processing method and an image processing device capable of displaying an image.

[0012]

SUMMARY OF THE INVENTION The above-mentioned problems are solved in an image processing method for executing image processing after performing mask processing on image data of a plurality of cross sections, for generating a mask for image data to be processed. Perform predetermined processing,
A step of generating a mask for all image data, a step of performing mask processing on all image data using the generated mask, and generating masked image data,
And a step of performing desired image processing using the mask-processed image data.

Image storage means for storing image data to be processed, and predetermined processing for mask generation are performed on the image data stored in the image storage means,
A mask generation unit that generates a mask for all image data; and a mask processing unit that performs mask processing on all image data using the mask generated by the mask generation unit and generates masked image data. And an image processing unit that performs desired image processing by using the masked image data masked by the mask processing unit.

[0014]

In this image processing method and image processing apparatus,
The mask generation unit performs a predetermined process for mask generation on the image data to be processed, generates a mask for all image data, and the mask processing unit uses the mask for all image data. Masking is performed to generate masked image data, and desired image processing is executed using the masked image data thus generated. Since the masked image data has unnecessary areas removed, various image processing can be performed to display an arbitrary three-dimensional image.

[0015]

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 1 is an explanatory diagram showing a schematic processing step of an image processing method according to an embodiment of the present invention. Further, FIG. 2 shows an image processing apparatus according to an embodiment of the present invention. First, the configuration of the image processing apparatus will be described with reference to FIG.

In FIG. 2, the image storage unit 1 stores CT or MR image data obtained by a tomographic imaging device (not shown). The image processing apparatus 2 performs an efficient masking process on the image data stored in the image storage unit 1 and then executes desired image processing.
° MIP processing unit 3, mask generation unit 4, mask processing unit 5,
It has an image processing unit 6. The 0 ° MIP processing unit 3 generates data for determining a mask common to all image data to be processed or a predetermined amount of image data, and the mask generation unit 4 generates a common mask from this data.
The mask processing unit 5 uses a common mask to perform efficient mask processing on all image data or a predetermined amount of image data. The image processing unit 6 performs various desired image processing (3D, MIP processing, etc.) on the mask-processed image data, and displays the processing result on the display unit 7. The control unit 8 controls the operation of the entire apparatus.

Next, an image processing method and an image processing step of the image processing apparatus according to the embodiment of the present invention will be described.
The image storage unit 1 stores a large number of image data obtained by CT or MR. For example, it is assumed that image data including a series of image groups of dozens of tomographic images in the body axis direction of the chest and abdomen obtained by the helical scan is accumulated.

Each of the image data of the tomographic images thus accumulated has a relatively small change (close cross section)
In this case, the change between the respective image data is small, and the unnecessary areas are present at the same position. Therefore, a process for collectively generating one mask for all the image data is performed.

Further, when each of the image data has a relatively large change, the unnecessary areas of the respective image data are present at positions that are gradually displaced, so that all the image data is processed by one mask. As a result, the mask covers even the necessary area. Therefore, a division process is performed such that each area having a small change is divided into groups and a plurality of masks are generated. Therefore, the control unit 8 controls the image processing device 2 based on an instruction from an operation unit (not shown) (see FIG. 1).
step 1) ).

First, description will be given of a case where the division processing is not performed and the processing is performed collectively. The 0 ° MIP processing unit 3 performs the MIP processing of the maximum value, the minimum value, the average value, the specific value, etc. using the dozens of image data stored in the image storage unit 1.

As a result, image data (mask generation data) in which the bones, heart, etc. of all the image data are superimposed is generated (FIG. 1 (2)). In this embodiment, the case where the subject is horizontally placed on the table is taken as an example, and therefore the 0 ° MIP process in the body axis direction is performed, but the subject is tilted on the table. If it is placed, the MIP process is performed with an inclination that matches the inclination of the subject instead of 0 °.

Further, in the 0 ° MIP processing, the processing is performed by using all the image data, and then the mask for removing the common unnecessary area is generated. MI using two image data
Even if the P process is performed to generate a mask in a range that covers the unnecessary areas of these two data, a similar mask is obtained as a result. Therefore, in order to shorten the processing time, it is possible to perform the 0 ° MIP processing using the first and last two image data.

FIGS. 3 and 4 show a cross section of the image data, and FIG. 5 shows the mask generation image data obtained by the 0 ° MIP processing as described above. Here, what is shown in FIG. 5 is image data for mask generation obtained by performing MIP processing suitable for a mask for removing bones. Therefore, the bones appear to overlap.

The mask generation section 4 generates a mask with reference to the mask generation data (FIG. 5) thus obtained. When removing the bones, a mask corresponding to a portion where the bones appear to overlap each other in FIG. 5 is generated to generate a mask for batch processing that removes the bones and leaves a lesion (step in FIG. 1). (3)). It is also possible to generate a mask that leaves bones. Furthermore, a mask that removes arteries and veins to leave a capillary vessel part and a mask that removes an air layer are also conceivable.

That is, various masks can be generated in accordance with a desired image desired by the operator. For masking bones, MIP (Maximum Inte
nsityProjection) process to generate a mask. Also,
If masking air, etc., use the minimum value Min_IP
(Minimum Intensity Projection) Generate a mask.

Using the mask for batch processing thus obtained, the mask processing unit 5 executes mask processing on the image data consisting of a series of image groups stored in the image storage unit 1 (see FIG. 1). In step (4), the masked image data is returned to the image storage unit 1 again. In this case, since the mask processing can be performed by using the common mask, the efficiency is high and the mask processing time can be significantly shortened. Further, since the common mask is generated for the image data having a relatively small change, it becomes possible to accurately extract the region of interest.

Next, the case where the division process is performed will be described. The 0 ° MIP processing unit 3 uses several tens of image data stored in the image storage unit 1 to define several areas for each image having a small change, and the image storage unit 1 stores the image data for each area. Then, MIP processing of maximum value, minimum value, average value, specific value, etc. is performed for each area (step (5) in FIG. 1). As a result, the mask generation data is generated for each divided area (FIG. 1 (6)). By referring to the mask generation data (FIG. 5) obtained in this way, a mask for division processing is generated for each of the regions divided by the mask generation unit 4 (step (7) in FIG. 1).

Using the mask for division processing obtained in this way, the mask processing section 5 divides the mask processing into image data of a series of image groups accumulated in the image accumulating section 1 for each of the divided areas. Run (Fig. 1 step (8)
), The image data that has undergone mask processing is again stored in the image storage unit 1.
Return to. In this case, since the mask processing can be performed using the common mask for each area, the efficiency is high, and the mask processing time can be shortened as compared with the conventional case where the mask processing is performed for each sheet. Also in this case, since a common mask is generated for a group of image data having a relatively small change, it is possible to accurately extract the region of interest.

Then, the image processing unit 6 performs desired image processing on the masked image data stored in the image storage unit 1 and displays it on the display unit 7. In this case, since the image storage unit 1 has already stored a group of image data from which unnecessary portions have been removed by mask processing, the image processing in the image processing unit 6 can be any image processing. Is. That is, even if the MIP process or the 3D process at an arbitrary angle is performed, the process can be performed without being influenced by an unnecessary object such as a bone, and a desired image can be easily observed.

As described in detail above, according to the present embodiment, a common mask can be generated and masked for a large number of tomographic image data, and extremely efficient masking and Image processing can be performed. Further, since various image processing can be performed on the image data after the mask processing, an arbitrary three-dimensional image can be displayed.

[0031]

As described in detail above, by generating a common mask for image data of a large number of tomographic images and performing the mask processing, extremely efficient and accurate mask processing and image processing can be performed. It is possible to realize an image processing method and an image processing apparatus that can perform processing and perform various types of image processing on image data after mask processing to display a desired image.

[Brief description of drawings]

FIG. 1 is a flowchart showing an outline of processing contents of an image processing method according to an embodiment of the present invention.

FIG. 2 is a configuration diagram showing a configuration of an apparatus according to an embodiment of the present invention.

FIG. 3 is an explanatory diagram showing how a mask is generated according to an embodiment of the present invention.

FIG. 4 is an explanatory diagram showing how a mask is generated according to an embodiment of the present invention.

FIG. 5 is an explanatory diagram showing how a mask is generated according to an embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Image storage unit 2 Image processing device 3 0 ° MIP processing unit 4 Mask generation unit 5 Mask processing unit 6 Image processing unit 7 Display unit 8 Control unit

Claims (4)

[Claims] 1. An image processing method for executing image processing after performing mask processing on image data of a plurality of sections, wherein predetermined processing for mask generation is performed on image data to be processed, A step of generating a mask for the image data, a step of performing mask processing on all the image data using the generated mask to generate mask processed image data, and a step of using the mask processed image data And a step of performing desired image processing. 2. The image data of a plurality of cross sections is divided into a plurality of groups, a mask is generated for each group, and mask processing is performed for each of the plurality of groups using the mask thus obtained. The image processing method according to claim 1. 3. An image storage unit (1) for storing image data to be processed, and a predetermined process for mask generation is performed on the image data stored in the image storage unit (1), Mask generation means (3, 4) for generating a mask for image data.
4), and a mask processing section (5) for performing mask processing on all image data using the mask generated by the mask generation means (3, 4) to generate mask-processed image data, and a mask processing section. An image processing apparatus comprising: an image processing unit (6) for executing desired image processing using the masked image data masked in (5). 4. The mask generation means (3, 4) divides image data of a plurality of cross sections into a plurality of groups and generates a mask for each group, and the mask processing section (5) generates a mask for each group. The image processing apparatus according to claim 3, wherein the mask processing is performed for each of a plurality of groups by using the mask processing.