JP2010029269A - Image processing method, image processing apparatus and x-ray ct apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing method, an image processing apparatus and an X-ray CT apparatus capable of swiftly positioning an image even if the image has few characteristic points. <P>SOLUTION: The image processing method is characterized by measuring the dispersion value of indexes forming the image within the range of a first template formed on the first image, and forming a second template different from the first template if the measured dispersion value does not reach a prescribed value. The image processing apparatus includes an image data storing means for storing the image as image data, and an image positioning means for executing the above image processing method based on the stored image data. The X-ray CT apparatus is characterized by having the above image processing apparatus. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an image processing method, an image processing apparatus, and an X-ray CT apparatus.

  When a doctor interprets an image, a plurality of images relating to the same anatomical part are arranged, and a comparative observation is performed by examining while moving the viewpoint between observation targets in the same anatomical part. Here, a plurality of images to be compared and observed may have a time-series relationship. For example, when an image of an affected area is captured by an X-ray CT (Computed Tomography) apparatus, imaging may be performed continuously at a predetermined interval after a contrast medium is placed in the body of the subject. In addition, photographing may be performed after a period of time for follow-up observation.

  In the case of performing comparative observation by arranging images taken at different times in this way, if the positions of the images are shifted from each other, the image interpreter must perform observation while searching for an observation target. As a result, the burden on the reader may increase. In addition, there is a risk that it is difficult to diagnose details.

For this reason, there has been proposed an image diagnosis support apparatus that aligns a current image and a past image at the same anatomical site and displays them simultaneously (see Patent Document 1).
Here, in the technique disclosed in Patent Document 1, the current image and the past image are aligned based on the position of the bronchial bifurcation in the past image. When there are clear feature points in the image of the target organ or the like in this way, the feature points can be designated in advance as a search reference, and alignment can be performed by performing pattern matching.

However, for an image such as an image of a liver with few feature points, it is difficult to specify the feature points in advance as a search reference and perform alignment by pattern matching. In this case, if pattern matching is performed on the basis of an unnecessarily wide range in order to include feature points, the amount of data to be processed becomes enormous, and there is a possibility that rapid image alignment cannot be performed.
JP 2005-124895 A

  The present invention provides an image processing method, an image processing apparatus, and an X-ray CT apparatus capable of performing quick image alignment even in the case of an image having a few feature points.

  According to one aspect of the present invention, when a dispersion value of an index for forming an image is measured within a range of a first template formed on a first image, and the measured dispersion value is less than a predetermined value Provides an image processing method characterized by forming a second template different from the first template.

  According to another aspect of the present invention, an image data storage unit that stores an image as image data, an image registration unit that executes the image processing method based on the stored image data, An image processing apparatus is provided.

  According to another aspect of the present invention, there is provided an X-ray CT apparatus including the above-described image processing apparatus.

  According to the present invention, there are provided an image processing method, an image processing apparatus, and an X-ray CT apparatus capable of performing quick image alignment even in the case of an image having a few feature points.

Hereinafter, embodiments of the present invention will be illustrated with reference to the drawings. In addition, in each drawing, the same code | symbol is attached | subjected to the same component and detailed description is abbreviate | omitted suitably.
FIG. 1 is a schematic configuration diagram for illustrating an image processing apparatus according to an embodiment of the present invention. FIG. 1 illustrates an image processing apparatus that can be provided in an X-ray CT apparatus as an example of an image processing apparatus according to an embodiment of the present invention.

  As shown in FIG. 1, the image processing apparatus 1 includes an image data storage unit 2, an image alignment unit 3, and a display unit 4.

  In this case, the image processing apparatus 1 may be built in an X-ray CT apparatus 100 described later, or may be provided separately from the X-ray CT apparatus so that data can be exchanged between them. . The X-ray CT apparatus 100 in which the image processing apparatus 1 is built will be described later.

  The image data storage unit 2 stores CT (Computed Tomography) tomographic images captured by the X-ray CT apparatus 100 as CT tomographic image data. The stored CT tomographic image data includes a plurality of CT tomographic image data relating to the same anatomical site of the same subject. In addition, a plurality of CT tomographic image data relating to the same anatomical part of the same subject includes data having a time-series relationship. Examples of CT tomographic image data with a time-series relationship include CT tomographic image data continuously taken at predetermined intervals after a contrast medium is placed in the body of a subject, follow-up observations, periodic examinations, etc. Thus, CT tomographic image data taken with a predetermined period can be exemplified.

  When the X-ray CT apparatus 100 and the image processing apparatus 1 are provided separately, CT tomographic image data is input and stored via a communication line or a recording medium. The image processing apparatus 1 is provided with an image scanner or the like, and a CT tomogram such as the liver is converted into image data using the image scanner or the like, and the converted data is stored in the image data storage means 2 as CT tomogram data. You can also

  The image alignment means 3 aligns a plurality of CT tomographic images relating to the same anatomical site of the same subject stored as CT tomographic image data in the image data storage means 2. The alignment of CT tomographic images will be described later.

  The display means 4 can be, for example, a flat panel display device, a CRT (Cathode Ray Tube) display device, or the like. Further, when the image processing apparatus 1 is built in an X-ray CT apparatus 100 (see FIG. 6) to be described later, the display unit 116 provided in the X-ray CT apparatus 100 can also be used.

  Although not necessarily required, an input unit (not shown) for inputting various data such as data for specifying a subject, an illustration for storing data related to a positional shift obtained by the image alignment unit 3 and the like. It is also possible to provide an output unit (not shown) for outputting data relating to the positional deviation obtained by the image storing unit 3 and the image registration unit 3 to the outside. In this case, for example, a keyboard or a mouse can be provided as input means (not shown).

Next, the operation of the image processing apparatus 1 is illustrated.
CT tomographic image data obtained by imaging the subject with the X-ray CT apparatus 100 is stored in the image data storage means 2. Further, based on the stored CT tomographic image data, a CT tomographic image to be displayed on the display unit 4 is formed by the image alignment unit 3, and the formed CT tomographic image is aligned. The CT tomographic image displayed on the display unit 4 can be appropriately selected by inputting data for specifying a subject or an anatomical site from an input unit (not shown), for example. Then, the plurality of CT tomographic images that have been aligned are displayed side by side on the display means 4.

  Here, according to X-ray CT, for example, bones and muscles have high CT values, and fat has a low CT value between water and air. The air portion has the lowest CT value. And in the part with high CT value, the brightness of the image of the part becomes bright. Therefore, depending on the anatomical region, CT tomographic image data (CT tomographic image data having a characteristic part) having a part having a specific CT value or a specific brightness may be obtained. On the other hand, depending on the anatomical site, there may be obtained only CT tomographic image data (CT tomographic image data with few characteristic parts) with a small part having a specific CT value or specific brightness.

FIG. 2 is a schematic diagram for illustrating a CT tomographic image of a lung.
Much of the lung volume is occupied by the alveoli, which consists of alveolar spaces that store gas and the surrounding alveolar epithelium. Therefore, in the CT tomographic image of the lung, the alveoli are a part having a specific CT value and a specific brightness. By using the characteristic part of the alveoli as a reference, it is possible to align a plurality of CT tomographic images.

  For example, based on the characteristic part included in the range of the template A1 in FIG. 2A and the characteristic part included in the range of the template A2 in FIG. The lung CT tomographic image shown in a) and the lung CT tomographic image shown in FIG. 2B can be aligned.

FIG. 3 is a schematic diagram for illustrating a CT tomographic image of the liver.
The liver tissue is composed of structural units called liver lobules, and the CT tomographic image has few high contrast portions. That is, it becomes a CT tomographic image having many portions with low contrast and almost uniform brightness. Therefore, it is difficult to align a plurality of CT tomographic images with few characteristic parts.

  For example, the portion included in the range of the template B1 in FIG. 3A and the portion included in the range of the template B2 in FIG. Then, it is not possible to align the CT tomographic image of the liver shown in FIG. 3A and the CT tomographic image of the liver shown in FIG. That is, there is no characteristic pattern with high contrast that serves as a mark for aligning the two images. In this case, if pattern matching is performed on the basis of an unnecessarily wide range in order to include feature points, the amount of data to be processed becomes enormous and rapid image alignment cannot be performed.

  In the present embodiment, when positioning is performed by pattern matching, whether or not a pattern having a high contrast serving as a mark is included in a template range is determined based on a variance value in an index for forming an image. Then, after confirming that a pattern having a characteristic part is included by expanding the range of the template when the variance value is small, that is, when a pattern with high contrast is not included, CT The alignment of tomographic images is performed.

  That is, the dispersion value of the index that forms the image within the range of the template created on the reference image is measured, and a pattern with high contrast is included in the template range based on the measured dispersion value. Determine whether or not. Here, as an index for forming an image, for example, at least one of hue, saturation, and brightness of the image can be used. Alternatively, RGB (red green blue) values forming an image may be used as an index. A large variance value of these indexes is related to a high contrast of the portion. If there is a pattern with high image contrast, it is easy to align the image using the pattern as a mark.

  When it is determined that a characteristic pattern having a small variance value, that is, a high contrast is not included, the range of the template is expanded. Alternatively, the position of the template may be changed. That is, at another location of the image, the variance value of the index is evaluated to check whether there is a portion with a high variance value, that is, a high contrast. On the other hand, when a high-contrast, that is, characteristic pattern is included, the pattern is extracted, and the CT tomographic image is aligned using the extracted pattern as a mark.

  Further, when the template range is expanded, the dispersion value of the index within the expanded template range is measured, and the high-contrast portion in the expanded template range based on the measured dispersion value, that is, It is determined whether or not a characteristic pattern is included.

On the other hand, even when the position of the template is changed, the variance value of the index within the template range at the position after the change is measured, and the contrast value is within the range of the template changed based on the measured variance value. It is determined whether or not a high portion, that is, a characteristic pattern is included.
When it is determined that a high contrast portion, that is, a characteristic pattern is included, the pattern is extracted. Then, a pattern corresponding to the extracted pattern is searched in the second image. That is, the second image is searched for a pattern having a characteristic portion in the image that serves as an alignment mark.

When a pattern is searched for in the second image, pattern matching is performed on the first image and the second image, so that the position shift of the second image to be aligned with the reference first image is performed. The quantity is measured. Based on the measured positional deviation amount, the position of the first image serving as a reference is aligned with the position of the second image serving as an alignment target.
If it is determined that a pattern having a characteristic portion is not included, the range of the enlarged template is further enlarged or the position is changed, and the above-described procedure is repeated.

  For example, as shown in FIG. 3C, when a characteristic part is not included in the range of the template B1, the range of the template is expanded to B3 and a characteristic part is included in the range. Make sure that Then, a plurality of CT tomographic images are aligned based on the characteristic part C included in the enlarged range of the template B3. Note that whether or not a characteristic portion is included in the template range can be determined by, for example, providing a threshold value or the like for the luminance dispersion value as described later.

Next, the image processing method according to the embodiment of the present invention will be further illustrated.
FIG. 4 is a flowchart for illustrating the image processing method according to the embodiment of the present invention.
FIG. 5 is a schematic diagram for illustrating the state of image processing. FIG. 5 illustrates image processing in a CT tomographic image as an example of the image processing method according to the embodiment of the present invention. In addition, arrows XYZ in FIG. 5 indicate three directions orthogonal to each other, XY indicates a cross-sectional direction, and Z indicates a body axis direction.

  First, as shown in FIG. 5A, the reference image is divided into a grid pattern (step S1). The reference image can be arbitrarily selected. For example, it can be a CT tomographic image taken first. Also, the number of divisions can be appropriately changed depending on the size of the target image. Note that the image to be aligned is similarly divided into a grid.

Next, as shown in FIG. 5B, a template D serving as a reference is created at the position of the lattice point (step S2).
The grid points for creating the template D can be arbitrarily selected. When a position where a characteristic part is present can be predicted in advance, it is preferable to select a lattice point in the vicinity thereof.

Next, the dispersion value of the index in the template D serving as a reference is measured (step S3). For example, a luminance dispersion value as an index is measured. In addition to the luminance (or lightness), as described above, the dispersion value of at least one index such as hue, saturation, or RGB value may be measured.
Next, based on the measured dispersion value, it is determined whether or not an appropriate pattern to be used for alignment is included in the range of the template D serving as a reference (step S4).
That is, it is determined whether or not a pattern having a characteristic portion is included in the range of the template D serving as a reference. In this case, for example, it is possible to determine whether or not an appropriate pattern is included (whether or not a pattern having a characteristic part is included) by measuring the luminance dispersion value using the index as the luminance. it can.

  In the portion where the CT value is high, the brightness of the image of that portion becomes bright, and in the portion where the CT value is low, the brightness of the image of that portion becomes dark. For this reason, if there is a characteristic part (changed part) within the range defined by the template, the luminance dispersion value also varies.

  If such a variation in the variance value of luminance is obtained, whether or not an appropriate pattern is included based on a predetermined threshold or the like (whether or not a pattern having a characteristic part is included). ) Can be determined.

  Here, when it is determined that an appropriate pattern is not included (a pattern having a characteristic part is not included), the range of the template is expanded as in template E shown in FIG. Then, the determination in step S4 is performed again.

In addition, when it is determined that an appropriate pattern is included (a pattern having a characteristic part is included), as shown in FIG. And the amount of positional deviation is measured by performing pattern matching (step S5).
That is, the amount of positional deviation at the position of the lattice point on the image to be aligned corresponding to the lattice point on which the template on the reference image is created is measured by performing pattern matching.

Next, the positions of other grid points on the image to be aligned are interpolated (step S6). Next, based on the positional shift amount measured in step S5 and the interpolation of the positions of other grid points performed in step S6, the positional shift amount at all grid points is obtained (step S7).
Then, based on the obtained misregistration amounts at all grid points, the entire image to be aligned is aligned as shown in FIG. 5C (step S8).

  According to the image processing method and the image processing apparatus according to the present embodiment, quick alignment can be performed even for an image with few characteristic parts (for example, a CT tomographic image of the liver). In addition, in the case of performing comparative observation by arranging CT tomographic images, the burden on the image interpreter can be reduced. Moreover, since it becomes easy to perform comparative observation, it is easy to make a detailed diagnosis.

Next, an X-ray CT apparatus according to an embodiment of the present invention will be illustrated.
FIG. 6 is a schematic block diagram for illustrating a schematic configuration of the X-ray CT apparatus according to the embodiment of the present invention.
As shown in FIG. 6, the X-ray CT apparatus 100 includes an imaging unit 100a and a processing / display unit 100b.

  The imaging means 100a exposes the subject to X-rays, detects the X-rays that have passed through the subject, and acquires projection data (or raw data). In general, the imaging means is a rotation / rotation (ROTATE / ROTATE) type in which an X-ray tube and a two-dimensional detector system are integrally rotated around a subject, and a large number of detection elements are attached in a ring shape. Various types, such as a fixed / rotated (STATIONARY / ROTATE) type in which only the X-ray tube rotates around the subject, and a type in which the position of the X-ray source is electronically moved on the target by deflecting the electron beam However, any type is acceptable. Here, for convenience of explanation, a rotation / rotation type X-ray CT apparatus will be described as an example.

  As shown in FIG. 6, the imaging unit 100a includes an X-ray tube 101, a rotating ring 102, a two-dimensional detector system 103, a data acquisition circuit (DAS) 104, a non-contact data transmission device 105, a gantry driving unit 107, a slip A ring 108 and a radiation detector (not shown) are provided.

An X-ray tube 101 that is an X-ray source is a vacuum tube that generates X-rays, and is provided on the rotating ring 102. The X-ray tube 101 is supplied with power (tube current, tube voltage) necessary for X-ray exposure from the high voltage generator 109 via the slip ring 108. The X-ray tube 101 emits X-rays toward a subject in the effective visual field region FOV by causing electrons accelerated by the supplied high voltage to collide with the target.
An X-ray (not shown) that shapes the shape of the X-ray beam exposed from the X-ray tube 101 into a cone shape (quadrangular pyramid shape) or a fan beam shape between the X-ray tube 101 and the subject. A tube side collimator is provided.

The two-dimensional detector system 103 is a detector system that detects X-rays transmitted through the subject, and is provided on the rotating ring 102 so as to face the X-ray tube 101. A plurality of radiation detectors (not shown) are attached to the two-dimensional detector system 103.
The X-ray tube 101 and the two-dimensional detector system 103 are provided on the rotating ring 102. The rotating ring 102 is driven by the gantry driving unit 107 and rotates around the subject.

  The data acquisition circuit (DAS) 104 has a plurality of data acquisition element arrays in which DAS chips are arranged, and receives data detected by the two-dimensional detector system 103 (hereinafter referred to as raw data). The input raw data is subjected to amplification processing, A / D conversion processing, and the like, and then transmitted to the preprocessing device 106 provided in the processing / display unit 100b via the data transmission device 105.

  The gantry driving unit 107 integrally rotates the X-ray tube 101 and the two-dimensional detector system 103 around a central axis parallel to the body axis direction of the subject inserted into the diagnostic aperture. Drive and control it.

  Next, the processing / display unit 100b is illustrated. The processing / display unit 100b includes a preprocessing unit 106, a high voltage generation unit 109, a host controller 110, a storage unit 111, a reconstruction unit 114, an input unit 115, a display unit 116, an image processing unit 118, a network communication unit 119, data. / A control bus 300 is provided. Further, as will be described later, the X-ray CT apparatus 100 according to the present embodiment also includes the functions of the elements included in the image processing apparatus 1 according to the present embodiment. Each element included in the image processing apparatus 1 can be connected to the data / control bus 300, or can be connected to the image processing apparatus 1 provided separately from the X-ray CT apparatus via the network communication apparatus 119. It can also be.

The preprocessing device 106 receives raw data from the data acquisition circuit (DAS) 104 via the data transmission device 105, and executes sensitivity correction and X-ray intensity correction. Note that the raw data preprocessed by the preprocessing device 106 is referred to as “projection data”.
The high voltage generator 109 supplies power necessary for X-ray exposure to the X-ray tube 101 via the slip ring 108. The high voltage generator 109 includes a high voltage transformer, a filament heating converter, a rectifier, a high voltage switch, and the like.
The host controller 110 performs overall control related to various processing such as photographing processing, data processing, and image processing according to the present embodiment.

The storage unit 111 stores image data such as collected raw data, projection data, and CT tomographic image data. That is, it also has the function of the image data storage means 2 described above.
The reconstruction device 114 performs reconstruction processing on projection data based on predetermined reconstruction parameters (reconstruction area size, reconstruction matrix size, threshold for extracting a region of interest, etc.), and thereby a predetermined slice. Create reconstructed image data. In general, reconstruction processing includes cone beam reconstruction (Feldkamp method, ASSR method, etc.) and fan beam reconstruction, and any method can be used.

  The input means 115 is provided with a keyboard, various switches, a mouse, and the like, and various scanning conditions such as slice thickness and the number of slices can be input by an operator.

  The image processing unit 118 performs image processing for display, such as window conversion and RGB processing, on the reconstructed image data created by the reconstructing device 114 and outputs the image processing to the display unit 116. Further, the image processing unit 118 creates a so-called pseudo three-dimensional image such as a tomographic image of an arbitrary cross section, a projection image from an arbitrary direction, or a three-dimensional surface image based on a command from the operator, and outputs it to the display unit 116. To do.

Further, based on a command from the operator, a plurality of CT tomographic images relating to the same anatomical site of the same subject displayed on the display unit 116 are aligned. In other words, the image processing unit 118 also has the function of the image positioning means 3 described above. The display means 116 also has the function of the display means 4 described above.
The image data output from the image processing unit 118 is displayed on the display unit 116 as a CT tomographic image.

  The network communication device 119 transmits / receives various data to / from other devices and a network system such as RIS (Ragiology Information System) via the network. The data / control bus 300 is a signal line for connecting various devices and transmitting / receiving various data, control signals, address information, and the like.

  In addition, since the known technique in the X-ray CT apparatus can be applied to elements other than the elements (for example, the image alignment means 3) provided in the image processing apparatus 1 described above, description of the operation thereof is omitted. To do.

The embodiment of the present invention has been illustrated above. However, the present invention is not limited to these descriptions.
As long as the features of the present invention are provided, those skilled in the art appropriately modified the design of the above-described embodiments are also included in the scope of the present invention.
For example, the shape, size, material, arrangement, and the like of each element included in the image processing apparatus 1, the X-ray CT apparatus 100, and the like are not limited to those illustrated, and can be changed as appropriate.
In addition, the case where the image processing method and the image processing apparatus according to the embodiment of the present invention are applied to the alignment of CT tomographic images has been described as an example, but the present invention is not limited thereto. For example, the image processing method and the image processing apparatus according to the embodiment of the present invention can be applied also in an inspection process of an electronic device such as a semiconductor device or a flat panel display. That is, the present invention can be widely applied when an image having a small number of feature points is aligned by pattern matching.

  Moreover, each element with which each embodiment mentioned above is combined can be combined as much as possible, and what combined these is also included in the scope of the present invention as long as the characteristics of the present invention are included.

1 is a schematic configuration diagram for illustrating an image processing apparatus according to an embodiment of the present invention. It is a schematic diagram for illustrating a CT tomographic image of the lung. It is a schematic diagram for illustrating a CT tomographic image of the liver. It is a flowchart for demonstrating about the image processing method which concerns on embodiment of this invention. It is a schematic diagram for demonstrating the mode of image processing. 1 is a schematic block diagram for illustrating a schematic configuration of an X-ray CT apparatus according to an embodiment of the present invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image processing apparatus, 2 Image data storage means, 3 Image alignment means, 4 Display means, 100 X-ray CT apparatus, 100a Imaging means, 100b Processing / display means, 111 Storage means, 116 Display means, 118 Image processing part

Claims (12)

Measuring a dispersion value of an index for forming an image within a range of the first template formed on the first image;
An image processing method comprising: forming a second template different from the first template when the measured variance value is less than a predetermined value.   The image processing method according to claim 1, wherein the second template is obtained by enlarging a range of the first template.   The image processing method according to claim 1, wherein the second template includes a portion different from the first template. Measuring a variance value of the indicator within the range of the second template;
When the measured variance value reaches a predetermined value, a pattern included in the second template is extracted,
Searching the second image for a pattern corresponding to the extracted pattern;
Measuring a positional deviation amount of the second image with respect to the first image by performing pattern matching on the first image and the second image;
The image processing according to any one of claims 1 to 3, wherein the position of the second image and the position of the first image are matched based on the measured positional deviation amount. Method. The first image and the second image are each divided into a grid pattern,
Creating the template at the position of an arbitrary grid point on the first image;
Measuring the amount of positional deviation at the position of the grid point on the second image corresponding to the grid point at which the template is created by performing the pattern matching;
Based on the measured positional deviation amount, by interpolating the position of another grid point on the second image, the position of the second image and the position of the first image are matched,
The image processing method according to claim 4. Measuring a variance value of the indicator within the range of the second template;
The image processing method according to claim 1, further comprising: forming a third template different from the second template when the measured variance value is less than a predetermined value.   The image processing method according to claim 6, wherein the third template is obtained by enlarging a range of the second template.   The image processing method according to claim 6, wherein the third template includes a portion different from the second template.   The image processing method according to claim 1, wherein the index is at least one of hue, saturation, and brightness.   The image processing method according to claim 1, wherein the image is a CT tomographic image. Image data storage means for storing an image as image data;
An image alignment unit that executes the image processing method according to any one of claims 1 to 9 based on the stored image data;
An image processing apparatus comprising: An X-ray CT apparatus comprising the image processing apparatus according to claim 11.

Images