JP4361268B2 - 3D image display device for directly creating a 3D image from projection data of an X-ray CT apparatus - Google Patents

Description

[0001]
[Technology to which the invention belongs]
The present invention relates to a three-dimensional image display apparatus that directly creates a three-dimensional image from projection data of an X-ray CT apparatus.
[0002]
[Prior art]
The X-ray CT apparatus irradiates the subject with X-rays from an X-ray tube that circulates around the subject lying on the top of the bed, and the intensity of the X-rays transmitted through the subject is detected by an X-ray detector. The projection data of the cross-section of the subject is collected by detecting at. Next, after moving the couch top, similarly, the projection data of the cross section of the subject is collected. By repeating the movement of the top plate and the collection of projection data, projection data of a plurality of cross sections of the subject are collected. The image data of the plurality of cross sections of the subject is obtained by performing image reconstruction processing on the collected projection data of the plurality of cross sections of the subject by the high-speed data processing apparatus.
[0003]
As a method of using the data inspected by the X-ray CT apparatus, a plurality of cross-sectional projection data are collected by the X-ray CT apparatus, and the high-speed data processing apparatus performs image reconstruction processing on the plurality of cross-sectional projection data. Create cross-sectional image data, display this cross-sectional image data on the image display device, burn the multi-cross-sectional image displayed with a laser imager, etc. on the film, and use the film observer to display the multi-cross-section image. By observing the baked film, it has been widely practiced to find an abnormal part of the subject. In the early X-ray CT apparatus, the number of reconstructed image data was 10 to several tens, so such a method was possible.
[0004]
As a method of using data inspected by an X-ray CT apparatus, recently, projection data of a plurality of cross sections is collected by an X-ray CT apparatus, and image reconstruction processing is performed on the projection data of the plurality of cross sections by a high-speed data processing apparatus. The image data of a plurality of cross sections is prepared to store the image data of the plurality of cross sections in a storage medium such as a magnetic disk device, a magneto-optical disk device, or an optical disk device. Reading out image data of a plurality of cross sections stored in the storage medium, displaying the image data on an image display device, and observing is performed. In some cases, image data of a plurality of cross sections stored in the storage medium is read out and image processing is performed on the image data by an image processing apparatus. For example, a three-dimensional image is constructed from this image data by a three-dimensional image display device, and the three-dimensional image is observed.
[0005]
In the early X-ray CT apparatus, the movement of the top plate on which the subject lay down and the collection of the projection data of the subject by X-ray irradiation were performed alternately. The technology of the X-ray CT apparatus has advanced rapidly, and in recent X-ray CT apparatuses, the X-ray irradiation is simultaneously performed while the subject lies on the tabletop and the projection data is collected. Such a projection data collection method is called helical scan.
[0006]
In the early X-ray CT apparatus, the movement of the table on which the subject lay down and the acquisition of the projection data of the subject by X-ray irradiation were performed alternately. It was a cross section of the image data obtained by performing the configuration. That is, the cross-section from which the projection data was collected and the cross-section of the image data obtained by image reconstruction coincide with each other, and the number of cross-sections from which projection data was collected and the image data obtained by image reconstruction. The position of the cross-section where the projection data was collected was distributed discretely along the patient's body axis. However, in the helical scan method, the projection data is collected by performing X-ray irradiation at the same time while the subject moves the tabletop lying down, so there is no discrete projection data collection cross section. Therefore, in general, a virtual cross section is set and projection data is interpolated to obtain virtual cross section projection data, and image reconstruction processing is performed using the virtual cross section projection data. Is going. Since the virtual cross section of this projection data can be set quite freely, the number of image data obtained by image reconstruction processing and the interval between the cross sections can be selected fairly freely depending on the setting of the virtual cross section of the projection data. it can.
[0007]
In early X-ray CT systems, X-ray detectors that detect the intensity of X-rays that have passed through a subject collect projection data of one cross section of the subject in a row in the body axis direction of the subject. Was. The technology of X-ray CT equipment has advanced rapidly. In recent X-ray CT equipment, multiple rows of X-ray detectors are arranged in the direction of the body axis of the subject, and projection data of multiple cross sections of the subject can be simultaneously transmitted. Use multi-row detector to collect. By using this multi-row detector and the helical scan method at the same time, much more projection data than the initial X-ray CT apparatus has been collected. In this multi-row detector / helical scan system, there is no discrete projection data acquisition cross section as in the early X-ray CT system. Generally, a virtual cross section is set and projection data interpolation processing is performed. The virtual cross-section projection data is obtained and image reconstruction processing is performed using the virtual cross-section projection data. Since the virtual cross section of the projection data can be set freely, the number of image data obtained by the image reconstruction process and the interval between the cross sections can be freely selected depending on the setting of the virtual cross section of the projection data.
[0008]
In an examination using an X-ray CT apparatus, currently, projection data is collected by an X-ray CT apparatus, an image reconstruction process is performed on the projection data by a high-speed data processing apparatus, and obtained as a result of the image reconstruction process. Image data of a plurality of cross sections is stored in a storage medium such as a magnetic disk device, a magneto-optical disk device, or an optical disk device. There are two methods for using the results of inspection with an X-ray CT apparatus: a method of printing an image created using image data of a plurality of cross sections obtained as a result of image reconstruction processing on a film, and observing the film; A method for displaying and observing an image on an image observation apparatus using image data of a plurality of cross sections obtained as a result of the reconstruction process, and an image for the image data of a plurality of cross sections obtained as a result of the image reconstruction process There is a method of performing image processing such as three-dimensional image processing with a processing device and observing the result. There are optimal image reconstruction parameters for each method of use, but currently the data stored for a long time is image data after image reconstruction, and long-term storage of projection data before image reconstruction is Reconstruction processing is performed again using the optimal image reconstruction processing parameters for each method of use due to the fact that it has not been performed, and that image reconstruction from projection data requires large computational resources. This is not done for research purposes.
[0009]
In particular, when constructing a 3D image using multiple cross-sectional image data, it is suitable for 3D image processing that is different from the image reconstruction processing parameters suitable for direct observation of the cross-sectional image. Although good results are often obtained by using image data of multiple cross-sections that have been reconstructed using image reconstruction processing parameters, it is generally stored for a long time after image reconstruction Since only the image data is stored and the projection data is not stored for a long time, such image reconstruction processing for each purpose of use cannot be performed except for research purposes.
[0010]
Even if the X-ray CT apparatus stores projection data for a long period of time, the X-ray CT apparatus routinely performs a large number of examinations continuously, so an external user can use this projection data. On the other hand, it is almost impossible to perform image reconstruction using the image reconstruction parameters desired by the user. For this reason, the image reconstruction of the projection data is performed by a predetermined image reconstruction parameter in the routine inspection.
[0011]
In recent X-ray CT apparatuses, there are models that can perform advanced image processing such as three-dimensional image processing on image data obtained by image reconstruction. However, even with such a function, an X-ray CT apparatus routinely performs a large number of examinations almost continuously, so an external user can use this function to perform 3D image processing. It is almost impossible to do in reality.
[0012]
In recent multi-row detectors and helical scan CT devices, compared to earlier CT devices, the noise of the X-ray detector is reduced, the spatial density of the X-ray detector is improved, and the helical scan pitch density is increased. , The spatial distribution of collected projection data is fine. For this reason, even if the spatial region for image reconstruction is reduced while maintaining the number of pixels for image reconstruction, a precise image with little increase in noise can be obtained. Therefore, by reducing the image reconstruction area for the same projection data, the spatial accuracy of the image of the region of interest of the subject can be improved. Moreover, even if the interval between the image reconstruction planes is narrowed, a meaningful image with little increase in noise can be obtained, so that the spatial resolution in the body axis direction can also be improved.
[0013]
With recent multi-row detectors and helical scan CT devices, the spatial distribution of the projection data collected has become finer, so CT values can be obtained not only within the cross-section of the subject but also at spatial positions with fine spacing in the body axis direction. It became possible to ask for. In a three-dimensional image display device using CT image data, voxel data is created by stacking cross-sectional image data obtained by image reconstruction of projection data in the body axis direction, and three-dimensional reconstruction processing is performed on this. 3D image is created and displayed. By using CT image data obtained by a recent multi-row detector / helical scan type CT apparatus, it is possible to obtain a very precise three-dimensional image.
[0014]
In order to create a 3D image from CT image data, projection data is reconstructed with an X-ray CT apparatus to create cross-sectional image data, and voxel data is created by stacking this image data in the body axis direction. . When 512 pieces of image data having a cross-sectional number of pixels of, for example, 512 × 512 pixels and a pixel size of 0.5 mm × 0.5 mm are stacked in the body axis direction, for example, at intervals of 0.5 mm, a spatial region of 256 mm × 256 mm × 256 mm is obtained. A cube of 512 × 512 × 512 voxels is created. Next, a three-dimensional image is created and displayed by performing three-dimensional reconstruction processing such as surface rendering and volume rendering on the cube of voxels.
[0015]
As described above, the conventional three-dimensional display apparatus generates CT image data by reconstructing an image of projection data with an X-ray CT apparatus, and uses this to construct voxel data. When interpreting a two-dimensional image of a CT image, a radiologist arranges a series of CT images representing a cross section of a subject in a plane and arranges them for observation, thereby interpreting the presence or absence of an abnormality. On the other hand, when interpreting as a three-dimensional image, voxel data is constructed from this series of CT image data, and opacity is given to the voxel based on the physical properties of each voxel constituting the voxel data, that is, the CT value. Given the direction of the line of sight of the observer observing this voxel data and the direction of the light source, the attenuation and reflection due to the light from the light source passing through the object are calculated from the opacity, and this is integrated to create an image To do. Therefore, integral calculation corresponding to the number of voxels of the image to be displayed is required, and the volume rendering calculation time increases as the number of voxels increases. In other words, the volume rendering method assigns opacity, which is a light transmission characteristic, to each CT value, and manages light source attenuation along the line of sight with all voxels. The gray level gradient calculated by the gray level gradient method at all points of the volume Based on the attachment, the luminance value is calculated by multiplying the amount of the incident light source and the opacity of the voxel. A three-dimensional image is obtained by sequentially integrating these in the line-of-sight direction (ray casting). In the volume rendering method, natural and smooth shading can be obtained even at an edge where the CT value changes rapidly, and the ability to depict fine tissues such as peripheral blood vessels is dramatically improved. Conventionally, surface rendering has been the mainstream for 3D display, but volume rendering is now widely used.
[0016]
Thus, conventional 3D display devices use a plurality of CT image data obtained by scanning a subject with an X-ray CT apparatus, collecting projection data, and reconstructing the projection data. ing. When observing a CT image as a two-dimensional image, density resolution rather than spatial resolution is important. On the other hand, when reconstructing a three-dimensional image from CT image data and observing, spatial resolution is more important than density resolution. When observing a CT image as a two-dimensional image, such as when resolution is required, and when reconstructing a three-dimensional image, the optimal parameters for reconstructing projection data are different. There are many cases. However, at present, the CT image data is reconstructed using the optimum image reconstruction parameters when the projection data is observed as a two-dimensional image in the X-ray CT apparatus, and the two-dimensional image is displayed in the three-dimensional image display apparatus. Three-dimensional image reconstruction is performed using CT image data optimized for observation. This is because a large computer resource is required to reconstruct the projection data, and a large storage capacity is required to store the projection data.
[0017]
In the X-ray CT system,
(1) From an X-ray source that moves on the outer periphery of the cross section of the subject, X-rays are radiated in a fan shape along the plane of the cross section, and the X-rays that have passed through the subject are, for example, 500 X-ray detectors Measure with Thereby, for example, 500 pieces of data are measured at one source position. By repeating this over 180 degrees or more of the outer periphery of the cross section of the subject, projection data is collected. For example, if it is repeated every time, data in directions of 180 or more are collected, so that projection data of 500 × 180 = 90000 is collected.
(2) The collected projection data is subjected to preprocessing such as noise removal, and then subjected to convolution processing.
(3) A pixel plane of 512 × 512 pixels composed of, for example, square pixels of 1 mm × 1 mm is set at a position corresponding to the cross section of the subject from which projection data has been collected, and for each pixel of this pixel plane Then, the projection data subjected to the convolution process from the position of the X-ray source at the time of collecting the projection data is back-projected in a fan shape.
(4) At this time, since the backprojected data does not necessarily cross the center of the pixel on the pixel plane, interpolation processing is performed using the backprojected data in the vicinity, and approximate data is assigned to each pixel.
(5) By repeatedly performing this backprojection process on all collected projection data, each pixel on the pixel plane of 512 pixels × 512 pixels has a value corresponding to the physical property of the subject with respect to X-rays. The image data it has is reconstructed.
(6) The position of the cross section of the subject is moved by, for example, 1 mm in the body axis direction, the projection data is collected in the same manner, and image reconstruction processing is performed on the projection data to reconstruct the cross section image data in the same manner. .
(7) By repeating this while moving the position of the cross section of the subject in the body axis direction, 500 pieces of image data of the cross section of the subject can be collected at intervals of 1 mm, for example. Using this data set, a voxel space of 512 × 512 × 500 voxels composed of, for example, cubic pixels (voxels) of 1 mm × 1 mm × 1 mm can be created.
(8) A 3D image is created and displayed by performing 3D image processing such as volume rendering on the voxel space created in this way.
[0018]
FIG. 1 is a block diagram schematically showing a three-dimensional display device connected to a conventional X-ray CT apparatus via a network. Reference numeral 101 denotes an X-ray CT apparatus, 111 denotes a scanner unit of the X-ray CT apparatus, and 112 denotes a data collection unit. The projection data collected by detecting the scan of the subject by the X-ray source of the scanner unit by the X-ray detector is digitized by the data acquisition unit. The digitized X-ray detector output data 122 is sent to a preprocessing unit 113 where preprocessing such as data noise removal and calibration is performed. The projection data 123 preprocessed by the preprocessing unit 113 is sent to the image reconstruction device 114. The image reconstruction device 114 performs an image reconstruction process on the preprocessed projection data 123. The reconstructed image data 124 is sent to and displayed on the console 115 of the X-ray CT apparatus and is stored in the image data storage device 116. Reference numeral 221 denotes image data sent to an external three-dimensional image display device or the like.
[0019]
Reference numeral 201 denotes a three-dimensional image display device. An image data storage device 211 stores image data 221 sent from the X-ray CT apparatus. Reference numeral 212 denotes a three-dimensional image processing apparatus. The three-dimensional image processing device 212 receives the image data 222 designated by the operator from the image data storage device 211, performs three-dimensional image reconstruction such as volume rendering on the image data storage device 211, and the created three-dimensional image 223 is displayed on the console 213. indicate.
[0020]
As shown in this example, a user of a conventional 3D image display apparatus creates a 3D image using image data reconstructed by an X-ray CT apparatus. It is not realized in the routine that a new image reconstruction process is performed using the projection data, and a three-dimensional image is created using the newly reconstructed image data.
[0021]
[Problems to be solved by the invention]
In the inspection using the X-ray CT apparatus, the projection data was collected by the X-ray CT apparatus, the image reconstruction process was performed on the projection data by the high-speed data processing apparatus, and the result of the image reconstruction process was obtained. Image data of a plurality of cross sections is stored in a storage medium such as a magnetic disk device, a magneto-optical disk device, or an optical disk device. There are appropriate image reconstruction parameters for each of the methods that use the results of inspection by the X-ray CT apparatus, but currently, the image data that has been stored for a long period of time is the image data after the image reconstruction processing. Since the data is not stored for a long period of time, it is not performed to perform the image reconstruction process with the optimal image reconstruction process parameter for each usage method. In particular, when constructing a 3D image using multiple cross-sectional image data, it is suitable for 3D image processing that is different from the image reconstruction processing parameters suitable for direct observation of the cross-sectional image. In many cases, it is desired to use image data reconstructed using image reconstruction processing parameters, but such image reconstruction processing for each purpose of use is not generally performed. The reason for this is that a large storage capacity is required to store projection data, so that projection data is not stored for a long period of time. Since it is not used by external users, the image reconstruction device cannot be used to reconstruct the projection data for 3D images, and a large computer is needed to reconstruct the projection data. Depending on the need for resources.
[0022]
[Means for Solving the Problems]
In order to solve the above problems, the present invention is a mechanism for storing projection data acquired by X-ray scanning by connecting to an X-ray CT apparatus, and a mechanism for creating and displaying a three-dimensional image using the projection data. A three-dimensional image display device that can directly create and display a three-dimensional image from projection data without using slice or volume image data obtained by image reconstruction processing in an X-ray CT apparatus Realized. In this three-dimensional image display device, a mechanism for setting a spatial region of interest to be displayed as a three-dimensional image, that is, a spatial region for creating a three-dimensional image by three-dimensional image reconstruction, and a subject used for three-dimensional image reconstruction And a mechanism for setting in advance the spatial distribution of a plurality of points for obtaining the physical quantity, that is, the CT value, and the spatial position thereof. Reconstruction processing is performed on the designated spatial position in the designated space of interest using projection data, and the CT value of the subject is acquired. A three-dimensional image reconstruction process is performed on the CT value data set thus obtained, and the obtained result is displayed as a three-dimensional image.
[0023]
In conventional 3D image creation and display, slice data with different cross-sectional positions is reconstructed from projection data in an X-ray CT apparatus, and 3D voxel data is created by stacking slice data with different cross-sectional positions. , 3D image reconstruction processing such as volume rendering processing is performed on the 3D voxel data to create and display a 3D image. On the other hand, in the 3D image display apparatus according to the present invention, a mechanism for setting a space area of interest to be displayed as a 3D image, that is, a space area of interest for performing 3D image reconstruction, is used for 3D image reconstruction. A mechanism for setting in advance the spatial distribution of points for obtaining physical quantities representing the characteristics of the object, that is, CT values, and their spatial positions was provided. Reconstruction processing is performed on the designated spatial position in the designated space of interest using projection data, and a physical quantity representing the characteristics of the subject, that is, a CT value is acquired. A three-dimensional image reconstruction process is performed on the CT value data set in the region of interest obtained in this way, and the obtained result is displayed as a three-dimensional image.
[0024]
Further, a physical quantity representing the characteristics of the object, that is, a spatial region of interest for obtaining a CT value data set, a spatial distribution of a plurality of points for obtaining CT values, and their spatial positions are three-dimensionally obtained using this CT value data set. By considering the effect on the image quality of the 3D image obtained by image reconstruction, the space of interest for reconstructing the image of the projection data to obtain the CT value data set and the image reconstruction of the projection data A mechanism is provided that makes it possible to determine the spatial distribution and the spatial position of a plurality of points that are obtained to obtain a CT value.
[0025]
In the conventional X-ray CT apparatus, image reconstruction using the projection data is performed for all regions where the projection data exists, but the three-dimensional display apparatus according to the present invention creates a three-dimensional image. A space of interest region is set in advance, and image reconstruction is performed only within that space region of interest. In the conventional X-ray CT apparatus, the number of pixels on the image plane to be reconstructed and the size thereof are determined in advance, and image reconstruction is performed for the center point of the pixel. In the original display device, a spatial distribution and a spatial position of a plurality of points for obtaining a CT value from the aspect of the image quality of the three-dimensional image to be created are planned in advance, and image reconstruction is performed on the points using projection data. This eliminates the need for spatial image reconstruction that is not related to the creation of 3D images. Also, in the spatial area where 3D images are created, the points are distributed densely in the spatial areas that have a large effect on the image quality of 3D images, and sparsely distributed in the spatial areas that have a small effect on the image quality of 3D images. It is possible to greatly reduce the amount of calculation for image reconstruction by using or not distributing points.
[0026]
In this way, the computational complexity of image reconstruction processing can be greatly reduced, so every time a 3D image is created, the image is reconstructed from the projection data, and the result is used to convert the 3D image. It became possible to create.
[0027]
The ray-tracing process used in volume rendering uses a ray-tracing process for the conical range radiated from the viewpoint to the target area of interest within the space area occupied by the object. If a CT value on a ray to be given is given, volume rendering can be performed. Accordingly, the ray used in the volume rendering ray tracing process is determined in advance, and the image reconstruction from the projection data may be performed only for the points on the ray used in this ray tracing. In the ray tracing process, light is attenuated by the opacity of the object, so that the spatial region contributing to the ray tracing process in the ray depth direction is limited. Further, the opacity can be set to 0 on the front side of the region of interest of the ray and can be ignored. Therefore, the image reconstruction from the projection data need only be performed for the spatial region contributing to the ray tracing process. That is, in the present invention, image reconstruction processing is performed on a point set on a ray of a ray that intersects the space of interest of the object using a set of projection data that passes through the vicinity of the point. Volume rendering is performed using CT values on the ray obtained as a result of the processing.
[0028]
Since the region of interest in 3D image display is usually quite small compared to the spatial region from which X-ray CT projection data is acquired, it is possible to significantly reduce the amount of calculation by using the method of the present invention. Therefore, it is possible to perform an image reconstruction process optimized from the projection data for each three-dimensional image display process.
[0029]
In image reconstruction in an X-ray CT apparatus, it is possible to set a reconstruction area smaller than the area from which projection data was acquired, and to perform image reconstruction processing on the reconstruction area to obtain image data having a small pixel size. Has been done. In the present invention, an area to be subjected to ray rendering processing of volume rendering, a ray direction and density, and a density of points giving CT values on the ray are given from the standpoint of volume rendering, and image reconstruction processing of projection data is performed using this data. A feature is to determine a spatial distribution and a spatial position of a plurality of points for obtaining a region of interest and a CT value to be performed, and to perform image reconstruction processing using projection data passing through this point and passing through this point. There is.
[0030]
In the conventional method, interpolation processing is performed a plurality of times in image reconstruction processing of projection data and 3D image reconstruction processing.
(1) When image reconstruction processing is performed from helical scan projection data, interpolation processing is performed on the helical scan projection data in order to obtain projection data passing through the cross section to be reconstructed.
(2) In the image reconstruction process, an interpolation process is performed on the projection data obtained in (1) in order to obtain projection data passing through pixels on the image plane.
(3) In order to construct voxel data used for 3D image processing, the reconstructed cross-sectional image data is stacked in the body axis direction, but the front and rear cross-sectional image data are used to obtain the desired axial interval. Interpolation may be performed.
(4) Interpolation processing is performed on voxel data in order to obtain a CT value on a ray by ray casting processing of volume rendering.
[0031]
In the method of the present invention, a point where a CT value is necessary is determined in advance by the ray rendering processing of volume rendering, and an image reconstruction is performed by performing an interpolation process of projection data passing through the neighborhood to this point, thereby obtaining an image. Volume rendering ray casting processing is performed using the obtained CT values. Therefore, the data interpolation process is only one of the projection data interpolation processes that pass through the vicinity in the image reconstruction process, and the number of interpolation processes can be greatly reduced. Compared with the image creation method, the image quality can be greatly improved.
[0032]
The method of the present invention can greatly reduce the amount of calculation of data processing.
(1) Since it is possible to determine in advance the spatial area and points for image reconstruction, the amount of calculation for image reconstruction processing can be greatly reduced.
(2) Since the number of data interpolation processes can be reduced, the amount of calculation in the image reconstruction process and the 3D image reconstruction process can be greatly reduced. In addition, by reducing the number of data interpolation processes, the image quality could be greatly improved.
(3) Since the data required for the ray rendering process of volume rendering is prepared in advance by the image reconstruction process, the calculation amount of the 3D image reconstruction process can be greatly reduced.
[0033]
by this,
(1) Since the calculation amount of data processing is greatly reduced, it has become possible to directly create a three-dimensional image from projection data that could not be routinely realized until now.
(2) Since it is possible to presuppose creation of a three-dimensional image, it is possible to use an image reconstruction parameter optimal for a three-dimensional image as an image reconstruction parameter of projection data.
(3) Since the amount of calculation for data processing has been reduced, it has become possible to reconstruct images at finer intervals than before, and the quality of 3D images could be further improved.
[0034]
In the present invention, a spatial area of volume data used in volume rendering and a spatial coordinate of a point in the spatial area used by volume rendering for calculation are obtained in advance, and each point is used in image reconstruction processing using projection data. The CT value is obtained by performing image reconstruction processing on the spatial coordinate position. Next, in volume rendering processing for creating a three-dimensional image, volume rendering processing is performed using the CT value obtained by image reconstruction processing for the spatial coordinate position of each point. From the standpoint of creating a 3D image, the volume area used for volume rendering is a part of the entire volume data, so it is possible to greatly reduce the space area for reconstructing projection data. Thus, the computational resources required for image reconstruction can be greatly reduced. Accordingly, the calculation time required for image reconstruction can be greatly shortened. In addition, since the CT values of points necessary for the volume rendering calculation are obtained in advance by image reconstruction processing, interpolating processing between voxels that is necessary in conventional volume rendering becomes unnecessary. As a result, the calculation time for 3D image reconstruction can be reduced, accuracy deterioration associated with interpolation can be prevented, and image quality can be improved.
[0035]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, a system of a three-dimensional image display apparatus that directly creates a three-dimensional image from projection data of the X-ray CT apparatus according to the present invention will be described. FIG. 2 is a block diagram schematically showing a three-dimensional image display apparatus according to the present invention connected to an X-ray CT apparatus via a network. Reference numeral 101 denotes an X-ray CT apparatus, 111 denotes a scanner unit of the X-ray CT apparatus, and 112 denotes a data collection unit. The projection data collected by detecting the X-ray scan of the subject in the scanner unit by the X-ray detector is digitized by the data collection unit. The digitized output data 122 of the X-ray detector is sent to a preprocessing unit 113 where preprocessing such as data noise removal and normalization is performed. The projection data 123 preprocessed by the preprocessing unit 113 is sent to the image reconstruction device 114 and sent to the projection data storage device 117 for storage. The image reconstruction device 114 performs an image reconstruction process on the preprocessed projection data 123. The reconstructed image data 124 is sent to and displayed on the console 115 of the X-ray CT apparatus and is stored in the image data storage device 116. Reference numeral 321 denotes projection data sent from the projection data storage device 117 to the three-dimensional image display device. In some cases, the X-ray CT apparatus does not have the projection data storage device 117. In this case, the projection data 123 preprocessed by the preprocessing unit 113 is directly sent to the three-dimensional image display device and stored.
[0036]
Reference numeral 301 denotes a three-dimensional image display device. A projection data storage apparatus 311 stores projection data 321 sent from the X-ray CT apparatus. 312 has a function of setting an image reconstruction area, and sets an image reconstruction space area based on a space area of interest for creating a three-dimensional image. This is physically included in the image reconstruction processing device 313. The image reconstruction processing device 313 obtains the CT value of the designated point in the designated image reconstruction space region by image reconstruction processing. Reference numeral 324 denotes a data set of CT values obtained by image reconstruction, which is sent to the three-dimensional image processing apparatus 314. The three-dimensional image processing device 212 receives the CT value data set 324 reconstructed from the image reconstruction processing device 313, performs three-dimensional image reconstruction such as volume rendering on this, and creates the generated three-dimensional image 325. Is displayed on the console 315.
[0037]
FIG. 3 is a diagram schematically showing processing and data flow in the three-dimensional image display apparatus according to the present invention. The X-ray 121 transmitted through the subject by the X-ray scanning of the scanner unit 111 of the X-ray CT apparatus is detected by the X-ray detector and digitized by the data acquisition unit 112. The digitized projection data 122 is sent to the preprocessing unit 113, where processing such as data noise removal and calibration is performed. The projection data 123 preprocessed by the preprocessing unit 113 is sent to the image reconstruction device 114 and sent to the projection data storage device 117 for storage. The image reconstruction device 114 performs an image reconstruction process on the preprocessed projection data 123. The reconstructed image data 124 is sent to and displayed on the console 115 of the X-ray CT apparatus and is stored in the image data storage device 116. Reference numeral 321 denotes projection data sent from the projection data storage device 117 to the three-dimensional image display device.
[0038]
The projection data storage device 311 of the three-dimensional image display device stores the projection data 321 sent from the X-ray CT device, and the projection data 322 of the examination specified by the operator is a reconstruction area setting unit 312 of the image reconstruction processing device. To send to. The reconstruction area setting unit 312 sets image reconstruction parameters and prepares necessary projection data 323 based on the spatial area for image reconstruction and the position of the points obtained from the spatial area for creating a 3D image. In the image reconstruction process 313, the CT value of the designated point in the designated image reconstruction space area is obtained by the image reconstruction process. Three-dimensional image processing 314 is performed on the data set 324 of CT values that have been reconstructed. The 3D image processing 314 performs 3D image reconstruction such as volume rendering on this, and displays the created 3D image 325 on the console 315.
[0039]
When the operator selects an examination for creating a three-dimensional image, an instruction 326 for designating necessary projection data is transmitted from the console to the projection data storage device. Based on this, the projection data storage device transmits the projection data 322 of the inspection designated by the operator to the reconstruction area setting unit 312 of the image reconstruction processing device. When the operator determines the space area of interest of the three-dimensional image to be created using the projection data, and determines the spatial distribution and the spatial position of the point for obtaining the CT value, an instruction 327 is transmitted from the console to the reconstruction area setting unit. Is done. Based on this instruction, the reconstruction area setting unit 312 sets image reconstruction parameters and prepares necessary projection data 323. In the image reconstruction process 313, the CT value of the designated point in the designated image reconstruction area is obtained by the image reconstruction process. Three-dimensional image processing 314 is performed using the CT value data set 324 obtained by image reconstruction. The 3D image processing 314 performs 3D image reconstruction such as volume rendering on the CT value data set, and displays the created 3D image 325 on the console 315.
[0040]
The operator observes the 3D image, changes the area and parameters for creating the 3D image as necessary, and transmits an instruction 327 from the console. Image reconstruction and 3D image reconstruction are performed again using the changed image reconstruction parameters. Since the image reconstruction time is greatly reduced by the present invention, this series of scanning and processing can be performed interactively.
[0041]
FIG. 4 is a diagram for schematically explaining the relationship between the spatial regions of the present invention. Reference numerals 511, 512, and 513 denote the XY plane, the XZ plane, and the YZ plane, respectively, of the spatial region where the projection data of the subject is collected. Reference numerals 511, 512, and 513 denote the XY plane, the XZ plane, and the YZ plane, respectively, of the spatial region where the projection data of the subject is collected. Reference numerals 514, 515, and 516 respectively denote an XY plane, an XZ plane, and a YZ plane of a space of interest for creating a three-dimensional image. Reference numerals 521, 522, and 523 denote an XY plane, an XZ plane, and a YZ plane, respectively, at the viewpoint positions for creating a three-dimensional image. Reference numerals 531, 532, and 533 denote the XY plane, the XZ plane, and the YZ plane of the ray space area of the ray tracing process for creating a three-dimensional image, respectively.
[0042]
Projection data in the spatial region indicated by 511, 512, and 513 is acquired by scanning with an X-ray CT apparatus. This projection data is transferred to the three-dimensional image display device according to the present invention. The operator determines a spatial region for creating a three-dimensional image, that is, a spatial region of interest 514, 515, 516. Then, the viewpoint positions 521, 522, and 523 are determined. Reference numerals 531, 532, and 533 denote the XY plane, the XZ plane, and the YZ plane of the conical space area including the line of sight when the region of interest is viewed from the viewpoint. Rays in the ray tracing process for creating a three-dimensional image are included in this spatial region. In consideration of the image quality of the three-dimensional image, the density of rays spreading radially from the viewpoint in this conical region and the position of the point for obtaining the CT value on each ray line are determined. That is, the position of the point for obtaining the CT value is determined on the ray line used for the ray tracing process.
[0043]
The reconstruction area setting unit is instructed to determine the position of the point for which the CT value thus determined is obtained. The image reconstruction apparatus obtains the CT value of the point for obtaining the designated CT value by using the projection data passing through the point for obtaining the designated CT value and passing through the vicinity of the point. The three-dimensional image processing apparatus performs three-dimensional image processing using a CT value data set at a point at which an instructed CT value is obtained. Since the CT value of the point on the ray line used for the ray tracing process of the three-dimensional image processing is obtained, the interpolation process is unnecessary.
[0044]
【Example】
In the above description, an example in which projection data is transferred from an X-ray CT apparatus to a 3D image display apparatus via a network has been described. However, projection data is transferred from an X-ray CT apparatus to a 3D image apparatus via an offline storage medium. The same applies when transferring.
[0045]
In the description so far, the X-ray CT apparatus and the three-dimensional image display apparatus have been described as separate apparatuses. However, when the X-ray CT apparatus includes the function of the three-dimensional image display apparatus of the present invention, the same applies. included.
[0046]
In the above description, the perspective projection method of the volume rendering process has been described, but various volume rendering processes such as a parallel projection method are similarly included.
[0047]
In the above description, volume rendering processing has been described as three-dimensional image processing. However, the present invention can be used for various three-dimensional image processing such as MIP and Latham. It can also be used for various 2D image processing such as 2D cross-section display and 2D curved image display.
[0048]
In the above description, the X-ray CT apparatus has been described. However, the present invention is not limited to the projection data of the X-ray CT apparatus, but also the so-called raw data of the nuclear medicine apparatus such as SPECT, the so-called raw data of the MR apparatus, and the line of the ultrasonic apparatus. A system for storing data and the like and performing image reconstruction by applying new parameters as necessary is also included.
[0049]
【The invention's effect】
In conventional 3D image creation / display, the X-ray CT apparatus reconstructs projection data and creates cross-sectional data. Next, three-dimensional voxel data is created by stacking slice data having different cross-sectional positions. A 3D image reconstruction process such as a volume rendering process is performed on the 3D voxel data to create and display a 3D image. On the other hand, in the 3D image display apparatus according to the present invention, a mechanism for setting a space area of interest to be displayed as a 3D image, that is, a space area of interest for performing 3D image reconstruction, is used for 3D image reconstruction. A mechanism for setting in advance a spatial distribution and a spatial position of points for obtaining physical quantities representing the characteristics of the object, that is, CT values, is provided. Reconstruction processing is performed on the designated spatial position in the designated space of interest using projection data, and a physical quantity representing the characteristics of the subject, that is, a CT value is obtained. A three-dimensional image reconstruction process is performed on the CT value data set thus obtained, and the obtained result is displayed as a three-dimensional image.
[0050]
In addition, a physical quantity representing the characteristics of the object, that is, a spatial region where a CT value data set exists and a spatial distribution of points where CT values exist are obtained by performing a three-dimensional image reconstruction process using this data set. By taking into account the effect on the image quality of the obtained 3D image, a spatial region of interest in which the CT value is obtained by performing image reconstruction of the projection data, and a spatial distribution of a plurality of points in which the CT value in the region of interest is obtained A mechanism is provided that makes it possible to determine the spatial position.
[0051]
In the conventional X-ray CT apparatus, image reconstruction using the projection data is performed on all areas where the projection data exists, but the three-dimensional display apparatus according to the present invention creates a three-dimensional image. An area is set in advance, and image reconstruction is performed only within that area. In the conventional X-ray CT apparatus, the number of pixels on the image plane to be reconstructed and the size thereof are determined in advance, and image reconstruction is performed for the center point of the pixel. In the original display device, the positions and distributions of points having CT values are planned in advance from the image quality of the three-dimensional image to be created, and image reconstruction is performed on the points using projection data. This eliminates the need for image reconstruction of spatial areas that are not related to the creation of 3D images. Also, in the spatial region where 3D images are created, it is important to distribute points densely in areas that have a large effect on the image quality of 3D images, and sparsely distribute areas in areas that have a small effect on image quality of 3D images. By not distributing it, the computational complexity of image reconstruction can be greatly reduced. Since the amount of image reconstruction calculations can be greatly reduced, every time a 3D image is created, it is possible to reconstruct the image from projection data and use the result to create a 3D image. Became.
[0052]
The ray-tracing process used in volume rendering uses a ray-tracing process for the conical range radiated from the viewpoint to the region of interest of the object within the space area occupied by the object. If a CT value on a ray to be given is given, volume rendering can be performed. Accordingly, the ray used in the volume rendering ray tracing process is determined in advance, and the image reconstruction from the projection data may be performed only for the points on the ray used in this ray tracing. In the ray tracing process, light is attenuated by the opacity of the object, so that the spatial region contributing to the ray tracing process in the ray depth direction is limited. Further, the opacity may be set to 0 on the near side of the region of interest of the ray and ignored. Therefore, the image reconstruction from the projection data need only be performed for the spatial region contributing to the ray tracing process. In other words, image reconstruction processing is performed on a point set on a ray that intersects the region of interest of the object, using a set of projection data passing through the vicinity of the point, and as a result of the image reconstruction processing. Volume rendering is performed using the CT value on the obtained ray. Since the region of interest in 3D image display is usually considerably smaller than the spatial region from which X-ray CT projection data is acquired, using the method of the present invention, each time projection data is processed for 3D image display processing. It is also possible to realize image reconstruction.
[0053]
In image reconstruction that can be placed on an X-ray CT apparatus, a region smaller than the region from which projection data is acquired is set, and image reconstruction is performed on that region. In the present invention, a volume rendering ray tracing process, a ray direction and density, and a density for giving a CT value on the ray are given in advance from the standpoint of volume rendering. It is characterized in that the position of a point to be obtained is determined and image reconstruction processing is performed using projection data passing through the vicinity of this point. In the second embodiment of the present invention, an area for performing ray tracing processing of volume rendering, a ray direction and density, and a density for giving a CT value on the ray are given from the standpoint of volume rendering, and space for image reconstruction from now on. It is characterized in that a region is determined, a voxel space is determined in the region, and image reconstruction processing is performed using projection data passing through the neighborhood for this voxel.
[0054]
In the method of the present invention, a point that requires a CT value in the volume rendering ray casting process is determined in advance, and an interpolation process is performed on projection data passing through the vicinity of this point to perform image reconstruction, thereby obtaining a CT value. The volume rendering ray casting process is performed by using the obtained CT value data set. Therefore, the data interpolation process is performed only once, and the number of interpolation processes can be greatly reduced. Accordingly, the image quality can be greatly improved.
[0055]
The method of the present invention can greatly reduce the amount of calculation of data processing. (1) Since the space area and the point to be reconstructed are determined in advance, the amount of calculation of the image reconstruction process can be greatly reduced. (2) Since the number of data interpolation processes can be reduced, the amount of calculation in the image reconstruction process and the 3D image reconstruction process can be greatly reduced. In addition, the image quality could be improved by reducing the number of data interpolation processes. (3) Since the CT value data of the position required for the ray rendering process of volume rendering is prepared in advance by the image reconstruction process, the calculation amount of the 3D image reconstruction process can be greatly reduced.
[0056]
As a result, (1) the amount of calculation for data processing has been reduced, and it has become possible to directly create a three-dimensional image from projection data that could not be realized routinely until now. (2) Since it is possible to assume the creation of a 3D image, it is possible to use the optimum parameters for the 3D image as the image reconstruction parameters of the projection data. (3) Since the amount of calculation for data processing has been reduced, it has become possible to reconstruct images at finer intervals than before, and the image quality of three-dimensional images has been improved.
[0057]
In the present invention, the spatial coordinates of the data used in the volume rendering process and the spatial coordinates of the points in the spatial area used by the volume rendering are calculated in advance, and the respective points are used in the image reconstruction process using the projection data. An image reconstruction process is performed on the spatial coordinate position of the point to obtain the CT value of that point. Next, in volume rendering processing for creating a three-dimensional image, volume rendering processing is performed using the CT value obtained by image reconstruction processing for the spatial coordinate position of each point. Since the spatial area of the data used for volume rendering is a part of the spatial area of the entire projection data, the spatial area for reconstructing the projection data can be greatly reduced, which is necessary for image reconstruction. Computational resources can be greatly reduced. Accordingly, the calculation time required for image reconstruction can be greatly reduced. In addition, since the CT value of the point necessary for calculation of volume rendering processing is obtained in advance by image reconstruction processing, interpolation processing between voxels, which was necessary in conventional volume rendering, becomes unnecessary, reducing calculation time, It has become possible to prevent the deterioration of accuracy associated with interpolation and greatly improve the image quality.
[Brief description of the drawings]
FIG. 1 is a block diagram for explaining a conventional X-ray CT apparatus and a three-dimensional image display apparatus.
FIG. 2 is a block diagram illustrating an X-ray CT apparatus and a three-dimensional image display apparatus according to the present invention.
FIG. 3 is a diagram schematically showing processing and data flow in a three-dimensional image display device that directly creates a three-dimensional image from projection data of an X-ray CT apparatus according to the present invention.
FIG. 4 is a diagram for schematically explaining the relationship between the spatial regions of the present invention.
[Explanation of symbols]
101 X-ray CT system
111 Scanner section of X-ray CT system
112 Data collection unit
113 Preprocessing section
114 Image reconstruction device
115 console
116 Image data storage device
117 Projection data storage device
121 X-ray scanning in X-ray CT system
122 Projection data collected by the scanner unit of the X-ray CT apparatus
123 Projection data preprocessed by the preprocessing unit
124 Reconstructed image data
201 Conventional three-dimensional image display device
211 Image data storage device
212 3D image processing device
213 console
221 Image data sent to 3D image display device
222 Image data for 3D image processing
223 3D image data
301 Three-dimensional image display device according to the present invention
311 Projection data storage device
312 Reconfiguration area setting unit
313 Image reconstruction processing device
314 Three-dimensional image processing device
315 console
321 Projection data sent to 3D image display device
322 Projection data for image reconstruction
323 Projection data required for image reconstruction of an image reconstruction space area set based on a space area of interest for creating a three-dimensional image
324 Data set of CT values of points included in the space of interest for creating a 3D image
325 3D image data
326 Instruct to select projection data
327 Instruction of image reconstruction area based on area to create 3D image
511 XY plane of the spatial region where the projection data of the subject is collected
512 XZ plane of spatial region where projection data of subject is collected
513 YZ plane of the spatial region where the projection data of the subject is collected
514 XY plane of region of interest for creating 3D image
515 XZ plane of region of interest for creating 3D image
516 YZ plane of region of interest for creating 3D image
521 Position of XY plane of viewpoint for creating 3D image
522 Position of XZ plane of viewpoint for creating 3D image
523 YZ plane position of viewpoint for creating 3D image
531 XY plane of spatial area including ray of 3D image processing
532 XZ plane of spatial region including ray of 3D image processing
533 YZ plane of spatial area including ray of 3D image processing

Claims (1)

A mechanism for collecting projection data by X-ray scanning, a mechanism for performing image reconstruction processing on this projection data, and performing image processing on the image data obtained by this image reconstruction processing and displaying the results mechanism and an X-ray CT apparatus and a mechanism for storing image data obtained by the image reconstruction, a mechanism for storing projection data collected by X-ray scanning, spatial region to collect projection data A mechanism for setting a space area of interest for reconstructing and displaying a 3D image within the space, and a plurality of CT values that are physical quantities representing the characteristics of the subject used for 3D image reconstruction within the space of interest area The density determined based on the image quality of the created 3D image from the spatial position of the point that observes the 3D image toward the internal region of the space area of interest that creates the 3D image By setting a plurality of radial lines of sight (rays), for each of the plurality of lines of sight (rays) set, at intervals and ranges determined based on the image quality of the three-dimensional image to be created, Performs image reconstruction processing using projection data for each of the designated spatial positions and the mechanism for designating the spatial positions of a plurality of points for which a CT value is to be obtained. A slice obtained by the image reconstruction process in the X-ray CT apparatus, and a mechanism for reconstructing and displaying a three-dimensional image using the CT value data set obtained by the image reconstruction process Alternatively, a three-dimensional image display device that can directly create and display a three-dimensional image from projection data collected by X-ray scanning without using volume image data.

Images