JP2010046102A - Tomographic image processor, x-ray ct apparatus, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To image an area including the outside of an imaging visual field of an X-ray CT apparatus as a tomographic image wherein the deviation of a pixel value and the deviation of the contour of a subject are suppressed. <P>SOLUTION: The area corresponding to the outside of the imaging visual field in a reconfiguration image is reconfigured by inversely projecting only actually measured projection data, and the pixel value of a pixel in the area is normalized by the number of view of the projection data inversely projected through the pixel position. Thus, the area to be reconfigured in terms of software is widened to the outside of the imaging visual field without being accompanied by the change in terms of hardware of the X-ray CT apparatus, and an imageable area in X-ray CT imaging is more easily widened. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a tomographic image processing apparatus for processing a tomographic image obtained by X-ray CT (Computed Tomography) imaging of a subject, an X-ray CT apparatus, and a program therefor.

  Conventionally, in X-ray CT imaging, from various viewpoints, there is a demand for securing a wide imageable region without increasing the number of X-ray detection elements constituting the X-ray detector.

In order to meet such a demand, for example, an X-ray CT apparatus using an asymmetric detector has been proposed (see, for example, Patent Document 1 and FIG. 4). In the asymmetric detector, the X-ray detection element is arranged so as to be asymmetric with respect to the channel direction about an extension line passing through the focal point of the X-ray tube and the rotation center of the gantry. It is a detector. In an X-ray CT apparatus using an asymmetric detector, the gantry is rotated 360 degrees and a full scan is performed, so that an imageable region can be extended to the outermost channel of the asymmetric detector. It can be expanded to the same area as when a symmetrical X-ray detector is used.
JP 2003-135443 A

  However, since an X-ray CT apparatus using an asymmetric detector has a special hardware design and is difficult to handle, there is a demand for a method for making an imageable area easier and wider.

  In view of the above circumstances, an object of the present invention is to provide a tomographic image processing apparatus, an X-ray CT apparatus, and a program therefor that can more easily widen an imageable area in X-ray CT imaging. To do.

  In a first aspect, the present invention back-projects projection data (data) in a plurality of view directions corresponding to the view angles required for reconstruction, thereby obtaining an imaging field of view and imaging of the X-ray CT apparatus. A tomographic image reconstruction means for reconstructing a tomographic image corresponding to a space including the outside of the visual field, and pixel values of pixels included in the tomographic image outside the imaging visual field obtained by the tomographic image reconstruction means, There is provided a tomographic image processing apparatus comprising: a pixel value normalizing unit that normalizes to a pixel value corresponding to a pixel value of the tomographic image of the imaging field obtained by the reconstruction unit.

  In a second aspect, the present invention provides the pixel value normalization unit, wherein projection data is substantially backprojected or backprojected among the plurality of view directions for each pixel of the tomographic image outside the imaging field of view. A tomographic image processing apparatus according to the first aspect of the present invention that normalizes pixel values based on the number of view directions that are not performed and the number of view directions.

  In a third aspect, the present invention provides the tomographic image processing apparatus according to the second aspect, wherein the pixel value is a CT value, and the pixel value normalizing means normalizes the CT value according to the following equation.

g ′ (x, y) = {g (x, y) +1000} × N / (N−M) −1000
However, g (x, y) is a CT value before normalization at a pixel at coordinates (x, y), and g ′ (x, y) is a CT value after normalization at a pixel at coordinates (x, y). , N is the number of the plurality of view directions, and M is the number of the view directions that are not substantially backprojected.

  In a fourth aspect, the present invention applies an extrapolation method for estimating and extrapolating data connected to an end of the measured projection data in the channel direction to the projection data in the plurality of view directions, Extrapolation method applied tomographic image for reconstructing an extrapolation method applied tomographic image corresponding to the space by obtaining extrapolated projection data of a plurality of views and backprojecting the extrapolated projection data of the plurality of views. Reconstructing means, and the image corresponding to outside the imaging field is a portion corresponding to the outside of the imaging field in the normalized tomographic image, and the image corresponding to the imaging field is the imaging in the extrapolation method application tomographic image There is provided a tomographic image processing apparatus according to any one of the first to third aspects, further comprising synthetic tomographic image generation means for generating a synthetic tomographic image that is a portion corresponding to a visual field.

  In a fifth aspect, the present invention provides the method according to the first to fourth aspects, wherein the tomographic image reconstruction means applies a filter back projection method to the projection data in the plurality of view directions. A tomographic image processing apparatus according to any one of the aspects is provided.

  In a sixth aspect, the present invention relates to a bone part region searching means for searching a bone part region in the region based on a pixel value of a region corresponding to the outside of the imaging field in the tomographic image, and the searched bone Pixel value correction means for converting a pixel value of a pixel in the periphery of the partial region into a value obtained by multiplying the pixel value by a gain coefficient that changes in accordance with a distance from the bone region to the pixel. A tomographic image processing device according to any one of the first to fifth aspects is provided.

  In a seventh aspect, X-ray CT imaging of a subject is performed, and imaging means for collecting projection data in a plurality of view directions corresponding to the view angles required for reconstruction, and the projection data in the plurality of view directions are reversed. Included in the tomographic image reconstruction means for reconstructing a tomographic image corresponding to the space including the imaging field of view and outside of the imaging field of view by projection, and the tomographic image outside the imaging field of view obtained by the tomographic image reconstruction means There is provided an X-ray CT apparatus comprising: a pixel value normalizing unit that normalizes a pixel value of a pixel to a pixel value corresponding to a pixel value of the tomographic image of the imaging field obtained by the tomographic image reconstruction unit.

  In an eighth aspect, the present invention provides a computer that backprojects projection data in a plurality of view directions corresponding to a view angle required for reconstruction, so that an imaging field of the X-ray CT apparatus and an imaging field outside the imaging field are displayed. A tomographic image reconstructing means for reconstructing a tomographic image corresponding to a space including a pixel value of a pixel included in the tomographic image outside the imaging field obtained by the tomographic image reconstructing means. The program for functioning as a pixel value normalization means which normalizes to the pixel value equivalent of the tomographic image of the said imaging visual field obtained by 1 is provided.

  Here, the imaging field of view is a so-called SFOV (scan field-of-view), and the object can be projected in a plurality of view directions corresponding to 360 degrees, that is, one scan rotation in the slice plane direction. It is a circular area. The imaging field of view is geometrically determined by the position of the X-ray detector, the width in the channel direction, and the like. In addition, the area outside the imaging field is an area outside the imaging field and in which the subject cannot be projected in at least a part of the view direction. That is, in the region corresponding to the imaging field, all of the projection data in the plurality of view directions are substantially back-projected to obtain a tomographic image, but in the region corresponding to the outside of the imaging field, the plurality of view directions. Substantially only a part of the projection data is backprojected to obtain a tomographic image.

  Further, the fact that the projection data is substantially back-projected means that when the projection data is back-projected, attention is paid to one pixel and a path passing through the pixel exists. Therefore, in the area corresponding to the imaging field, all the projection data in the plurality of view directions are substantially backprojected, and in the area corresponding to the outside of the imaging field, only a part of the projection data in the plurality of view directions is used. Substantially backprojected.

  According to the present invention, the region corresponding to the outside of the imaging field of view in the reconstructed image is reconstructed by back projecting only the actually measured projection data, and the pixel value of the pixel in the region is reversed through the pixel position. Normalize by the number of views of projection data to be projected. As a result, the area reconstructed in software can be expanded outside the field of view of imaging without changing the hardware of the X-ray CT apparatus, and the area that can be imaged in X-ray CT imaging can be expanded. It can be widened more simply.

  Embodiments according to the present invention will be described below. Here, an X-ray CT apparatus is taken up as an embodiment of the present invention. This X-ray CT apparatus has a general X-ray detector that is not an asymmetric detector, and scans a subject that partially protrudes outside the imaging field of view, and part of the projection information of the subject is missing. The projection data of a plurality of views including the projected data is collected. Then, a tomographic image corresponding to a space including the imaging field and the outside of the imaging field is reconstructed and generated using the filter back projection method. At this time, the pixel value of the pixel corresponding to the region outside the imaging field is corrected by a predetermined normalization method. In addition, the vicinity of the missing portion of the projection information of the subject in the projection data is improperly processed by the filter function, and in the region corresponding to the vicinity of the boundary of the imaging field in the reconstructed image, the pixel value is extremely increased, resulting in an artifact. Which corrects this artifact.

  FIG. 1 is a block diagram showing an X-ray CT apparatus 100 according to an embodiment of the present invention. The X-ray CT apparatus 100 includes an operation console 1, an imaging table 8, and a scanning gantry 9.

  The operation console 1 includes an input device 2, a central processing unit 3, a control interface (interface) 4, a data collection buffer (buffer) 5, and a monitor 6. The input device 2 accepts operator instructions and information. The central processing unit 3 executes a scan control process, a tomographic image reconstruction process, a pixel value normalization process, and the like. The control interface 4 outputs a control signal and the like to the imaging table 8 and the scanning gantry 9 under the control of the central processing unit 3. The data collection buffer 5 collects data acquired by the scanning gantry 9. The monitor 6 displays an operation screen, a tomographic image, and the like.

  The imaging table 8 moves the placed subject in the body axis direction (hereinafter referred to as z direction) and conveys it to the cavity of the scanning gantry 9.

  The scanning gantry 9 has a cavity including an imaging space, and includes a rotating unit 7 that rotates around the cavity. The rotation unit 7 includes an X-ray controller 10, an X-ray tube 11, a collimator 12, an X-ray detector 13, a data collection unit 14, and a rotation controller 15. The X-ray controller 10 controls tube voltage and tube current of the X-ray tube 11, on / off of X-ray irradiation, and the like. The collimator 12 shapes the X-ray beam emitted from the X-ray tube 11. The X-ray detector 13 has a detector row in which a plurality of detection elements are arranged symmetrically in the channel direction about an extension line passing through the X-ray tube 11 and the iso-center ISO. Each detection element outputs a signal corresponding to the detected X-ray intensity. The data collection unit 14 performs A / D (analog-digital) conversion on the output signal of the X-ray detector 13 to collect projection data of the subject and sends it to the data collection buffer 5. The rotation controller 15 controls the rotation speed of the rotating unit 7, on / off of rotation, and the like. The X-ray tube 11 and the X-ray detector 13 are disposed to face each other with the cavity portion interposed therebetween.

  The central processing unit 3 and the scanning gantry 9 are examples of imaging means in the present invention. The central processing unit 3 includes a tomographic image reconstruction unit, a pixel value normalization unit, an extrapolation method applied tomographic image reconstruction unit, a synthetic tomographic image generation unit, a bone region search unit, and a pixel value correction unit. It is an example.

  FIG. 2 is a flowchart showing a flow of tomographic image generation processing in the X-ray CT apparatus 100 according to the present embodiment.

  In step S1, projection data collection processing is executed. That is, for example, an axial scan is performed on a predetermined part of the subject, and projection data in a plurality of view directions corresponding to view angles necessary for image reconstruction with respect to a predetermined slice is obtained as a projection data set P. Collect as. Here, the view angle required for image reconstruction is 360 degrees, that is, one scan rotation, and the plurality of views are 984 views. The projection data is data after so-called fan-parallel conversion, and is obtained when an object is irradiated with a parallel beam of X-rays in one view direction. Is a collection of projection data values corresponding to each channel.

  In step S2, a tomographic image reconstruction process is executed. That is, the image is reconstructed by the filter back projection method based on the projection data set P, and the tomographic image G1 corresponding to the space including the imaging field of view and the outside of the imaging field of view is generated. More specifically, the projection data is subjected to Fourier transform, filtering, and inverse Fourier transform, and backprojected 360 degrees for each view direction in the reconstruction space to obtain a tomographic image G1. Reconstruct the image.

  FIG. 6 is a diagram showing an example of a tomographic image generated in this way. 6 (1) and 6 (2) show the same tomographic image when a predetermined phantom is scanned as the subject H, and FIG. 6 (1) shows the window level ( FIG. 6B is a tomographic image when the window level WL and the window width WW are displayed with general values. FIG. 6B is a tomographic image when the window level WL is displayed with a smaller value. Among the tomographic images G1, in the imaging field image G1f corresponding to the imaging field, an artifact A1 having a high pixel value occurs in the vicinity of the circumference Ba corresponding to the boundary of the imaging field, but other than that of the subject H The image is accurately depicted. On the other hand, as shown in FIG. 6 (1), in the tomographic image G1, the image-out-field-of-view image G1fg corresponding to the outside of the field-of-view, at first glance, the subject information seems to be almost missing, but the portion with a low pixel value. When the window level WL is lowered so that the image can be seen, an image of the subject emerges as shown in FIG. As described above, in the imaging out-of-view image G1fg, the back-projected projection data is insufficient, so that the pixel value is reduced and some artifacts are generated, but only in the actual measurement portion of the projection data without estimation. Since the image is reconstructed based on the image, it can be seen that the image of the subject is depicted in a state that is very close to the actual image.

  In step S3, a pixel value normalization process outside the imaging field of view is executed. That is, the pixel value of the imaging out-of-view image G1fg of the tomographic image G1 is normalized by the number of projection data views that are not substantially backprojected for the pixel (hereinafter also referred to as the number of missing views) and the number of views. To generate the normalized imaging out-of-view image G1′fg.

  Here, details of the imaging out-of-view pixel value normalization process will be described.

  FIG. 3 is a flowchart illustrating an example of the imaging field-of-view pixel value normalization process. The imaging out-of-view pixel value normalization process includes, for example, a plurality of steps as shown in FIG.

  In step S31, an image outside the imaging visual field is specified. That is, the to-be-photographed image G1fg is specified as a processing target in the tomographic image G1.

  In step S32, a target pixel is set. That is, among the pixels constituting the identified imaging out-of-view image G1fg, one pixel that has not yet been normalized, which will be described later, is selected, and that pixel is the target pixel G1 (xt, yt) to be processed ).

  In step S33, the number of missing views is read out. That is, the missing view number M (xt, yt) corresponding to the target pixel G1 (xt, yt) is read with reference to a prepared missing view number map (map) M (x, y). The missing view number map M (x, y) is a map in which coordinates in the reconstruction space are associated with the number of missing views corresponding to the coordinates. The number of missing views is a view in which projection data is not back-projected with respect to one pixel when attention is paid to one pixel out of the total number of views N (here, N = 984) of projection data back-projected in the reconstruction space. It can be obtained from the geometric positional relationship of the X-ray tube 11 and the X-ray detector 13.

  FIG. 7A is a diagram illustrating an example of a missing view number map. FIG. 7B is a diagram showing changes in the number of missing views in the diameter direction of the imaging field in the map, in which the horizontal axis corresponds to the coordinate rx in the diameter direction and the vertical axis corresponds to the pixel of the coordinate rx. This is expressed as the number of missing views M (rx). The missing view number map M (x, y) represents, for example, the size of the number of missing views corresponding to each pixel by the shading of the pixel (increasing the brightness as the number of missing views increases), as shown in FIG. ), The pixel region Rf corresponding to the imaging field of view becomes true black, and the pixel region Rfg corresponding to the outside of the imaging field of view becomes whiter as it goes away from the iso-center ISO, that is, goes outward. Is obtained. The relationship between the coordinates (relative value) on the straight line L set in the diameter direction of the imaging field in the missing view number map M (x, y) and the number of missing views at the coordinates is shown in FIG. As shown, the coordinates of the pixel region Rf corresponding to the imaging field of view are zero, the pixel region Rfg corresponding to the outside of the imaging field of view increases as the distance from the isocenter ISO, and exceeds 400 views at the end.

  Note that when the number of views N changes, the number of missing views for each pixel also changes, so if there are multiple conditions that can be set as the number of views N, a plurality of types of missing view number maps are prepared according to the conditions. It is necessary to keep it.

  In step S34, the pixel value is normalized. Here, the pixel value of the target pixel G1 (xt, yt) is normalized according to the following equation.

g1 '(xt, yt) = {g1 (xt, yt) +1000} × N / {NM (xt, yt)}-1000 (Formula 1)
Where g1 '(xt, yt): normalized pixel value at the target pixel
g1 (xt, yt): Pixel value before normalization at the target pixel
N: Total number of views of projection data backprojected in reconstruction space
M (xt, yt): Number of missing views in the pixel of interest In other words, add 1000 to the pixel value and convert the pixel value from a CT value with water as a zero reference to a pixel value with air as a zero reference The pixel value is once linear with respect to the magnitude of the X-ray absorption rate. Thereafter, a value obtained by dividing the total number of views of projection data to be backprojected by the number of views of projection data to be actually backprojected is obtained as a weighting coefficient, and the converted pixel value is multiplied by the weighting coefficient. Then, 1000 is subtracted from the obtained pixel value to return the pixel value to the CT value. As a result, the pixel value dropped due to the lack of backprojected projection data is raised to the original level.

  In step S35, it is determined whether or not there is a pixel whose pixel value has not been normalized among the pixels constituting the imaging out-of-view image G1fg. If the determination is affirmative, the process returns to step S31, and normalization of the pixel value in the image outside the field of view is continued. On the other hand, if this determination is negative, the imaging out-of-view pixel value normalization process is terminated.

  FIG. 8 is a diagram illustrating an example of a normalized tomographic image obtained by such an imaging field-of-view pixel value normalization process. For example, as shown in FIG. 8, the normalized tomographic image G <b> 1 ′ is drawn in a state in which the image of the subject H is quite close to the actual image in the normalized imaging out-of-view image G <b> 1 ′ fg corresponding to the outside of the imaging field. At the same time, the pixel value dropped due to the lack of backprojected projection data is raised to the original level.

  In step S4, a bone periphery pixel value correction process is executed. That is, correction of a decrease in pixel value caused by a beam hardening effect or the like occurring around the bone portion in the normalized imaged out-of-view image G1′fg to generate a corrected imaged out-of-view image G1 ″ fg To do.

  Here, details of the bone periphery pixel value correction processing will be described.

  FIG. 4 is a flowchart illustrating an example of a bone periphery pixel value correction process. The bone peripheral pixel value correction process includes, for example, a plurality of steps as shown in FIG.

  In step S41, a bone region is searched. In other words, the bone region Bj (j = 1, 2,...) Is searched for in the normalized imaged out-of-view image G1′fg by determining the threshold value of the pixel value. For example, as shown in FIG. 9 (1), the bone region B1 is detected by the threshold value determination of the pixel value in the normalized imaging out-of-view image G1′fg.

  In step S42, it is determined whether a bone region has been detected. If the determination is affirmative, the process proceeds to step S43. If the determination is negative, the tomographic image generation process is terminated.

  In step S43, the target bone part region is set. That is, among the detected bone region Bj, a bone region that has not yet been corrected for pixel values to be described later is set as the target bone region Bt to be processed. For example, the bone part region B1 is set as the target bone part region Bt.

  In step S44, concentric circles sandwiching the target bone region are set. That is, two concentric circles Dt1 and Dt2 sandwiching the target bone part region Bt with the isocenter ISO as the center are set. For example, as shown in FIG. 9B, concentric circles D11 and D12 are set with the bone region B1 as the target bone region Bt.

  In step S45, the angular range of the target bone region is measured. That is, when the target bone region Bt is sandwiched between two straight lines having the isocenter ISO as one end, an angle range αt formed by the two straight lines is measured. For example, as shown in FIG. 9B, an angle range formed by two straight lines sandwiching the bone region B1 is α1.

  In step S46, the neighboring pixel value of the target bone region is corrected. That is, in the ring-shaped region Dt sandwiched between the concentric circles Dt1 and Dt2, each pixel value g1 ′ (xk, xk, included in the angle range of αt × Scale centering on a straight line passing through the center of the target bone region Bt. yk) is multiplied by a gain coefficient Gain for correction. At this time, the target bone part region Bt may or may not be included as a correction target. The gain coefficient Gain changes so as to decrease as the distance or angle difference from the target bone region Bt increases. For example, a value of 2 is taken when the distance or angle difference is zero, and changes so as to approach a value of 1 as they increase. The correction of the neighboring pixel value of the target bone region is expressed by the following equation.

g1 "(xk, yk) = {g1 '(xk, yk) +1000} × Gain-1000 (Formula 2)
Where g1 "(xk, yk): corrected pixel value
g1 '(xk, yk): Pixel value before correction
Gain: Gain coefficient For example, the scale is set to about 3, and as shown in FIG. Each pixel value g1 ′ (xk, yk) included in the angle range of × Scale is corrected by multiplying it by a gain coefficient Gain.

  In step S47, it is determined whether or not there is a bone region that has not been subjected to pixel value correction among the detected bone region Bj. If the determination is affirmative, the process returns to step S43, and the correction of the pixel values around the bone portion is continued. On the other hand, when the determination is negative, the bone periphery pixel value correction process is terminated.

  FIG. 10 is a diagram showing an example of a corrected tomographic image obtained by such a bone periphery pixel value correction process. In the corrected tomographic image G1 ″, for example, as shown in FIG. 10, the decrease in pixel value due to the beam hardening effect or the like is corrected around the bone region B1 of the corrected imaging out-of-view image G1 ″ fg.

  In step S5, extrapolation method applied tomographic image generation processing is executed. Specifically, an extrapolation method is applied to the projection data set P to generate an extrapolation method applied tomographic image G2. The extrapolation method-applied tomographic image G2 corresponds to the vicinity of the boundary of the imaging field of view since the image is reconstructed using the filter back projection method after artificially extrapolating the missing portion of the projection information of the subject in the projection data. There are almost no whiteout artifacts in the area.

  Here, the details of the extrapolation method applied tomographic image generation processing will be described.

  FIG. 5 is a flowchart illustrating an example of an extrapolation method tomographic image generation process. The extrapolation method applied tomographic image generation process includes, for example, a plurality of steps as shown in FIG.

  In step S51, attention projection data is set. That is, one projection data in which extrapolation of a later-described missing portion has not yet been performed among the projection data Pv (v = 1, 2,..., N) constituting the projection data set P is a processing target. It is set as attention projection data Pt.

  In step S52, the feature amount of the measured part of the target projection data is calculated. That is, gradients (tilts) SLlt, SLrt, signal strengths MGlt, MGrt, etc. at the left and right ends of the curve are calculated on the profile of the target projection data Pt. For example, as shown in FIG. 11, the gradient SLlt and the signal intensity MGlt at the left end of the curve are calculated on the profile of the projection data Pt based on the actual measurement part pm of the target projection data Pt.

In step S53, the missing portion of the target projection data Pt is estimated and extrapolated. That is, the missing portion of the projection data Pt of interest is estimated and inserted based on the gradients SLlt, SLrt, signal strengths MGlt, MGrt, and the like. For example, as shown in FIG. 11, based on the gradient SLlt and the signal strength MGlt, a circle Clt is defined in which the center position in the channel direction is Xlt and the radius is Rlt. Next, a cylindrical water substance formed by extending this circle Clt in the body axis direction (z direction) of the subject H is virtually generated, and this is fitted to the end of the profile of the target projection data Pt. (fitting) Then, a projection data portion em obtained by forward projection of this water substance with a parallel beam in the view direction corresponding to the target projection data Pt is extrapolated as a missing portion of the projection information of the subject H. For details of this method, refer to non-patent literature, “A novel reconstruction algorithm to extend the CT scan.
field-of-view ", J. Hsieh, et al, Medical Physics, Vol. 31, No. 9, September
See 2004, P.2385-2391, or Japanese Patent Application Laid-Open No. 2004-121853.

  In step S54, it is determined whether there is projection data in which the missing portion has not yet been extrapolated in the projection data Pv. If the determination is affirmative, the process returns to step S51, and extrapolation for the missing projection data is continued. On the other hand, if negative, the process proceeds to step S55.

  In step S55, an extrapolation method applied tomographic image is generated. That is, based on the extrapolated projection data set P ′ including the projection data with extrapolation of the missing part, the image is reconstructed by the filter back projection method, and corresponds to the space including the imaging field in the real space and the outside of the imaging field. An extrapolation method applied tomographic image G2 is generated.

  FIG. 12 is a diagram illustrating an example of an extrapolation method applied tomographic image obtained by such extrapolation method applied tomographic image generation processing. In the extrapolation method tomographic image G2, for example, as shown in FIG. 12, over-exposed artifacts hardly occur in a portion corresponding to the vicinity of the boundary of the imaging field of view.

  In step S6, an image composition process is executed. Specifically, as shown in FIG. 13, the extrapolated imaging field image G2f corresponding to the imaging field in the extrapolation method applied tomographic image G2 and the imaging field-corrected imaging field-of-view image G1 ″ fg are combined, A synthetic tomographic image G3 is generated in which the image corresponding to the imaging field is the extrapolated imaging field image G2f and the image corresponding to the outside of the imaging field is the imaging field corrected G1 ″ fg. This generates a tomographic image that suppresses pixel value shifts and subject contour shifts in areas corresponding to areas outside the imaging field of view, and suppresses whiteout artifacts in areas corresponding to the vicinity of the boundary of the imaging field of view. Is done.

  As described above, according to the present embodiment, the region corresponding to the outside of the imaging field in the reconstructed image is reconstructed by back projecting only the actually measured projection data, and the pixel value of the pixel in the region is determined as the pixel. Normalize by the number of views of projection data backprojected through position. As a result, the area reconstructed in software can be expanded outside the field of view of imaging without changing the hardware of the X-ray CT apparatus, and the area that can be imaged in X-ray CT imaging can be simplified. Can be wide.

  By the way, when the subject protrudes out of the imaging field, the projection information (data) in a part of the view direction obtained by scanning causes the projection information of the subject to be interrupted at the end in the channel direction, resulting in truncation. Arise. Based on a projection data set composed of projection data in a plurality of view directions including the projection data in which this omission occurs, an image is reconstructed using, for example, a filter back projection method. Then, in the region corresponding to the outside of the imaging field of view in the reconstructed image, the projection data is not back-projected because there is no path for at least a part of all the views, and the projection data is insufficient and the artifact (artifact ) And CT value shift. Further, the vicinity of the missing portion of the projection data is improperly processed by the filter function, and in the region corresponding to the vicinity of the boundary of the imaging field in the reconstructed image, the pixel value is extremely increased to form an artifact.

  As a method capable of reducing such artifacts and CT value shifts, a method of correcting projection data called an extrapolation method has been proposed. The extrapolation method is a method of estimating a missing portion that is interrupted based on an actually measured portion or the like of projection data, and extrapolating the estimated missing portion to be added to the actually measured portion. When the image is reconstructed based on the projection data corrected by such extrapolation method, in the region corresponding to the outside of the imaging field of view, the projection data portion of a part of the view direction that is missing at the time of back projection is temporarily supplemented, Artifacts and CT value shifts in the reconstructed image are reduced.

  However, since the projection data portion extrapolated by the extrapolation method is pseudo data based on estimation to the last, it may deviate greatly from the projection data portion obtained when the subject is actually projected. In such a case, for example, the pixel value in the reconstructed image may deviate more than the value that should actually be obtained. In addition, for example, the position of the contour of the subject in the reconstructed image becomes inaccurate, and the contour in the body axis direction of the subject may greatly fluctuate in a three-dimensional image composed of a plurality of continuous reconstructed images.

  According to this embodiment, an image that is not significantly different from the actual image is not reconstructed due to inappropriate estimation of the missing portion in the projection data as in the case of applying the extrapolation method, and the backprojection is performed. The pixel values that have fallen due to a lack of the number of projection data can be restored, and the area including the outside of the imaging field is imaged as a tomographic image in which the deviation of the pixel values and the deviation of the contour of the subject are suppressed. It becomes possible.

  In the present embodiment, since the pixel values around the bone portion in the normalized imaging field outside the visual field are raised, it is possible to correct the decrease in the pixel value caused by the beam hardening effect or the like that occurs around the bone portion. it can.

  Further, in the present embodiment, the composite tomographic image using the tomographic image generated by applying the extrapolation method is generated as the image corresponding to the imaging visual field in the tomographic image by the image synthesis processing, and thus the imaging visual field. This not only suppresses pixel value shifts and object contour shifts in the area corresponding to the outside, but also causes overexposure artifacts in areas corresponding to the vicinity of the boundary of the imaging field due to loss of projection data. Can be obtained. This facilitates interpretation by a doctor and improves diagnosis efficiency.

  In addition, this invention is not limited to said embodiment, A various change and addition are possible in the range which does not deviate from the meaning.

  For example, the extrapolation method is not limited to the method proposed in the non-patent document, and any method may be used as long as the method extrapolates data to a missing portion of projection data.

Also, for example, as multiple views necessary for reconstruction, so-called full scan (full scan)
The view may be not only a view corresponding to 360 degrees corresponding to (scan) but also a view corresponding to 180 + α degrees corresponding to so-called half scan.

  Further, for example, the backprojection method used for image reconstruction is not particularly limited to the filter backprojection method, and various backprojection methods or techniques applying them can be used.

  Further, for example, the scan performed when collecting projection data is not limited to an axial scan, but may be a helical scan or the like.

  Note that a tomographic image that does not have imaging means for X-ray CT imaging of a subject and reconstructs a tomographic image corresponding to a space including an imaging field of view and an outside of the imaging field of view by back projecting the projection data set. The reconstruction unit and the pixel value of the pixel in the region corresponding to the outside of the imaging field in the reconstructed tomographic image are normalized based on the number of views of the projection data back-projected by the reconstruction unit with respect to this pixel. A tomographic image processing apparatus such as a workstation provided with pixel value normalization means is also an embodiment of the present invention.

  A program for causing a computer to function as the tomographic image reconstruction means and the pixel value normalization means is also an embodiment of the present invention.

It is a block diagram which shows the X-ray CT apparatus by this embodiment. It is a flowchart which shows an example of the tomographic image generation process in the X-ray CT apparatus by this embodiment. It is a flowchart which shows an example of an imaging visual field pixel value normalization process. It is a flowchart which shows an example of a bone part periphery pixel value correction process. It is a flowchart which shows an example of an extrapolation method application tomographic image generation process. It is a figure which shows an example of the tomographic image obtained by a tomographic image reconstruction process. It is a figure showing an example of a missing view number map, and the change of the number of missing views on the said map. It is a figure which shows an example of the normalized tomographic image obtained by an imaging visual field pixel value normalization process. It is a figure for demonstrating an example of a bone part periphery pixel value correction process. It is a figure which shows an example of the corrected tomographic image obtained by a bone part periphery pixel value correction process. It is a figure for demonstrating an example of the extrapolation method. It is a figure which shows an example of the extrapolation method application tomographic image obtained by the extrapolation method application tomographic image generation process. It is a figure for demonstrating the outline | summary of an example of an image composition process.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 X-ray CT apparatus 1 Operation console 2 Input device 3 Central processing unit 4 Control interface 5 Data acquisition buffer 6 Monitor 7 Rotating part 8 Imaging table 9 Scanning gantry 10 X-ray controller 11 X-ray tube 12 Collimator 13 X-ray detector 14 Data Collection unit 15 Rotation controller

Claims (8)

A tomogram for reconstructing a tomographic image corresponding to a space including the imaging field of view and the outside of the imaging field of the X-ray CT apparatus by back projecting projection data in a plurality of view directions corresponding to the view angles required for reconstruction. Image reconstruction means;
Pixel value that normalizes the pixel value of the pixel included in the tomographic image outside the imaging field obtained by the tomographic image reconstruction unit to the pixel value of the tomographic image of the imaging field obtained by the tomographic image reconstruction unit A tomographic image processing apparatus comprising normalization means.   The pixel value normalization means includes, for each pixel of the tomographic image outside the imaging field of view, the number of view directions in which projection data is substantially or is not backprojected among the plurality of view directions, and the plurality of views. The tomographic image processing apparatus according to claim 1, wherein the pixel value is normalized based on the number of directions. The pixel value is a CT value;
The tomographic image processing apparatus according to claim 2, wherein the pixel value normalizing unit normalizes the CT value according to the following expression.
g ′ (x, y) = {g (x, y) +1000} × N / (N−M) −1000
However, g (x, y) is a CT value before normalization at a pixel at coordinates (x, y), and g ′ (x, y) is a CT value after normalization at a pixel at coordinates (x, y). , N is the number of the plurality of view directions, and M is the number of the view directions that are not substantially backprojected. The extrapolated projection data of the plurality of views is applied to the projection data of the plurality of view directions by applying an extrapolation method for estimating and extrapolating data connected to the end of the measured projection data in the channel direction. Obtaining extrapolation method tomographic image reconstruction means for reconstructing an extrapolation method application tomographic image corresponding to the space by backprojecting the extrapolated projection data of the plurality of views;
The image corresponding to the outside of the imaging field is a portion corresponding to the outside of the imaging field of view in the normalized tomographic image, and the image corresponding to the imaging field of view corresponds to the imaging field of view in the extrapolation method application tomographic image The tomographic image processing apparatus according to any one of claims 1 to 3, further comprising a synthetic tomographic image generation unit configured to generate a synthetic tomographic image.   5. The tomographic image processing apparatus according to claim 1, wherein the tomographic image reconstruction unit applies a filtered back projection method to the projection data in the plurality of view directions. Based on a pixel value of a region corresponding to the outside of the imaging field of view in the tomographic image, a bone region search means for searching for a bone region in the region;
Pixel value correction means for converting a pixel value of a pixel around the searched bone region into a value obtained by multiplying the pixel value by a gain coefficient that changes in accordance with a distance from the bone region to the pixel; The tomographic image processing apparatus according to any one of claims 1 to 5, further comprising: Imaging means for collecting projection data in a plurality of view directions corresponding to the view angles necessary for reconstruction by X-ray CT imaging of the object;
A tomographic image reconstruction means for reconstructing a tomographic image corresponding to a space including an imaging field of view and an outside of the imaging field of view by back projecting the projection data in the plurality of view directions;
Pixel value that normalizes the pixel value of the pixel included in the tomographic image outside the imaging field obtained by the tomographic image reconstruction unit to the pixel value of the tomographic image of the imaging field obtained by the tomographic image reconstruction unit An X-ray CT apparatus comprising normalizing means. Computer
A tomogram for reconstructing a tomographic image corresponding to a space including the imaging field of view and the outside of the imaging field of the X-ray CT apparatus by back projecting projection data in a plurality of view directions corresponding to the view angles required for reconstruction. Image reconstruction means;
Pixel value that normalizes the pixel value of the pixel included in the tomographic image outside the imaging field obtained by the tomographic image reconstruction unit to the pixel value of the tomographic image of the imaging field obtained by the tomographic image reconstruction unit A program for functioning as a normalization means.