JP6858391B2 - X-ray CT equipment, image correction method and image correction program - Google Patents

Description

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

The X-ray CT apparatus can obtain a three-dimensional image including the internal structure of the object by reconstructing an image of the object taken by X-ray from various directions. Conventionally, by utilizing the features of this X-ray CT apparatus, it has been used for observing minute internal defects such as voids and cracks in metal parts and resin parts, measuring the complicated internal shape of electronic parts, and analyzing the cause of failure. It was.

At present, with the progress of digital technology, attempts to use the X-ray CT apparatus as the core of the digital engineering system have begun. The digital engineering system is a technology that integrates a high-performance CAD / CAM system, a three-dimensional modeling system, and a three-dimensional measurement system to realize efficiency and high quality from development to manufacturing. By combining these technologies, it is possible to repeat trial production and design without making a mold, and to bring the product to the market in a short time and at low cost. Sharing accurate and complete production and technical data is expected to reduce development risk.

As a three-dimensional measurement system, a digitizer, an optical cutting method, and the like have been proposed. However, although these methods can accurately measure the surface shape, it is extremely difficult to measure the internal shape of the object to be measured. For these measurement methods, an ultrasonic diagnostic method capable of determining the presence or absence of an internal space has also been proposed, but it is difficult to accurately grasp the internal shape of this method as well. Therefore, the X-ray CT apparatus is expected to be the only three-dimensional measurement system capable of grasping the internal structure.

By the way, in a general X-ray CT apparatus, as shown in FIG. 10, X-rays are irradiated from a point light source type X-ray source to the measurement object O in a cone beam shape, and the X-rays are the measurement objects. Characteristic X-rays generated by passing through O reach the flat panel type detector 200. Therefore, the error of the projection image I _P that is large becomes apparent as will end away from the central portion of the detector 200 of the flat panel type. In recent years, in order to correct an error of such a projection image I _P, techniques using calibration tool having a grid pattern arranged balls have been proposed (e.g., see Patent Document 1).

U.S. Patent Application Publication No. 2013/0230150

In the technique described in Patent Document 1, the geometry between the position of the point light source, the size and position of the sphere of the calibrator, and the size and position of the projected image of the sphere projected on the two-dimensional plane. The error of the projected image is corrected by utilizing the above relationship. However, since such correction is not based on an actual X-ray CT image (three-dimensional image), there is a problem that an error still remains.

The present invention has been made in view of such circumstances, and in an X-ray CT apparatus including a point light source type X-ray source and a flat panel type detector, at the end of the flat panel type detector. The purpose is to correct the error of the projected image with high accuracy.

In order to achieve the above object, the present invention includes a point light source type X-ray source that irradiates an object to be measured with X-rays, and a flat panel type detector that detects characteristic X-rays that have passed through the object to be measured. , An X-ray CT device that generates an X-ray CT image of the object to be measured based on the characteristic X-rays detected by the flat panel detector, and converges the generated X-ray CT image synogram to an ideal synogram. Provided is an X-ray CT apparatus provided with an image correction means for correcting an X-ray CT image.

Further, the present invention includes a point light source type X-ray source that irradiates an object to be measured with X-rays and a flat panel type detector that detects characteristic X-rays that have passed through the object to be measured, and is a flat panel type detector. It is an image correction method of an X-ray CT apparatus that generates an X-ray CT image of an object to be measured based on the characteristic X-rays detected in, so that the synogram of the generated X-ray CT image is converged to an ideal synogram. Provided is an image correction method including an image correction step of correcting an X-ray CT image.

Furthermore, the present invention presents a computer with an X-ray CT image so that the synogram of the X-ray CT image of the object to be measured generated by using a point light source type X-ray source and a flat panel type detector is converged to an ideal synogram. Provided is an image correction program for executing an image correction step for correction. Further, a computer-readable recording medium on which the image correction program according to the present invention is recorded is also provided.

It is a block diagram of the X-ray CT apparatus which concerns on embodiment of this invention. It is explanatory drawing for demonstrating the synogram of an X-ray CT image. It is explanatory drawing for demonstrating the maximum likelihood estimation / expected value maximization reconstruction method. It is a figure which shows the comparison result of the cross-sectional image reconstructed by using the maximum likelihood estimation / expected value maximization reconstruction method, and the cross-sectional image reconstructed by using a filter correction back projection method. It is explanatory drawing for demonstrating the method of correcting an X-ray CT image using an ideal synogram. An example of a calibration phantom is shown, where (A) is a perspective view and (B) is a sectional view. It is a graph which shows the relationship between the position of a calibration phantom from the center of a sphere, and the error between the centers of a sphere. It is a flowchart for demonstrating the image correction method which concerns on embodiment of this invention. It is a perspective view which shows another example of a calibration phantom. It is explanatory drawing for demonstrating the error of the projected image at the end part of the flat panel type detector of the conventional X-ray CT image.

First, the configuration of the X-ray CT apparatus 1 according to the embodiment of the present invention will be described with reference to FIGS. 1 to 7. The X-ray CT apparatus 1 irradiates the measurement target O with X-rays and detects the projection data for each rotation angle of the measurement target O, thereby detecting the X-ray CT image (three-dimensional image) of the measurement target O. Is to get.

As shown in FIG. 1, the X-ray CT apparatus 1 includes an X-ray source 10 for irradiating X-rays, a detector 20 for detecting characteristic X-rays transmitted through an object O to be measured, an X-ray source 10 and a detector. An installation table 30 arranged between the 20 and for installing the measurement object O, a signal processing means 40 for quantifying the characteristic X-ray dose (characteristic X-ray peak) measured by the detector 20, and a signal processing means. The image reconstructing means 50 for reconstructing an image based on the numerical data according to the above, and the image correcting means 60 for correcting the X-ray CT image obtained by the image reconstructing means 50 are provided.

In this embodiment, a point light source type X-ray source 10 is adopted, and a flat panel type detector 20 is adopted. The installation table 30 is configured to rotate about a predetermined rotation axis by a moving mechanism (not shown) and to perform a linear motion along an axis orthogonal to the rotation axis. The installation base 30 is preferably made of granite or ductile cast iron having high rigidity.

The signal processing means 40 and the image reconstructing means 50 are composed of hardware such as a computer and software such as a program implemented therein. Specifically, when the programs for the signal processing means 40 and the image reconstructing means 50 are read into a computer via a communication medium such as the Internet or a recording medium such as USB, an arithmetic processing unit such as a CPU, a memory, or the like Various processes are executed by the storage unit and the like. Various data and result data necessary for such execution are appropriately input via the input unit and the communication unit, and are output via the output unit and the display unit.

The image reconstruction means 50 in the present embodiment is the maximum likelihood estimation / expected value maximization reconstruction method (hereinafter, “ML-EM reconstruction method”) in the sequential approximation reconstruction method, similarly to the image correction means 60 described later. The X-ray CT image of the object O to be measured is reconstructed based on the numerical data of the detected X-ray dose, but other algorithms (for example, filter-corrected back projection method, additive ART method, etc.) Images can also be reconstructed using the multiplying ART method, SIRT method, gradient method, steepest descent method, conjugate gradient method, MAP-EM method, Convex method, etc.).

A linear scale may be arranged between the X-ray source 10 and the detector 20. In this way, the position of the installation table 30 can be accurately grasped, and the X-ray CT image of the measurement object O can be accurately acquired.

The image correction means 60 corrects the X-ray CT image so that the sinogram (actual measurement synogram) of the X-ray CT image of the measurement object generated by using the X-ray source 10 and the detector 20 converges to the ideal synogram. It is a thing. As shown in FIG. 1, the image correction means 60 includes a display means 61 that displays the data of the X-ray CT image generated by the image reconstruction means 50 on the display screen as a synogram, and a synogram (actual measurement synogram) of the X-ray CT image. It has a correction means 62 for correcting an X-ray CT image by reconstructing an image using the ML-EM reconstruction method in the sequential approximation reconstruction method so as to converge the image to an ideal synogram. The display means 61 and the correction means 62 are composed of hardware such as a computer and software such as a program implemented therein, and the programs for the display means 61 and the correction means 62 are read into the computer. Then, various processes are executed by an arithmetic processing unit such as a CPU or a storage unit such as a memory.

Here, the synogram used for image correction will be described. The synogram is an image in which the detection signal for each angle when the measurement object O is rotated by 360 ° is expressed by a sine wave, and is acquired for each cross section of the measurement object O. The synogram of the X-ray CT image in the predetermined cross section of the measurement object O having an elliptical plan view generated by the image reconstructing means 50 is represented by an image as shown in FIG. 2, for example.

Further, the ML-EM reconstruction method used for image correction will be described with reference to FIGS. 3 and 4. The ML-EM reconstruction method is a method of repeatedly calculating what kind of image can obtain calculated projection data close to the measured projection data. As shown in FIG. 3, it is assumed that the projection data (synograms) of 0 °, 90 °, 180 °, and 270 ° are obtained. At this time, it is possible to predict what the cross-sectional image obtained from these projection data will look like. For example, with respect to the outer shape, it is expected to be an ellipse from the outermost synogram shape. In addition, the 90 ° and 270 ° synograms suggest that a bright substance is present in the upper part of the ellipse and an air layer is present in the lower part. Since there is no information about the material inside the ellipse at 180 ° and 270 °, it is expected that the high-intensity material and the air layer cancel each other out. The outline of the ML-EM reconstruction method is a method of constructing a consistent cross-sectional image by repeatedly performing such an operation at the same time.

FIG. 4 shows a comparison result between the cross-sectional image reconstructed by using the ML-EM reconstruction method and the cross-sectional image reconstructed by using the filter-corrected back projection method (hereinafter referred to as “FBP method”). In the cross-sectional image reconstructed by the FBP method, the occurrence of streak-like artifacts was confirmed. It was also clarified that the contrast between the hole inside the sample and the air layer outside was different. On the other hand, in the ML-EM method, such a phenomenon was not confirmed, but the outline of the hole was blurred. The FBP method is an effective reconstruction method for samples containing elements having significantly different linear attenuation coefficients, but it is less effective for artifacts derived from complex shapes such as plates and many protrusions. This is because the FBP method processes the bokeh correction filter at the time of reconstruction. In addition, there are problems such as the edge being emphasized and the contrast being different due to the influence of the correction filter. These problems cause a measurement error, and the measurement error may increase depending on the shape of the object to be measured. On the other hand, the ML-EM reconstruction method can suppress the occurrence of artifacts manifested by the FBP method.

However, since the ML-EM reconstruction method is a method of deriving a statistically most probable image based on projection data, it may not converge due to (1) statistical method, and (2) reconstruction image. Three problems have been pointed out: the edge of the image becomes unclear, and (3) the amount of analysis is huge and the reconstruction time is long. In order to actually use the ML-EM reconstruction method, it has been required to develop a method for solving these problems. Therefore, the inventor of the present invention corrects the entire X-ray CT image synogram (measured synogram) generated by the image reconstructing means 50 so as to converge to the ideal synogram created in advance, thereby recreating the ML-EM. The above problem of the construction method was solved. As the ideal synogram, a synogram of an X-ray CT image generated when a surface light source type X-ray source is used instead of the point light source type X-ray source 10 can be adopted.

FIG. 5 is an explanatory diagram for explaining a method of correcting an X-ray CT image using an ideal synogram. In this embodiment, an example of performing image correction using a calibration phantom F in which three spheres B as shown in FIG. 6 are arranged in a straight line as a measurement object O will be described. When an X-ray CT image of the calibration phantom F is acquired using a point light source type X-ray source 10 and a flat panel type detector 20, as shown in FIG. 7, the center of the detector 20 is obtained. It has been clarified that the farther the sphere B of the calibration phantom F is from the above, the larger the deviation (intercenter error) between the center position of the sphere B and the actual center position in the X-ray CT image. In the present embodiment, as shown in FIG. 5, the image is corrected so that _{the synogram (measured synogram) S M} of the X-ray CT image generated by the image reconstructing means 50 converges to the ideal synogram S _I. The center-to-center error of the sphere B can be reduced (or made zero). Such a method using a synogram can also be applied when correcting an image of another measurement object O.

The X-ray CT apparatus 1 preferably has a vibration isolating function as a countermeasure against vibration from the outside. Further, the X-ray CT apparatus 1 is preferably shielded by a shield made of lead, tungsten, or the like, and the temperature and humidity inside the X-ray CT device 1 are preferably maintained constant by air conditioning means. By doing so, it is possible to suppress the influence of the external environment when acquiring the image information, and it is possible to obtain a more accurate three-dimensional image.

Next, a method of correcting the X-ray CT image of the measurement object O will be described with reference to FIGS. 5 to 7 with reference to the flowchart of FIG.

First, the X-ray source 10 irradiates the measurement target O with X-rays, and the detector 20 detects the projection data for each rotation angle of the measurement target O, whereby the X-ray CT image of the measurement target O is detected. (Image acquisition step: S1). Then, sinogram (measured sinogram) of X-ray CT image of the measuring object O that has acquired the S _M, the display screen on the display unit 61 as shown in FIG. 5, for example (measured sinogram display step: S2).

Then, an ideal sinogram S _I prepared in advance, for example in advance and stored in storage means such as a hard disk or a memory, or receives from the outside through the appropriate communication means, display means as shown in FIG. 5 Displayed on the display screen at 61 (ideal synogram display step: S3). The ideal synogram display step S3 may be performed before the image acquisition step S1 and the actual measurement synogram display step S2.

Incidentally, the ideal sinogram S _I measures the surface and center coordinates of the ball B in advance the calibration phantom F in the three-dimensional measuring instrument (CMM), obtain an accurate surface coordinates and the center coordinates as the point cloud data for each ball B It can be obtained by doing. That is, the central axis of the center coordinates of the ball B is located in central calibration phantom F, when the one rotation of the calibration phantom F, can obtain the ideal sinogram S _I point group data in each ball B Is. The ball B and CT center coordinates of positioned at the center of the calibration phantom F match on software, and the ideal sinogram S _I of each sphere B of the calibration phantom F (point group data), sinogram obtained from the CT data And, the position of the flat panel type detector 20 is corrected on the software so that they match.

Then, by reconstructing the image using the ML-EM reconstruction method to converge the measured sinogram S _M to the ideal sinogram S _I, to correct the X-ray CT image (image correcting step: S4).

In the X-ray CT apparatus 1 according to the embodiment described above, a synogram (actual measurement) of an X-ray CT image of the measurement object O generated by using the point light source type X-ray source 10 and the flat panel type detector 20. The X-ray CT image can be corrected so that the synogram) converges to the ideal synogram. Therefore, it is possible to accurately correct the error of the projected image at the end of the flat panel type detector 20 that occurs when the point light source type X-ray source 10 is used.

In the above embodiment, as shown in FIG. 6, an example in which the calibration phantom F in which the three spheres B are arranged in a straight line is adopted is shown, but the configuration of the calibration phantom F is limited to this. It's not a thing. For example, as shown in FIG. 9, a calibration phantom F in which a plurality of spheres B are arranged radially can also be adopted.

Further, in the above embodiment, an example in which the X-ray CT image is corrected by using the ML-EM reconstruction method is shown, but another reconstruction method can be obtained by converging the actually measured synogram to the ideal synogram. (For example, filter-corrected back projection method, additive ART method, multiplication ART method, SIRT method, gradient method, steepest descent method, conjugate gradient method, MAP-EM method, Convex method, etc.) It can also be corrected.

The present invention is not limited to the above embodiments, and those skilled in the art with appropriate design changes are also included in the scope of the present invention as long as they have the features of the present invention. .. That is, each element included in the embodiment and its arrangement, material, condition, shape, size, etc. are not limited to those exemplified, and can be appropriately changed. In addition, the elements included in the embodiment can be combined as much as technically possible, and the combination thereof is also included in the scope of the present invention as long as the features of the present invention are included.

1 ... X-ray CT apparatus 10 ... X-ray source 20 ... detector 60 ... image correcting means O ... measuring object S _I ... ideal sinogram S _M ... Found sinogram S4 ... image correction step

Claims (9)

A point light source type X-ray source that irradiates an object to be measured with X-rays and a flat panel type detector that detects characteristic X-rays that have passed through the object to be measured are provided, and the detection is performed by the flat panel type detector. An X-ray CT apparatus that generates an X-ray CT image of the object to be measured based on characteristic X-rays.
An image correction means for correcting the X-ray CT image so as to converge the generated X-ray CT image synogram to an ideal synogram is provided .
The ideal sinogram, Ru sinogram der X-ray CT image generated in the case of using the surface light source type X-ray source as X-ray source, X-ray CT apparatus. The X-ray CT apparatus according to claim 1 , wherein the image correction means corrects the X-ray CT image by using the maximum likelihood estimation / expected value maximization reconstruction method. The image correction means uses the filter correction back projection method, the additive ART method, the multiplication ART method, the SIRT method, the gradient method, the steepest descent method, the conjugate gradient method, the MAP-EM method, or the Convex method. correcting the line CT image, X-rays CT apparatus according to claim 1. A point light source type X-ray source that irradiates an object to be measured with X-rays and a flat panel type detector that detects characteristic X-rays that have passed through the object to be measured are provided, and the detection is performed by the flat panel type detector. An image correction method for an X-ray CT apparatus that generates an X-ray CT image of the object to be measured based on characteristic X-rays.
An image correcting step of correcting said X-ray CT images so as to converge the sinogram generated the X-ray CT image to the ideal sinogram seen including,
The ideal synogram is an image correction method that is a synogram of an X-ray CT image generated when a surface light source type X-ray source is used as an X-ray source. The image correction method according to claim 4 , wherein in the image correction step, the X-ray CT image is corrected by using the maximum likelihood estimation / expected value maximization reconstruction method. In the image correction step, the X is used by using the filter correction back projection method, the addition type ART method, the multiplication type ART method, the SIRT method, the gradient method, the steepest descent method, the conjugate gradient method, the MAP-EM method, or the Convex method. The image correction method according to claim 4 , wherein the line CT image is corrected. Image correction that corrects the X-ray CT image so that the computer converges the synogram of the X-ray CT image of the measurement object generated by using the point light source type X-ray source and the flat panel type detector to the ideal synogram. An image correction program that executes a process
The ideal synogram is an image correction program which is a synogram of an X-ray CT image generated when a surface light source type X-ray source is used as an X-ray source. The image correction program according to claim 7 , wherein in the image correction step, the X-ray CT image is corrected by using the maximum likelihood estimation / expected value maximization reconstruction method. In the image correction step, the X is used by using the filter correction back projection method, the addition type ART method, the multiplication type ART method, the SIRT method, the gradient method, the steepest descent method, the conjugate gradient method, the MAP-EM method, or the Convex method. The image correction program according to claim 7 , which corrects a line CT image.

Images