JP4769487B2 - X-ray measuring device - Google Patents

Description

  The present invention relates to an X-ray measurement apparatus capable of correcting a saturation phenomenon and a protrusion phenomenon of a subject that occur in a detector and obtaining a good three-dimensional image with improved value uniformity.

  There is an X-ray measurement apparatus in which an X-ray source and a two-dimensional X-ray detector are installed at both ends of a C-shaped support (hereinafter referred to as C arm). There are shapes that suspend the C-arm from the ceiling and shapes that support the C-arm from the floor. In addition, there is an X-ray measurement device in which an X-ray source and a two-dimensional X-ray detector are installed on a gantry so as to face each other. In these apparatuses, by moving the C-arm or gantry, it is possible to perform X-ray measurement while rotating the X-ray source and detector around the subject. In addition, a plurality of measurement data obtained by rotational measurement are respectively corrected to obtain one set of projection data for three-dimensional reconstruction, and a three-dimensional reconstruction algorithm is obtained for the obtained one set of projection data. It is possible to obtain a three-dimensional reconstructed image by performing reconstruction processing using

  In these three-dimensional measuring devices, if the detector is saturated at the time of measurement or if the subject protrudes from the field of view of the detector, non-uniform values occur in the three-dimensional reconstructed image, and the image quality deteriorates. Japanese Patent Application Laid-Open No. 2004-151867 describes a method of extrapolating extra area data with an elliptic function in projection data of a subject. Japanese Patent Application Laid-Open No. 2004-259542 describes a method of extrapolating extrapolated area data with a function approximated using data in the vicinity of the boundary in the projection data of the subject.

JP 09-066051

JP2004-121853

  In the prior art disclosed in Patent Document 1, the projection data of the subject is supplemented by extrapolating the data of the region protruding from the detector field of view with an elliptic function centered on the position of the rotation axis. When the object is approximated by an ellipse with the rotation axis as the centerline, the profile of the projection data is an elliptic function, and this approximation is valid. However, in Patent Document 1, the elliptic function is obtained as a function that becomes 0 at a predetermined distance from the start point through the value of the start position of the extrapolation area centered on the position of the rotation axis. That is, the subject is an elliptical cylinder having the same cross-sectional shape predicted in advance along the rotation axis with the rotation axis as the center. For this reason, there has been a problem that the correction of the protrusion is not satisfactorily performed on the subject having a different cross-sectional shape along the rotation axis. Further, when measuring a subject having a large individual difference, there is a problem that correction cannot be applied to an arbitrary subject because a deviation from a basic shape predicted in advance occurs.

  In the prior art disclosed in Patent Document 2, the projection data of the subject is extrapolated by extrapolating the data of the region protruding from the detector field of view with the function approximated using the data in the vicinity of the protruding start position. In this method, the data of only the vicinity of the start position is used, and the inclination of the function is obtained by assuming these as points on the arc, so that the subject shape is not necessarily restored. For this reason, in order to perform the correction satisfactorily, there is a problem that it is necessary to rescale the amount of protrusion using data that does not protrude.

  In the prior arts shown in Patent Documents 1 and 2, there is a problem that when the data near the start position of the protrusion is saturated, a good extrapolation function cannot be obtained and correction cannot be performed satisfactorily. there were.

  An object of the present invention is to correct a non-uniformity in the value of a three-dimensional reconstructed image by correcting a saturation phenomenon and a protrusion phenomenon of a subject that occur in a detector, and an X that can obtain a good three-dimensional image. The object is to provide a line measuring device.

  The above-described object is to provide an X-ray source that generates X-rays to be irradiated on an inspection object, an X-ray detector that detects measurement data relating to the inspection object, and a holding device that holds the X-ray source and the X-ray detector facing each other. And an X-ray measurement apparatus having a rotation device that changes the relative positions of the X-ray source and the X-ray detector with respect to the inspection target and a control processing device that performs calculation processing of measurement data. Means for dividing the sensitivity data and subject measurement data, means for logarithmically transforming the divided data, means for determining the saturation area in the logarithmic transformation data, and extrapolating the value in the saturation area from the boundary point of the saturation area Means for converting the value in the protruding area of the subject into a value extrapolated from the end of the logarithmic conversion data, and performing reconstruction calculation using the logarithmic conversion data after the value conversion. It is achieved by X-ray measuring apparatus characterized by having a means for obtaining a dimension image.

  According to the present invention, it is possible to correct a saturation phenomenon and an object protrusion phenomenon that occur in a detector, and obtain a good three-dimensional image with improved value uniformity.

Hereinafter, embodiments of the present invention will be described.
FIG. 2 shows a side view of the X-ray measuring apparatus according to the present invention. The X-ray measuring apparatus includes an X-ray source 201, a detector 202, a support 203, a rotating device 204, a bed 205, and a control processing device 206. The X-ray source 201 and the detector 202 are installed on the column 203. For the support 203, a C-shaped arm, a U-shaped arm, a gantry, or the like is used. FIG. 2 shows a C-shaped arm. The form which suspends the support | pillar 203 from a ceiling and the form which supports the support | pillar 203 from a floor can be considered. The support column 203 is rotated by the rotating device 204 around the subject 208 lying on the bed 205 around the rotation axis 207. In FIG. 2, the case where the rotating shaft 207 and the bed 205 are parallel to the floor is shown as the most general form. It is also possible to set the rotating shaft 207 obliquely with respect to the body axis by moving the column 203 and the bed 205. Further, a configuration in which the rotation shaft 207 is perpendicular to the floor and the column 203 rotates around the subject 208 sitting on a chair is also possible.

  As the detector 202, a planar X-ray detector, a combination of an X-ray image intensifier and a CCD camera, an imaging plate, a CCD detector, a solid state detector, or the like is used. As a flat type X-ray detector, there is one in which an amorphous silicon photodiode and a TFT are paired and arranged on a square matrix, and this is directly combined with a fluorescent plate.

X-rays generated from the X-ray source 201 pass through the subject 208, are converted into electric signals corresponding to the X-ray intensity by the detector 202, and are input to the control processing device 206 as measurement data. The control processing device 206 controls the X-ray generation in the X-ray source 201, the acquisition of data in the detector 202, and the rotation of the column 203 in the rotating device 204. As a result, the X-ray measurement apparatus can perform rotation measurement for generating X-rays and acquiring measurement data while rotating the support column 203. The control / repair device 206 can perform logarithmic conversion processing, reconstruction calculation processing, and the like on the measurement data to acquire three-dimensional data.
A filter 210 can be installed between the X-ray source 201 and the detector 202. The filter 210 is made of a metal such as aluminum, copper, or brass, ceramic, resin, or the like. The filter shape is arbitrary, and generally a filter shape that matches the basic object shape is used.

  The control processing device 206 executes processing for correcting saturation and protrusion. The control processing device 206 has a storage means 209 inside, and stores functions, parameters, conditions, etc. necessary for correction processing. As input means, key input from a keyboard, reading from a file, and replacement of a storage chip can be considered. The control processing means 206 has a mode for inputting the presence / absence of execution of correction processing, a switch, or the like as an operation menu.

  FIG. 1 shows a processing procedure in this embodiment. Rotational measurement is performed to obtain a series of measurement images (101). An arbitrary image is selected from the series of measurement images (102). In the selected measurement image, one row of data is selected (103). A saturation region is extracted from the selected measurement data and converted to an arbitrary value to obtain saturation extraction data (104). When there is a variation in sensitivity for each detection element in the detector, the variation in the sensitivity of the detector is corrected for each pixel in the saturation extraction data to obtain sensitivity correction data (105). If there is a difference in imaging intensity such as X-ray intensity between the sensitivity data and the measurement data, the intensity deviation is corrected to obtain imaging intensity correction data (105). Logarithmic conversion is performed on the photographing intensity correction data to obtain logarithmic conversion data (hereinafter, projection data) (106). In the projection data, the value in the saturation region is converted to a value extrapolated from the boundary point of the saturation region to obtain saturation correction data (107). Saturation correction data is obtained for all rows on the selected measurement image. When discontinuity is observed in the column direction in the saturation region of the saturation correction data, the saturation correction data is smoothed in the column direction to obtain a smoothed image (108). If the subject is protruding from the detection area of the detector, values are extrapolated from the end of the data in the smoothed image to obtain a protruding corrected image (109). Overflow correction images are obtained for all measurement images. The projected image is back-projected to obtain a reconstructed image (110).

  A procedure of sensitivity correction processing and logarithmic conversion processing is shown. Measurement is performed by irradiating X-rays with no subject placed, and sensitivity data of the detector is acquired. Measurement is performed without X-ray irradiation, and detector offset data is acquired. The offset data is subtracted from the subject data to obtain subject data after offset correction. Subtract offset data from sensitivity data to obtain sensitivity data after offset correction. Subject data after offset correction is divided by sensitivity data after offset correction to obtain subject data after sensitivity correction (sensitivity correction processing). A logarithmic conversion process is performed on the subject data after sensitivity correction, and multiplied by −1 to obtain projection data (logarithmic conversion process).

  A procedure for obtaining saturation extraction data from measurement data will be described. In arbitrary measurement data, an arbitrary pixel is set as a target position. When the value of the target position is smaller than the threshold value S and a value greater than the threshold value S continues to the left side for N points or more, the target position is set as the left boundary point. A pixel that is on the left side of the left boundary point and has a value larger than the threshold value S is determined as a saturated region, and is replaced with a specific numerical value, for example, -1000. Similarly, when the value of the position of interest is smaller than the threshold value S and a value greater than the threshold value S continues on the right side for N points or more, the position of interest is set as the right boundary point. A pixel that is on the right side of the right boundary point and has a value larger than S is determined as a saturated region, and is replaced with a specific numerical value, for example, -1000.

  Since the pixel in the saturation region has a specific numerical value, for example, -1000, it can be distinguished during logarithmic conversion. During logarithmic conversion, a pixel having a value of −1000 sets the converted value to a specific numerical value, for example, 0. This makes it possible to determine pixels in the saturation region in the projection data. By setting the pixel value of the saturated region to 0 in the projection data, it can be set to a value when no subject exists, that is, a reference value.

  A procedure for obtaining saturation correction data from projection data will be described. The pixel value and the gradient of the pixel value are obtained at the boundary point of the saturation region. An extrapolation function is determined from the pixel value and the slope. An extrapolation value is calculated using an extrapolation function for a region where the pixel value is a singular number, for example, 0, and the pixel value is replaced with the extrapolation value. When the extrapolated value can be calculated from both the right and left sides of the saturation region, an average value of the extrapolated values is obtained, and the pixel value is replaced with the average value.

  When a filter for adjusting the intensity of irradiated X-rays is installed between the X-ray source and the subject, the saturation value in the projection data depends on the position if the filter thickness is not uniform. Therefore, saturation is detected on the measurement data. On the other hand, when the thickness of the filter is uniform, the saturation value in the projection data is constant without depending on the position, so the saturation can be determined using data after logarithmic conversion.

  FIG. 3 shows an example of data when the filter thickness is constant. As shown in (a), in sensitivity data measured without setting a subject under a condition that does not cause saturation, the value of a pixel that has passed air (hereinafter, air value) is constant. As shown in (b), when there is saturation in the measurement data of the subject, a value larger than the saturation value So such as the air value At sticks to the saturation value So. In this case, as shown in (c), the saturation value S is expressed by Equation 1 in logarithmically converted data.

In the logarithm conversion data, if the pixel value is near the saturation value, it is determined as the saturation region. Specifically, the saturation value S is set as a threshold value, and a pixel having a value smaller than the threshold value is determined to be a saturation region. Alternatively, by setting a value obtained by multiplying the saturation value by, for example, a coefficient of 1.05 as the threshold value, it is possible to make it less susceptible to data fluctuations.

  FIG. 4 shows an example of data when the filter thickness is not uniform. As shown in (a), the case where the cross section of the filter is concave, the cross section of the subject is circular, and the center of the subject is shifted from the center of the filter is shown. As shown in (b), the sensitivity data is measured data of the filter without saturation and is represented by a dotted line. Subject measurement data is represented by a solid line when measured under X-ray conditions without saturation. In the measurement of the subject, if the intensity of the irradiated X-ray is increased, the value of the measurement data exceeds the saturation value, for example, the range from position A to B is the saturation region. As shown in (c), the projection data is represented by a thick line in the saturation region, and deviates from the true data represented by the broken line.

  FIG. 5 shows the extrapolation of projection data. Assume that the X-ray absorption coefficient of the subject is μ, the cross section of the subject is a circle with a radius a, and the center is at position c. The projection data y is expressed by the elliptic function of Equation 2 with respect to the position x.

Therefore, projection data at a position larger than the center is approximated by Expression 3, and projection data at a small position is approximated by Expression 4.

Specifically, for each line of each projection data, parameters a and c are calculated from the value y _o and the slope y ′ (x _o ) at the position x _o that is the start point of extrapolation, that is, the boundary point of the saturation region. Find the elliptic function formula. In this case, the boundary condition of Equation 3 is expressed by Equation 5 and Equation 6.

Therefore, the parameters a and c are obtained by Equation 7 and Equation 8.

The boundary conditions of Equation 4 are expressed by Equations 9 and 10, and parameters a and c are obtained by solving these equations in the same manner as Equation 3.

A method of calculating the slope at the position x _o that is the start point of extrapolation will be described. A pixel value at the start point position is obtained on the projection data, and similarly, a pixel value at a position separated by an arbitrary pixel in the direction opposite to the extrapolation direction is obtained. The slope of the line connecting the values of these two points is obtained and used as the slope of the starting point. When calculating the value, data of an arbitrary region is extracted centering on the extrapolation start point, and the addition average value is obtained and used as the start point value, thereby making it less susceptible to data fluctuations. it can. Specifically, data of 5 pixels is extracted from the position x _{o in the} direction opposite to the extrapolation direction and 9 pixels in the direction orthogonal to the extrapolation direction.

After the extrapolation process, the projection data is smoothed. Specifically, for projection data extrapolated in the row direction, data in an arbitrary area is extracted around the start point of the extrapolation, and the average value is obtained, and the pixel value at the start point is replaced. To do. For example, data of 5 points in the row direction and 9 points in the column direction are used. Smoothing makes it less susceptible to data fluctuations, and prevents data values from becoming discontinuous, particularly in the column direction.
When obtaining the average value, each data may be weighted. For example, when the weight for the pixel value at the extrapolation start point position is 1.0 and the weight for the surrounding pixel values is 0.5, it is possible to prevent the resolution of the projected image from being lowered.

  FIG. 6 shows an example of data when there are both saturation and protrusion in the projection data. Assuming that the projection data in which the region from the position D to E is measured, the region from the position C to D and the region from the position E to F are protruding regions. The projection data is extrapolated in these protruding areas. The extrapolation function is the same as that for saturation correction, and is calculated using equations (3) and (4). By using the same function, saturation and protrusion can be corrected at a time. It is also possible to first correct the saturation in the saturation region from the position A to B, and then correct the protrusion in the protrusion region from the position C to D. In this case, it is also possible to use a function different from the extrapolation function for saturation correction as the extrapolation function for protrusion correction.

  The effects of the above-described embodiment are summarized as follows. (1) It is possible to correct a saturation phenomenon and an object protrusion phenomenon that occur in the detector, and obtain a good three-dimensional image with improved value uniformity. (2) By determining the saturation region in the measurement data and correcting the saturation in the logarithmic conversion data, the saturation phenomenon can be corrected even when the filter shape is not uniform. (3) In the saturation region, the extrapolated value is obtained from both the right side and the left side of the saturation region and is replaced with the average value of both, thereby improving the correction accuracy. (4) By making the extrapolation functions used for the saturation correction and the protrusion correction the same, the saturation and the protrusion can be corrected at a time, and the speed can be increased. (5) By performing smoothing in the direction orthogonal to the direction of extrapolation after extrapolation processing, it is made less susceptible to data fluctuations, and in particular prevents discontinuity of data values in the column direction. be able to.

It is explanatory drawing of the correction | amendment procedure which concerns on this invention. It is a side view of the apparatus which concerns on this invention. It is an example of data concerning the present invention. It is an example of data concerning the present invention. It is explanatory drawing of the extrapolation process which concerns on this invention. It is an example of data concerning the present invention.

Explanation of symbols

201: X-ray source, 202: detector, 203: support, 204: rotating device, 205: bed, 206: control processing device, 207: rotating shaft, 208: subject, 209: storage device, 210: filter.

Claims (4)

An X-ray source that irradiates the subject with X-rays, an X-ray detector that is disposed opposite to the X-ray source and detects transmitted X-rays of the subject, and supports the X-ray source and the X-ray detector, A rotating device that rotates around the subject, and image reconstructing means that obtains transmission X-rays of the subject at multiple angles by the rotation as measurement data, and reconstructs a three-dimensional image from the obtained measurement data. In the X-ray measuring apparatus provided,
Means for extracting a saturation region from each measurement data of the subject, Sensitivity data measurement unit for measuring sensitivity data of the X-ray detector for the extracted saturation region, and correcting variations in the measured sensitivity data Sensitivity data correction means, photographing intensity deviation correcting means for correcting photographing intensity deviation between the corrected sensitivity data and the measurement data, and calculating projection data by logarithmically converting the measurement data with the photographing intensity deviation corrected. Projection data calculating means for converting the value of the saturation region in the calculated projection data from the boundary point of the saturation region into an extrapolated value using an extrapolation function for obtaining saturation correction data, and obtaining saturation correction data When the saturation correction data calculation means and the subject protrude from the detection area of the X-ray detector, the saturation correction data calculation means corrects the protrusion. Using the converted extrapolated value using the extrapolation function of use, protruding obtain a corrected image off the edge of the data it is determined that protrudes a correcting image calculating means,
The image reconstruction means includes
An X-ray measurement apparatus, wherein the obtained protrusion correction image is backprojected to reconstruct a three-dimensional image. When there is discontinuity in a predetermined one-dimensional direction in a saturation region of the saturation correction data, the saturation correction data is further smoothed in the one-dimensional direction, and further includes a smoothed image calculation unit that obtains a smoothed image. The X-ray measurement apparatus according to claim 1, wherein The protrusion correction image calculation means obtains a pixel value at the boundary point of the protrusion area in the projection data, extracts an arbitrary number of pixel values in the direction opposite to the protrusion area from the boundary point, and outputs the pixel at the boundary point. The slope of the value is obtained, and an elliptic function having a pixel value and a slope substantially equivalent to the boundary point at the boundary point position of the protrusion around the rotational axis position of the rotating device is obtained, and the obtained elliptic function is used. 2. The X-ray measurement apparatus according to claim 1, wherein an extrapolation value is calculated in the projection area, and a pixel value in the projection area is set to an extrapolation value, thereby correcting the projection in the projection data. . The X-ray measurement apparatus according to claim 1, wherein the saturation correction data and the overhang correction image are calculated using the same extrapolation function.

Images