JP6663094B2 - Scattered-ray intensity calculating apparatus, scattered-ray correcting apparatus including the same, and scattered-ray intensity calculating method - Google Patents

Description

  The present invention relates to a scattered radiation intensity calculation device and a scattered radiation intensity calculation method for calculating the intensity of a scattered radiation component included in an image obtained by X-ray photography.

  An image obtained by X-ray imaging includes a direct ray component caused by direct radiation transmitted through a subject and a component other than the direct ray component. As components other than the direct ray component, there are a scattered ray component caused by scattered rays, and a component caused by light scattering of the X-ray detection element and electric noise generated from the system of the X-ray imaging apparatus. The scattered ray component is dominant. Examples of the scattered radiation include scattered radiation from a subject and scattered radiation from carbon provided in front of a flat panel detector.

  In particular, at the pixel position corresponding to the position where the X-ray attenuation of the subject becomes large, most of the X-rays captured by the detector may become scattered radiation, and the linearity is largely lost, which causes an artifact to appear. Then, this causes deterioration of the image quality of the image obtained by the X-ray imaging, which may cause a misdiagnosis or miss a lesion to be originally treated. Therefore, in order to suppress the deterioration of the image quality of the image obtained by X-ray imaging, it is necessary to derive an image with reduced scattered radiation components from the image obtained by X-ray imaging.

Toshimitsu Sakuma, 7 others, `` Scattered radiation measurement by the lead disk method '', Bulletin of Tohoku University College of Medical Technology, Tohoku University College of Medical Technology, 1995, Vol. 4, No. 2, p. 119-124

  Non-Patent Document 1 proposes scattered radiation measurement by the lead disk method. However, there is no disclosure of a specific method of calculating the scattered radiation intensity that can be applied even when an arbitrary different living body is used as a subject.

  The present invention has been made in view of the above circumstances, and has an object to provide a scattered radiation intensity calculation apparatus that can be applied even when an arbitrary living body is used as a subject, a scattered radiation correction apparatus including the same, and a scattered radiation intensity calculation method. Is what you do.

  In order to achieve the above object, in the scattered radiation intensity calculation apparatus according to the present invention, the calculation result of the scattered radiation intensity with respect to the total radiation intensity in a specific frame at a specific pixel position on a detector used in X-ray imaging of a subject The ratio and the X-ray image of the subject in the specific frame with respect to the total radiation intensity detected at the specific pixel position in the X-ray image without the subject in the specific frame. A storage unit that stores a parameter of a curve expression indicating a relationship with a ratio of the detected total radiation intensity, and at the specific frame at the specific pixel position, the total radiation intensity detected by X-ray imaging of the subject or And a calculation unit that calculates the ratio of the scattered radiation intensity to the direct radiation intensity using the curve equation (first configuration).

  In the scattered radiation intensity calculating apparatus according to the first configuration, a plurality of the specific pixel positions may be set, and a plurality of the specific frames may be set (a second configuration). Further, in the scattered radiation intensity calculation device of the first or second configuration, the storage unit stores the scattered radiation intensity at the pixel position other than the specific pixel position on the detector in the specific frame at the specific pixel position. A configuration (third configuration) for storing an interpolation formula for performing interpolation based on the calculation result of the scattered radiation intensity or interpolation data calculated from the interpolation formula may be adopted.

  In order to achieve the above object, a scattered radiation correction apparatus according to the present invention comprises: a scattered radiation intensity calculation apparatus according to any one of the first to third configurations; and an X-ray imaging apparatus for the subject based on a calculation result of the calculation unit. And a correction unit that performs scattered radiation correction on the measurement image obtained in (4).

  In order to achieve the above object, in the scattered radiation intensity calculation method according to the present invention, the calculation result of the scattered radiation intensity relative to the total radiation intensity in a specific frame at a specific pixel position on a detector used in X-ray imaging of a subject is described. The ratio and the X-ray image of the subject in the specific frame with respect to the total radiation intensity detected at the specific pixel position in the X-ray image without the subject in the specific frame. A step of reading out from a storage medium a parameter of a curve expression indicating a relationship with a ratio of a detected total radiation intensity, and a total radiation detected by X-ray imaging of the subject in the specific frame at the specific pixel position The intensity or the ratio of the scattered radiation intensity to the direct radiation intensity is determined by the parameter read in the reading step. A calculation step of calculating using whole the curve formula, the structure comprising a (fifth configuration).

  ADVANTAGE OF THE INVENTION According to this invention, the scattered-ray intensity | strength calculation apparatus which can be applied even when arbitrary different living bodies are made into a subject, the scattered-ray correction apparatus provided therewith, and the scattered-ray intensity | strength calculation method can be implement | achieved.

Diagram showing positional relationship between lead disk, X-ray source, and subject Graph showing the luminance value of all radiation when the diameter of the lead disk is changed and the luminance value of scattered radiation without the lead disk Table showing errors in extrapolated data obtained by each approximation method Diagram showing an example of a specific pixel location on a detector Diagram showing the positions of 10 profiles Schematic diagram of pixel coordinates and luminance values of one block Table showing known brightness values The figure which shows the profile of the brightness value in one block Image of brightness value due to scattered radiation The figure which shows the profile of the brightness value by scattered radiation Diagram showing an example of a specific pixel position on a measurement image Graph showing the luminance value of all radiation when the diameter of the lead disk is changed and the luminance value of scattered radiation without the lead disk Diagram showing axial image A graph showing a measurement image showing the pixel area of interest and a calculation result of parameters in the pixel area of interest. A graph showing a measurement image showing the pixel area of interest and a calculation result of parameters in the pixel area of interest. A graph showing a measurement image showing the pixel area of interest and a calculation result of parameters in the pixel area of interest. A graph showing a measurement image showing the pixel area of interest and a calculation result of parameters in the pixel area of interest. A graph showing a measurement image showing the pixel area of interest and a calculation result of parameters in the pixel area of interest. A graph showing a measurement image showing the pixel area of interest and a calculation result of parameters in the pixel area of interest. A graph showing a measurement image showing the pixel area of interest and a calculation result of parameters in the pixel area of interest. A graph showing a measurement image showing the pixel area of interest and a calculation result of parameters in the pixel area of interest. A graph showing a measurement image showing the pixel area of interest and a calculation result of parameters in the pixel area of interest. A graph showing a measurement image showing the pixel area of interest and a calculation result of parameters in the pixel area of interest. A graph showing a measurement image showing the pixel area of interest and a calculation result of parameters in the pixel area of interest. A graph showing a measurement image showing the pixel area of interest and a calculation result of parameters in the pixel area of interest. Diagram showing an example of a specific pixel location on a detector Graph showing the luminance value of total radiation and the luminance value of scattered radiation Graph showing the calculation results of parameters in the pixel area of interest Graph showing a difference from a luminance value due to scattered radiation calculated from a quartic curve Graph showing the luminance value of all radiation and the luminance value of scattered radiation after correction A diagram showing an axial image when a living body is set as a subject FIG. 1 is a diagram showing a configuration of a scattered radiation correction device according to an embodiment of the present invention.

  An embodiment of the present invention will be described below with reference to the drawings. The scattered radiation intensity calculation apparatus according to the present invention and the result of examination and evaluation for developing the scattered radiation correction apparatus including the same will be described, and finally the scattered radiation intensity calculation apparatus according to the present invention and the scattered radiation including the same The correction device will be described. In the present embodiment, a pixel value which is a value counted for each frame by each pixel of the detector is referred to as a luminance value. The luminance value includes the contribution of direct radiation and the contribution of scattered radiation. In the present embodiment, the former is referred to as a luminance value due to direct radiation or a luminance value of direct radiation, and the latter is referred to as a luminance value due to scattered radiation or a luminance value of scattered radiation. The luminance value of the direct radiation or the luminance value of the direct radiation corresponds to the intensity of the direct radiation, and the luminance value of the scattered radiation or the luminance value of the scattered radiation corresponds to the intensity of the scattered radiation.

<1. Calculation of scattered radiation captured by each pixel of the detector>
<1-1. Brightness value of scattered radiation at pixel corresponding to position of lead disk>
The brightness value of the scattered radiation in the sensitive area of the detector can be determined experimentally. As a method, several types of lead disks 1 having different diameters are installed between an X-ray source 2 and a subject 3 as shown in FIG. 1 to perform CT imaging, and a detector corresponding to the center position of the lead disk 1 is used. The luminance value at the pixel position of No. 4 is compared and extrapolated.

  It can be considered that the lead disk can substantially block X-rays if it has a sufficient thickness. Therefore, it can be considered that the brightness value at the pixel position of the detector corresponding to the center position of the lead disk does not directly include a radiation component but reflects only scattered radiation. The scattered radiation reflected on the luminance value at the pixel position of the detector corresponding to the center position of the lead disk is considered to be scattered radiation generated when X-rays passing outside the lead disk irradiate the subject. Can be

  However, since the X-rays applied to the lead disk are absorbed and not applied to the subject, the scattered radiation is reduced accordingly. And, the decrease becomes smaller as the diameter of the lead disk becomes smaller. That is, the smaller the diameter of the lead disk, the more scattered radiation. Then, extrapolating the relationship between the diameter of the lead disk and the scattered radiation and calculating the luminance value (extrapolated data) of the scattered radiation when the diameter is 0, ordinary CT imaging without the lead disk is performed. This can be considered as the luminance value of the scattered radiation reflected at that pixel when performed.

<1-2. Study on extrapolated data>
1-1. In order to study a specific method for obtaining extrapolated data described in Section 2, the behavior of X-rays when a mathematical phantom simulating a head phantom is irradiated with an X-ray flux of an X-ray spectrum emitted from an X-ray tube is The number of direct and scattered radiation captured by each pixel of the detector was calculated by using the Monte Carlo method. Since the detector is of the accumulation type, it is strictly different, but it can be considered that the number of X-rays captured by the pixel and the obtained luminance value are substantially proportional to each other.

The head phantom created the geometry such that the shape and composition of the anterior and posterior teeth, upper jaw, lower jaw, cervical vertebra, soft tissue, etc. were as close as possible to the real one. In actual CT imaging, the X-ray flux is cut by a lead collimator directly under the X-ray tube so that X-rays are directly incident only on the sensitive area of the detector in order to suppress extra exposure of the patient. The same calculation was also cut. Direct X-rays and scattered X-rays were calculated when X-rays were detected at pixels that should be captured without being scattered by the subject, and were considered as scattered radiation when X-rays were captured at different pixels. The sum of the direct radiation and the scattered radiation corresponds to the radiation detected by the detector in actual CT imaging. Here, the sum of the direct radiation and the scattered radiation is defined as the total radiation.

  When the diameter of the lead disk placed between the X-ray source and the head phantom is changed to 2 mm, 4 mm, 8 mm, 12 mm, 16 mm, 20 mm, and 24 mm, the luminance value of the total radiation when calculated by the Monte Carlo method is excluded. In the case of 0 mm calculated by insertion, that is, the luminance value of the scattered radiation when the lead disk is not installed is compared with the luminance value of the scattered radiation in the case of 0 mm calculated by the Monte Carlo method.

  FIG. 2 is a graph showing the luminance value of all radiation when the diameter of the lead disk is changed and the luminance value of scattered radiation when there is no lead disk in front irradiation, side irradiation, and back irradiation, respectively. . As an approximation method for calculating a value extrapolated from the former value, (I) when approximating with an axially symmetric 6th-order curve, (II) an axially symmetric 6th-order curve and two points of 2 mm and 4 mm in diameter are used. Three patterns were examined: an approximation by an average value between the straight lines to be connected, and (III) an approximation by the least square method of a linear equation.

  FIG. 3 shows the difference between the luminance value of scattered radiation obtained by using each approximation method without using a lead disk and the luminance value of scattered radiation calculated by the Monte Carlo method at 0 mm in each irradiation pattern. FIG.

  The approximation method (II) has the most stable and small error as a whole. Therefore, in the following description, the luminance value of the scattered radiation is calculated by the approximation method (II).

  However, an approximation method other than the approximation method (I), the approximation method (III), or the approximation methods (I) to (III) may be employed. Note that the error is reduced by using the sixth-order equation as in the approximation method (II) or the approximation method (I). Therefore, it is desirable to perform the approximation using the sixth-order equation.

  The luminance value due to scattered radiation at a specific pixel position on the detector can be obtained as described above. In the following description, the position of the lead disk set in FIG. 1 is adjusted such that the pixel position on the detector corresponding to the position is one of 5 × 5 = 25 black circles shown in FIG. Imaging is performed, and scattered radiation is calculated from the obtained experimental data.

<1-3. Calculation of brightness value of scattered radiation in all pixels on detector>
On the detector, the scattered radiation at each pixel position does not change so rapidly even if the pixel coordinates change, but changes slowly. Since the brightness values of the scattered radiation at 25 points in FIG. 5 are known, it is possible to calculate the brightness values of the scattered radiation in the remaining pixels so that the brightness values change smoothly. Prior to the photographing experiment, this was demonstrated by calculation and is shown below.

  1-2. Similarly to the above, the number of direct radiation and scattered radiation taken in each pixel of the detector when X-rays are irradiated with the geometry from which the lead disk has been removed is calculated using the Monte Carlo method. At this time, considering a profile of the luminance value of the scattered radiation taken on one row of pixels parallel to the horizontal axis or the vertical axis direction with respect to the pixel number, these draw a gentle curve. This profile can be approximated by a quartic curve by the least squares method if the values of the five points are known. Considering the 25 black circles shown in FIG. 4, a total of 10 profiles can be approximated in the vertical and horizontal directions. Then, the luminance value of the scattered radiation at the pixel on the solid line shown in FIG. 5 can be calculated by approximation with a quartic curve. From these luminance values, the luminance values of the scattered radiation in the other pixels are calculated. The procedure is shown below.

  Since the pixel on the detector always belongs to one of the 16 blocks separated by the solid line shown in FIG. 5, the luminance value Bv (x, y) of the scattered radiation at the pixel at the coordinates (x, y) in the block to which the pixel belongs Is calculated. FIG. 6 is a schematic diagram showing the coordinates and luminance values of the pixels of this one block.

The coordinates of the corners of the block are (x ₀ , y ₀ ), (x ₁ , y ₀ ), (x ₀ , y ₁ ), and (x ₁ , y ₁ ). Then, the brightness value Bv (x _0, y ₀₎ of the corner 4 points of a _{block, Bv (x 1, y 0} ), Bv (x 0, y 1), Bv (x 1, y 1) and, focused pixel ( luminance values Bv (x, y ₀ ), Bv (x ₁ , y), Bv (x, y ₁ ) at the pixel at the boundary position of the block and the straight line parallel to the horizontal and vertical axes passing through (x, y) , Bv (x ₀ , y) are already known because they are all luminance values at pixels for which luminance values can be calculated by approximation with the above-described quartic curve.

When moving the pixel position from (x, y ₀ ) to (x, y), (x, y ₁ ), Bv (x, y ₀ ), Bv (x, y), Bv (x, y ₁ ) change in the luminance value to _{the, Bv (x 0, y 0} ), Bv (x 0, y), the change in luminance value of the _{Bv (x 0, y 1)} , _{and, Bv (x 1, y 0} ) , Bv (x ₁ , y) and Bv (x ₁ , y ₁ ). At this time, since the brightness value changes slowly, the brightness value ratio Bv (x, y) / Bv (x ₀ , y) is Bv (x, y ₀ ) / Bv (x ₀ , y ₀ ) and Bv It is reasonable to think that the value between (x, y ₁ ) / Bv (x ₀ , y ₁ ) changes slowly.

Assuming that the luminance value with changes in the y coordinate varies monotonically, it is possible to determine the ratio of the luminance value Bv (x, y) / Bv (x 0, y), Bv (x 0, y ) Is known, so that Bv (x, y) can be calculated. This is Bvs1 (x, y). On the other hand, Bv (x, y) / Bv (x ₁ , y) is Bv (x, y ₀ ) / Bv (x ₁ , y ₀ ) and Bv (x, y ₁ ) / Bv (x ₁ , y ₁ ) Can be considered to change gradually, so that Bv (x, y) can be calculated from this, and this is set as Bvl1 (x, y). In general, Bvs1 (x, y) and Bvl1 (x, y) do not match, so this is set as Bv1 (x, y) as a temporary value of Bv (x, y).

However, an attempt to apply it to all the pixels in the block, the mismatch of the luminance value occurs at the boundary portion of the block (when x = x ₀ and x = x _1). Therefore, Bv00 x = x ₀ and x = Bv when the x ₁ (x, y) the value of each (y), and Bv01 (y). Originally, the Bv1 (x _0, y) is Bv00 (y), Bv1 (x 1, y) is because it must match the Bv01 (y), so as to respectively coincide with x = x ₀ and x = x ₁ Bv1 (x, y) is corrected as follows, and the corrected value is set as Bv (x, y).

Here, pixel numbers of 0 to 50 are given to two-dimensional pixel coordinates (x, y) with a detector of 51 pixels × 51 pixels. Of the detected luminance values, all the pixels on the pixel lines of x = 0, x = 50 and y = 0, y = 50 are known, and when applied to FIG. 7, x ₀ = 0, x ₁ = 50, y ₀ = 0, y ₁ = 50. For example, assuming that x = 25 and y = 20, the known luminance value is a value shown in FIG.

  Bv1 was calculated from the known luminance values shown in FIG. Then, the value of Bv (25,20) was calculated by applying Expression (2), and it was 1135. When the luminance values of the pixels inside the pixel line are calculated in the same manner and the positions of the pixels are represented on a horizontal plane and the luminance values are taken in the height direction and expressed in three dimensions, the profile of FIG. It became so. The change according to the position of the profile shown in FIG.

  Regarding the luminance value obtained in this way, an error was estimated from a luminance value due to scattered radiation calculated by the Monte Carlo method using a mathematical phantom simulating a head phantom as a subject. In the 25 pixel regions, the luminance values due to the scattered radiation calculated by the Monte Carlo method are used, and the profiles of the luminance values of a total of 10 pixel lines, each of which is 5 lines in the vertical and horizontal directions, are represented by a quartic curve. Approximately, the luminance values of the other pixels are calculated from the equation (2) by the method described above.

  Then, a profile (hereinafter, referred to as a “first profile”) in which the luminance values of the pixels other than the above-described ten pixel lines are calculated by Expression (2) is used as the luminance values of all the pixels that have already been calculated by the Monte Carlo method. (Hereinafter, referred to as “second profile”). FIG. 9A is an image image of a luminance value due to scattered radiation in the case of front irradiation obtained from the second profile, and FIG. 9B is a luminance value due to scattered radiation in the case of front irradiation obtained from the first profile. It is an image image of. Comparing the two, it can be seen that FIG. 9 (b) reproduces FIG. 9 (a) well.

For the first profile and the second profile, when the profile of the luminance value due to the scattered radiation at Y = 350 in the case of the front irradiation, the side irradiation, and the back irradiation was examined, FIG. 10 (a), FIG. The result is as shown in FIG. FIG. 10 also shows a third profile obtained by smoothing the second profile. The average error between the second profile and the first profile was calculated to be 4.09%, 4.02%, and 6.44%. On the other hand, the average errors of the third profile and the first profile were 0.86%, 2.76%, and 3.03%, respectively, and it is considered that the variation between pixels of the luminance value calculated by the Monte Carlo method influenced the error. When errors were similarly calculated for Y = 300 and 400, the results are as shown in Table 1 and show the same tendency. Here, in Table 1, MC indicates the second profile, Sm indicates the third profile, and F (1) indicates the first profile. The reason why the backside irradiation at Y = 400 has a large error is that there is a large shift between both end portions that does not significantly affect the dentition portion. When the error was calculated using only the 350 pixels near the center, the error was 1.42%, so that the influence on the CT reconstructed image was small.

<2. Experimental method>
As an X-ray imaging apparatus, AUJE SOLIO manufactured by Asahi Roentgen Industries, Inc. was used. In a dental X-ray CT apparatus, an arm provided with an X-ray tube and a detector at both ends rotates around a subject by 360 degrees. The number of frames is 510 and the frame rate is 30 fps. The sensitive area of the detector is 156.96 mm × 159.36 mm (654 pixels × 664 pixels), the element of the detector is CsI, and the thickness is 700 μm. The distance from the focal point to the detector is 600 mm, and the distance from the focal point to the rotation center is 375 mm. The tube voltage of the X-ray tube was 85 kV, and the tube current was 4 mA.

  For shooting, when the head phantom is set at the optimum position, when it is shifted by -5 mm, -3 mm, +3 mm, and +5 mm in the horizontal direction, when it is shifted by -5 mm and +5 mm in the front-rear direction, it is shifted by -5 mm and +5 mm in the vertical direction 9 patterns were performed. In practice, this was achieved by fixing the phantom and adjusting the center position and height of the arm.

  For each pattern, a normal photograph without a lead disk was taken, and a photograph was taken with lead disks of different diameters placed one by one so that the center position was at the position corresponding to the 25 positions shown in FIG. . The diameters of the lead disks are five types of 2.3 mm, 4.7 mm, 8.2 mm, 11.8 mm, and 15.5 mm. Therefore, photographing was performed 5 × 25 + 1 = 126 times in one pattern, and 9 × 126 = 1134 times in total. For each shooting, there are 510 (the number of frames) measurement image data.

<3. Experimental results>
<1. Calculation of the scattered radiation captured by each pixel of the detector> Based on the method described in <1>, for each pattern, each position of the lead disk, and the imaging data of each frame, the position of the lead disk with respect to the change in the diameter of the lead disk The average luminance value of the corresponding pixel area was graphed, and the luminance value in the case where the diameter was 0 was extrapolated to calculate the luminance value due to scattered radiation at 25 places in the sensitive area. Of these, in the pattern in which the head phantom is installed at the optimal position, frame Nos. 1 (front illumination), 128 (side illumination), 256 (back illumination) shown in FIGS. 11 (i) to 11 (iv). 12 (a) to 12 (b) are graphs in which the abscissa represents the diameter of the lead disk and the ordinate represents the luminance value for four pixel regions a to d of the measurement image of 384 (side irradiation). It is shown in d). However, the luminance value when the diameter is 0 mm is a value calculated by extrapolation using the luminance values at five points when the diameter is 2.3 mm to 15.5 mm. Also, FIG. 11 (i) shows the case where the lead disk corresponds to the position c, FIG. 11 (ii) shows the position d, FIG. 11 (iii) shows the position a, and FIG. 11 (iv) shows the position b. A measurement image is shown as an example. Note that the luminance value due to the dark current has been subtracted.

  In the graphs shown in FIGS. 12A to 12D, the luminance value of the scattered radiation is large in a portion where the thickness of the subject is small and the luminance value of the measurement image is high, and is small in a portion where the subject is thick. The characteristic of scattered radiation appears. In addition, in all cases, the extrapolated luminance value at 0 mm and the other luminance values obtained by imaging draw a smooth curve, and it is considered that an appropriate luminance value of scattered radiation is calculated by this method. May be. Then, the luminance value of the scattered radiation in the sensitive area of the detector can be similarly calculated for other patterns in which the position of the head phantom is shifted.

<4. Effect of scattered radiation correction>
It is necessary to confirm how much the image quality of the CT reconstructed image is improved by performing the scattered radiation correction using the luminance value of the scattered radiation calculated by the above-described method.

  Therefore, the luminance values of the scattered radiation in 25 pixel regions for each frame were calculated from data obtained by photographing the head phantom in a pattern placed at the optimum position. Then, for each frame, the luminance value of the scattered radiation of the entire pixel of the detector is calculated using Expression (2), and the luminance value of the radiation (direct radiation + scattered radiation) at each pixel of the measurement image obtained by imaging is calculated. The scattered radiation correction was performed by subtracting the luminance value of the scattered radiation from.

  Then, a three-dimensional CT image obtained by reconstructing the measurement image after the processing was compared with an image to which scattered radiation correction proposed in Japanese Patent Application Laid-Open No. 2015-65593 was applied. The results are shown below.

  FIG. 13A shows an axial image picked up at one point when a measured image is reconstructed without performing scattered radiation correction.

  FIG. 13B shows an axial image picked up at one point when the image is reconstructed after applying the scattered radiation correction proposed in Japanese Patent Application Laid-Open No. 2015-65593 to the measured image. is there.

  FIG. 13C shows one axial image picked up when the image is reconstructed after performing scattered radiation correction on the measured image using the luminance value of the scattered radiation calculated by the method described above. Things.

  In FIG. 13B, artifacts appear in the direction along the inside of the dentition and between the teeth and between the molars, and there is not much change compared to FIG. 13A and the image quality is not much improved. 13 (c), the artifact has been eliminated. That is, it can be seen that the scattered radiation correction using the luminance value of the scattered radiation calculated by the above-described method is effective.

<5. Curve approximation of luminance value ratio of scattered radiation obtained by different positioning of head phantom in CT imaging>
<5-1. Calculation method of approximate curve>
In order to apply the above-mentioned scattered radiation correction to any position of the head phantom, and further to be able to apply it to any biological data, for each frame of 25 pixel regions corresponding to the position of the lead disk It is necessary to be able to calculate the luminance value of the obtained scattered radiation using a calculation formula. In this section, a curve calculation formula was derived using data obtained in nine types of positioning patterns of the head phantom, and parameters for each frame of 25 pixel regions were calculated. The method will be described below. It is assumed that the influence of the dark current has already been removed from the pixel luminance values described below.

The horizontal axis of the curve is the image taken with the head phantom installed for the brightness value obtained for each pixel when the image was taken without setting the subject (the measurement image obtained at this time is called a white image). In this case, since the radiation includes both direct radiation and scattered radiation, the radiation obtained at this time will be referred to as total radiation, and a lead disk is not provided. The vertical axis of the curve indicates the above <3. Experimental results>
Is the ratio of the luminance values of the scattered radiation calculated in the above, and that is Y.

  Then, both X and Y have values smaller than 1, and when the subject passes through a thin path, X approaches 1 and conversely, when the subject is thick, approaches 0. When the ratio of scattered radiation is large, Y approaches 1, and when it is small, it approaches 0. In a large subject such as a head phantom, the ratio of the scattered radiation tends to increase when the subject is thick, and to decrease when the subject is thin. In addition to the above tendency, taking into account these distributions in the case of graphing, the equation of the curve representing the relationship between X and Y is represented by the following equation (3) using the parameters a and b.

As described above, this applies to the nine data distributions obtained in each frame of each pixel area. Equation (3) can be transformed into the following equation (4) by taking logarithms on both sides and setting Y _log = _log Y and X _log = _log X, and Y _log can be expressed by a linear equation of X _log. .

  Therefore, by applying the least squares method, it is possible to calculate the value of the parameter a and the value of the parameter b by approximating the frame by the equation (3) for each frame of each pixel region. Therefore, since the value of X and the value of Y are uniquely determined by the equation (3), the value of Y can be obtained for any value of X. That is, in a specific frame of a specific pixel region, a luminance value due to scattered radiation can be uniquely calculated for an arbitrary luminance value obtained by imaging.

<5-2. Calculation of Parameters Using Least Square Method>
All the data groups obtained for the positions of the nine head phantoms photographed are approximated by equation (3), and the value of parameter a and the value of parameter b are determined for each of them. Compared. Of these, a graph of a characteristic curve is taken up, and an error between data obtained by photographing and the curve is calculated.

  FIGS. 14A to 25A show measurement images of patterns in which the position of the inner phantom of the measurement image of the frame of interest is located at the optimum position, and the pixel region of interest is indicated by a square mark. I have.

  FIGS. 14 to 25B are graphs showing distributions of X and Y in a focused pixel region of the focused frame and an approximate expression of a curve.

  FIG. 14 to FIG. 18 show curve approximation of a pixel area near the dentition. 15 and 18, since the thickness of the bone on the path of the transmitted X-ray changes due to the displacement of the position of the head phantom, the X-ray attenuation greatly changes, and the change amount of X is large. However, the distribution of these data groups can be approximated by the curve represented by the equation (3), and it can be seen from the graph that they fit well. The error is within about 5%. The accuracy of the error is expected to improve as the number of data increases.

  The pixel area of interest in FIGS. 16 and 17 is a portion where the thickness of the head phantom is large. Even if the position of the phantom is shifted, the thickness of the phantom in the X-ray transmission path hardly changes. Therefore, both X and Y have a small change amount. Therefore, the luminance value due to the scattered radiation hardly changes.

  FIG. 14 (b) is a curve approximation in the case where the tooth row is just located at the overlapping portion and the periphery thereof, and when the overlapping position is reached, the ratio of the scattered radiation becomes extremely high. However, even in this case, the data group is distributed along the curve.

  (B) of FIG. 19 to FIG. 22 show a curve approximation of a pixel region away from the vicinity of the tooth row. Also in these, all the data groups are distributed along the curve, and the error has a better value than in the vicinity of the dentition. Since the change in the luminance value along the pixel line is more gradual than in the vicinity of the dentition, the change in the ratio of the scattered radiation is considered to be small.

  (B) of FIG. 23 to FIG. 25 show a curve approximation of a region where the thickness of the head phantom is very small at the end of the detector. As for these, all data groups are distributed along the curve.

<6. Handling of parts with large X-ray attenuation, such as overlapping parts of molars>
In a pixel on a detector corresponding to a portion where X-ray attenuation is large, such as a dentition portion, particularly a portion where teeth overlap, a luminance value due to scattered radiation is smaller than a case where X-ray attenuation is small. Then, when applied to arbitrary biometric data, in an actual projection image, when the tooth overlap portion is located between 25 pixel regions, etc., the above <1-3. Calculation of luminance value of scattered radiation in all pixels on the detector> In the profile based on the quartic curve obtained in the above, the luminance value due to the scattered radiation may be calculated as a value larger than the actual value. Then, the luminance value of the tooth row portion after the correction of the scattered radiation is reduced, which has a considerable effect on the CT reconstructed image. Therefore, it is necessary to correct the luminance value of the scattered radiation. An example of this correction method will be described below.

  FIG. 26 shows a projected image of one frame and 25 regions. Corresponding to the dentition is between points 7 and 12, between points 8 and 13, and between points 9 and 14. FIG. 27 shows the profile of the luminance value of all radiation and the luminance value of scattered radiation calculated from the quartic curve on the pixel line passing through points 2, 7, 7, 12, and 22 in FIG. It is.

  From FIG. 27, it can be seen that there is a pixel whose luminance values are almost equal between pixel Nos. 200 to 300 (this portion corresponds to between points 7 and 12). This means that no radiation is directly detected and no X-rays are transmitted, and this does not occur unless there is a substance such as a metal having a large X-ray attenuation coefficient, which is contradictory.

  Therefore, in order to strictly correct the scattered radiation in this portion, the above-mentioned <1-3. Calculation of luminance values of scattered radiation in all pixels on the detector>, a quaternary curve is calculated for a total of 10 lines each of 5 lines in the vertical and horizontal directions, and then the profile of the luminance value between the above two points It is necessary to perform the calculation of each in addition, and to correct the luminance value of the scattered radiation.

  It is possible to conduct an experiment using a lead disk also in the pixel between these two points. However, in the region indicated by the arrow, the parameter of the above equation (3) does not change so much, so this is utilized. For example, for both points 7 and 12 in frame No. 40, a graph showing the ratio of the luminance value of all the radiation to the white image on the horizontal axis and the ratio of the luminance value of the scattered radiation to the total radiation on the vertical axis (FIG. 14 to 25) are shown in FIG. These are all distributed along one curve, and it can be seen that a common value can be applied to the parameter of equation (3). The profile of the luminance value due to the scattered radiation in FIG. 27 is corrected using the value calculated by this curve.

  Assuming that the relationship between the scattered radiation and the luminance value of the total radiation between points 7 and 12 is represented by this curve, three points are obtained by dividing the two points into four, and the scattered radiation is determined for each of the three points. Is calculated. Then, it is assumed that the calculated luminance value is correct.

  On the other hand, the above <1-3. Calculation of luminance value of scattered radiation in all pixels on the detector> Since the luminance value due to scattered radiation has already been calculated from the quartic curve calculated based on The profile of the brightness value difference obtained from the points is corrected by subtracting the profile of the brightness value due to the scattered radiation calculated from the quartic curve.

First, the magnitude relationship between the luminance value differences of scattered radiation at three points is
(1) It becomes smaller in the order of pixel numbers,
(2) increases in order of pixel number,
(3) the middle pixel is the largest,
(4) the middle pixel is minimized,
Can be roughly divided into four patterns. In order to make the profile of the difference in the luminance value with respect to the pixel number smooth, a Sin curve is adopted. This is shown in FIGS. 29 (a) to 29 (d). The horizontal axis is the pixel number and the vertical axis is the difference between the luminance values of the scattered radiation calculated by the above two methods. The pixel number 0 in the figure corresponds to the point 7, and the pixel number 165 corresponds to the point 12. Since the luminance value difference between these two pixels is 0 and both ends of the profile gradually increase from 0, the phase of the Sin curves at the points 7 and 12 is -π / 2 + 2nπ (n; an integer). The areas obtained by dividing the points 7 and 12 into four equal parts are sequentially referred to as an area I, an area II, an area III, and an area IV (see FIG. 29A).

  As for the pattern (1), since the luminance value difference reaches the maximum value from 0 in the region I, the phase of the Sin curve is changed from -π / 2 to + π / 2 with half as the amplitude. In the regions II to IV, the parameters of the phase constant are adjusted so that the phase changes from + π / 2 + 2nπ to (3/2) π + 2nπ in each region. The amplitude is one half of the luminance value difference of each area.

  Regarding the pattern (2), the profile for the pixel is reversed left and right with respect to the pattern (1), but can be modified in the same manner as above.

  For the pattern (3), parameters are determined so that the phase changes from -π / 2 + 2nπ to π / 2 + 2nπ in the regions I and II, and from π / 2 + 2nπ to (3/2) π + 2nπ in the regions III and IV.

  For the pattern (4), -π / 2 to π / 2 in the region I, π / 2 to (3/2) π in the region II, (3/2) π to (5/2) π in the region III, In the region IV, the phase is changed from (5/2) π to (7/2) π, and similarly corrected.

  It is determined which of the above-mentioned patterns corresponds from the magnitude relation of the difference of the three brightness values, the profile of the brightness value difference is calculated, and the brightness value due to the scattered radiation already calculated from the quartic curve is calculated. By subtracting the luminance value difference, it is possible to calculate a luminance value of scattered radiation that is closer to ideal.

  As described above, as a result of adding processing to the calculation of the luminance value of the scattered radiation, the profile of FIG. 27 is as shown in FIG. As a result, it is possible to avoid that the luminance value after the scattered radiation correction in the tooth row portion becomes extremely small, and that the profile of the scattered radiation also changes smoothly.

<7. Effect of scattered radiation correction when living body is used as subject>
FIG. 31A shows an axial image picked up at one point when a measured image of a living body as a subject is reconstructed without performing scattered radiation correction.

  FIG. 31B shows a case where an axial image is picked up at one point in the case of reconstructing an image after applying scattered radiation correction proposed in Japanese Patent Application Laid-Open No. 2015-65993 to a measurement image of a living body as a subject. It is shown.

  FIG. 31C shows the above-mentioned <1. Calculation of Scattered Radiation Captured by Each Pixel of Detector> Axial image picked up and shown when the image is reconstructed after performing scattered radiation correction using the luminance value of scattered radiation calculated by the method described in <1. It is a thing.

  In FIG. 31 (b), artifacts appear in the direction along the inside of the dentition and between the teeth and between the molars, and there is not much change compared to FIG. 31 (a), and the image quality is not much improved. At 31 (c), the artifact is eliminated. That is, <1. Calculation of Scattered Radiation Captured by Each Pixel of Detector> It is understood that the scattered radiation correction using the luminance value of the scattered radiation calculated by the method described above is effective even when a living body is used as a subject.

<8. Scatter correction device>
FIG. 32 is a diagram showing a configuration of the scattered radiation correction device 100 according to one embodiment of the present invention.

  The scattered radiation correction apparatus 100 includes a CPU 11 that controls the entire scattered radiation correction apparatus 100 according to programs stored in the ROM 12 and the HDD 17, a ROM 12 that records fixed programs and data, a RAM 13 that provides a working memory, A communication interface unit 14 for performing communication with the VRAM 15 that temporarily stores image data, a display unit 16 that displays an image based on the image data stored in the VRAM 15, an HDD 17 that will be described in detail below, An operation unit 18 such as a keyboard and a pointing device is provided.

  The communication method between the communication interface unit 14 of the scattered radiation correction apparatus 100 and the outside may be wired communication, wireless communication, or communication combining wired and wireless. As the scattered radiation correction device 100, for example, a personal computer can be used.

  The HDD 17 stores various programs such as an image reconstruction processing program, a scattered radiation correction program, CT imaging data (measurement image) by the dental X-ray imaging apparatus, and a reconstruction volume generated by the image reconstruction processing program. Data and various data such as set values of various parameters used when executing various programs are stored.

  Further, the scattered radiation correction apparatus 100 may acquire information from the detector to generate a measurement image, store the generated measurement image in the HDD 17, and acquire the measurement image from outside using the communication interface unit 14. Then, the acquired measurement image may be stored in the HDD 17. An external device may perform CT reconstruction. In this case, unlike the present embodiment, it is possible to adopt a configuration in which the HDD 17 does not store the image reconstruction processing program, and the scattered radiation correction apparatus 100 outputs the measured image after the scattered radiation correction to an external What is necessary is just to transmit to a device.

  The image reconstruction processing program is a program for reconstructing CT imaging data by a dental X-ray imaging apparatus and generating reconstructed volume data. The scattered radiation correction program is a program for performing scattered radiation correction. Note that the scattered radiation correction program includes a scattered radiation intensity calculation program for calculating the scattered radiation intensity.

  Each program stored in the HDD 17 may be preinstalled in the scattered radiation correction device 100, or may be distributed in a form stored in a storage medium such as an optical disk and installed in the scattered radiation correction device 100. It may be distributed via a network and installed in the scattered radiation correction apparatus 100. A part of each program and each data stored in the HDD 17 may be stored in the ROM 12 instead of the HDD 17.

  The HDD 17 transmits scattered radiation with respect to the total radiation intensity in a specific frame (for example, the 510 frames described above) at a specific pixel position (for example, 25 black circles shown in FIG. 4) on a detector used in X-ray imaging of a subject. The ratio of the calculation result of the intensity (for example, the extrapolated data described above) and the X-ray of the subject in the specific frame with respect to the total radiation intensity detected at the specific pixel position in X-ray imaging without the subject in the specific frame A curve-type parameter (for example, the above-described parameter a and parameter b) indicating the relationship with the ratio of the total radiation intensity detected at the specific pixel position in imaging is stored. The scattered radiation correction program stored in the HDD 17 is executed by the CPU 11 to cause the hardware of the scattered radiation correction device 100 to detect the total radiation detected by X-ray imaging of the subject in a specific frame at a specific pixel position. The calculation unit calculates the intensity or the ratio of the scattered radiation intensity to the direct radiation intensity, and functions as a correction unit that performs scattered radiation correction on the measured image based on the calculation result of the calculation unit. Further, the HDD 17 may store, for example, the above-described equation (2) or the interpolation data calculated from the equation (2). Note that the scattered radiation intensity calculation program stored in the HDD 17 causes the hardware of the scattered radiation correction device 100 to function as the above-described calculation unit by being executed by the CPU 11. That is, when only the scattered radiation intensity calculation program among the scattered radiation correction programs is executed, the scattered radiation correction device 100 is a scattered radiation intensity calculation device that calculates the scattered radiation intensity.

  As described above, one embodiment of the present invention has been described. However, the scope of the present invention is not limited to this, and various modifications can be made without departing from the gist of the invention.

  For example, instead of a lead disk, a lead plate material having a shape other than the circular shape (for example, a rectangular shape) may be used. Further, an X-ray absorbing material other than lead may be used. Further, instead of obtaining the extrapolated data of the scattered radiation as in the above-described embodiment, the direct radiation at a specific pixel position may be obtained using an X-ray absorber having a minute through-hole.

  For example, in the embodiment described above, the head phantom does not include a metal such as an implant, but the head phantom may include a metal such as an implant.

100 scattered radiation correction device 11 CPU
12 ROM
13 RAM
14 communication interface unit 15 VRAM
16 Display 17 HDD
18 Operation unit

Claims (7)

A first ratio, which is a ratio of a calculation result of the scattered radiation intensity to a total radiation intensity in a specific frame at a specific pixel position on a detector used in X-ray imaging of the subject, and the subject in the specific frame are provided. A second ratio that is a ratio of the total radiation intensity detected at the specific pixel position in the X-ray image of the subject in the specific frame to the total radiation intensity detected at the specific pixel position in the non-radiographic image . A storage unit that stores a parameter of a curve expression indicating a relationship with the ratio ,
In the specific frame at the specific pixel position , a third ratio, which is a ratio of scattered radiation intensity to total radiation intensity or direct radiation intensity detected by X-ray imaging of the subject , is calculated using the curve equation. A calculating unit;
Equipped with a,
The scattered radiation intensity calculation device , wherein the first ratio is calculated using extrapolation data of a luminance value of scattered radiation in the specific frame at the specific pixel position . A first ratio, which is a ratio of a calculation result of the scattered radiation intensity to a total radiation intensity in a specific frame at a specific pixel position on a detector used in X-ray imaging of the subject, and the subject in the specific frame are provided. A second ratio that is a ratio of the total radiation intensity detected at the specific pixel position in the X-ray image of the subject in the specific frame to the total radiation intensity detected at the specific pixel position in the non-radiographic image . A storage unit that stores a parameter of a curve expression indicating a relationship with the ratio ,
In the specific frame at the specific pixel position , a third ratio, which is a ratio of scattered radiation intensity to total radiation intensity or direct radiation intensity detected by X-ray imaging of the subject , is calculated using the curve equation. A calculating unit;
Equipped with a,
The curve equation is represented by Y = aX ^b (where Y is the first ratio, X is the second ratio, and a and b are each the parameter), Scattered ray intensity calculator. The scattered radiation intensity calculation device according to claim 1 or 2 , wherein a plurality of the specific pixel positions are set and a plurality of the specific frames are set. The storage unit, the scattered radiation intensity at the pixel position other than the specific on the detector, from the interpolation formula or the interpolation formula to interpolate based on the calculation result of the scattered radiation intensity in the specific frame at the specific pixel position The scattered radiation intensity calculation device according to claim 1, wherein the calculated interpolation data is stored. A scattered radiation intensity calculation device according to any one of claims 1 to 4 ,
A correction unit that performs scattered radiation correction on a measurement image obtained by X-ray imaging of the subject based on the calculation result of the calculation unit;
A scattered radiation correction device comprising: A first ratio, which is a ratio of a calculation result of the scattered radiation intensity to a total radiation intensity in a specific frame at a specific pixel position on a detector used in X-ray imaging of the subject, and the subject in the specific frame are provided. A second ratio that is a ratio of the total radiation intensity detected at the specific pixel position in the X-ray image of the subject in the specific frame to the total radiation intensity detected at the specific pixel position in the non-radiographic image . A reading step of reading from a storage medium a parameter of a curve expression indicating a relationship with the ratio ;
In the specific frame at the specific pixel position , a third ratio that is a ratio of the scattered radiation intensity to the total radiation intensity or the direct radiation intensity detected in the X-ray imaging of the subject is read in the reading step. A calculation step of calculating using the curve equation determined by the parameter,
Equipped with a,
The scattered radiation intensity calculation method , wherein the first ratio is calculated using extrapolation data of a luminance value of scattered radiation in the specific frame at the specific pixel position . A first ratio, which is a ratio of a calculation result of the scattered radiation intensity to a total radiation intensity in a specific frame at a specific pixel position on a detector used in X-ray imaging of the subject, and the subject in the specific frame are provided. A second ratio that is a ratio of the total radiation intensity detected at the specific pixel position in the X-ray image of the subject in the specific frame to the total radiation intensity detected at the specific pixel position in the non-radiographic image . A reading step of reading from a storage medium a parameter of a curve expression indicating a relationship with the ratio ;
In the specific frame at the specific pixel position , a third ratio that is a ratio of the scattered radiation intensity to the total radiation intensity or the direct radiation intensity detected in the X-ray imaging of the subject is read in the reading step. A calculation step of calculating using the curve equation determined by the parameter,
Equipped with a,
The curve equation is represented by Y = aX ^b (where Y is the first ratio, X is the second ratio, and a and b are each the parameter), Scattered ray intensity calculation method.