JP4047420B2 - Image reconstruction device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image reconstruction device that reconstructs a distribution image corresponding to an X-ray absorption coefficient inside a subject based on a plurality of projection data from different angles.
[0002]
[Prior art]
As a method for reconstructing the distribution image of the X-ray absorption coefficient of the subject from the projection data obtained by measuring the dose distribution of the radiation transmitted through the subject in a plurality of directions, refer to Reference Document ("Medical Imaging Engineering Handbook" Some are introduced in the Society supervised by the Japanese Society for Medical Image Engineering, edited by the Medical Image Engineering Handbook Editorial Committee, Shinohara Publishing, I. Basic Logic, Chapter 5, Image Reconstruction). Among them, a filtered back projection method (Filtered Backprojection) which is one of direct (analytic) reconstruction methods is well known.
[0003]
The image reconstruction method applied in the current CT (Computer Tomography) apparatus is different in that some approximations are made to facilitate hardware implementation and a spatial filter is added to the correction filter. Basically, such a method is adopted.
[0004]
This image reconstruction method assumes that projection data is collected at equiangular intervals, ie, each with equal pitch. However, even in the most major CT apparatus as an apparatus for image reconstruction using X-rays, the rotational speed of the X-ray tube or the irradiation angle of the X-rays has a minute blur caused by the weight of the X-ray tube. In reality, there are cases where projection data is collected at unequal angular intervals. Furthermore, when the three-dimensional CT that has become a hot topic in recent years is realized by a circulatory X-ray imaging apparatus having a rotating arm mechanism, such an effect becomes remarkable. More specifically, the rotation speed changes dynamically (dynamically) during the period from the start of rotation to the acceleration and from the deceleration to the end of rotation. However, the X-ray exposure interval is always an equal time interval. For this reason, the angular pitch of projection data collection changes dynamically during this time.
[0005]
Moreover, even if the rotation speed is almost constant, the projection data collection is not possible due to the direct influence on the rotation speed due to the weight of the X-ray tube or the detector or the indirect influence due to the deflection of the rotation arm. It is also conceivable that the angle pitch changes.
[0006]
Therefore, there is a problem that the reconstructed image obtained by the reconstruction processing by the filter-corrected back projection method is significantly deteriorated due to the non-uniformity of the angle pitch in the projection data collection.
[0007]
Reference literature (LAFeldkamp et al., “Practical cone-beam algorithm”, J. Opt. Soc. Am. A, No. 6, 612-619 (1984)) or the above-mentioned image reconstruction method for 3D CT (Feldkamp inverse filter correction described in Satoshi Oishi et al., “Three-dimensional reconstruction using cone beam projection system”, Med. Imag. Tech., Vol. 8, No. 2, 127-130 (1990)) There is a projection method. This method was proposed by Feldkamp et al. And is considered one of the most famous and practical methods in the three-dimensional reconstruction method. After convolution filtering is applied to the projection data, cone beam spread correction is performed. While performing, a three-dimensional backprojection operation is performed.
[0008]
However, even with this Feldkamp filter-corrected backprojection method, the above-mentioned problems are not solved, and a non-uniform angular pitch in the projection data collection works to degrade the reconstructed image.
[0009]
[Problems to be solved by the invention]
The present invention has been made in consideration of the above-described circumstances, and an object of the present invention is to deteriorate a reconstructed image due to a false image (artifact) generated due to non-uniformity of an angle pitch at the time of projection data collection. Is to provide an image reconstruction apparatus capable of effectively suppressing the processing time without significantly increasing the processing time.
[0010]
[Means for Solving the Problems]
  In order to solve the above problems and achieve the object, the present invention comprises the following means.
(1) The image reconstruction apparatus of the present invention includes a collection unit that collects a plurality of projection data at predetermined projection angle intervals by irradiating a subject with X-rays from multiple directions, and an X-ray inside the subject. Reconstructing means for reconstructing an image representing an absorption coefficient distribution based on the projection data;By multiplying each backprojection data by a correction coefficient for correcting the power difference according to the nonuniformity of the projection angle interval in the backprojection calculation performed in the reconstruction means.Suppression means for suppressing deterioration of the reconstructed image.
For this reason, the back projection calculation can be performed by correcting the difference in power according to the non-uniformity of the projection angle interval.
(2The image reconstruction apparatus of the present invention is the above (1The correction coefficient is calculated based on a ratio of a projection angle interval at a predetermined projection angle to a reference projection angle interval.
  For this reason, the correction coefficient for each projection angle can be calculated.
[0011]
For this reason, it is possible to suppress the degradation of the reconstructed image caused by the unevenness of the projection angle interval.
(2) The image reconstruction apparatus according to the present invention is the apparatus according to (1) above, and the suppression unit performs non-uniformity of the projection angle interval during back projection calculation performed in the reconstruction unit. And means for multiplying each backprojection data by a correction coefficient for correcting a power difference according to.
[0012]
For this reason, the back projection calculation can be performed by correcting the difference in power according to the non-uniformity of the projection angle interval.
(3) The image reconstruction device of the present invention is the device according to (2) above, and the correction coefficient is based on a ratio of a projection angle interval at a predetermined projection angle to a reference projection angle interval. It is calculated.
[0013]
For this reason, the correction coefficient for each projection angle can be calculated.
(4) The image reconstruction device according to the present invention is the device according to (1) above, and the suppression unit interpolates projection data at unequal angular intervals collected by the collection unit. Interpolation means for creating projection data at equiangular intervals is provided, and the reconstruction means reconstructs the image based on the projection data created by the interpolation means.
For this reason, the non-uniformity of the projection angle interval can be improved before the reconstruction calculation.
[0014]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of an image reconstruction device of the present invention will be described with reference to the drawings.
(First embodiment)
1 and 2 are an external view and a block diagram showing the configuration of the circulatory X-ray diagnostic system according to the first embodiment of the image reconstruction apparatus of the present invention. X-rays irradiated in a cone beam shape from the X-ray tube 1 shown in FIG. 1 pass through a subject (not shown) and enter the two-dimensional detector 2. The two-dimensional detector 2 detects the intensity distribution of transmitted X-rays that have passed through the subject. The X-ray tube 1 and the two-dimensional detector 2 are connected to a C-shaped rotary arm 3 and X-rays are irradiated (or continuously exposed) a plurality of times while rotating the rotary arm 3. Thus, the projection data from many directions around the subject is collected.
[0015]
The holder 3a that supports the rotary arm 3 so as to be slidable is provided with a drive mechanism 3b (not shown), whereby the rotary arm 3 is slid and rotated in the direction of the arrow R1. The holder 3a is rotatably supported with respect to the ceiling, and is driven to rotate in the direction of arrow R2 by a drive mechanism 3c (not shown). These drive mechanisms 3b and 3c are controlled by a separately provided control device (not shown).
[0016]
As shown in FIG. 2A, the X-ray diagnostic system of this embodiment includes a data acquisition unit 4, an A / D converter 5, a projection data memory 6, a reconstruction unit 7, an image memory 8, and a D / A. A converter 9 and a monitor 10 are provided. The data collection unit 4 includes the two-dimensional detector 2 of the rotary arm 3. 2B is a block diagram showing the internal configuration of the reconstruction unit 7. The reconstruction unit 7 includes a distortion correction unit 71, a logarithmic (log) conversion unit 72, a filtering unit 73, and a coefficient correction unit. 74 and a back projection unit 75.
[0017]
FIG. 3 is a diagram illustrating a state in which the projection data collection angle interval changes with the rotation speed of the arm 3.
Strictly speaking, the rotation speed of the arm 3 is not constant, and therefore the X-ray irradiation angle interval is not constant. Therefore, the projection data collection angle interval dynamically changes between the start and end of the slide rotation of the arm 3. For example, at 0 to 15 degrees, the rotation speed of the arm 3 is relatively slow, and the collection angle interval at this time is Δθ₁ In the range of 145 to 160 degrees, the rotation speed of the arm 3 is relatively fast, and the collection angle interval at this time is Δθ₂   It is. For the sake of understanding, FIG.₁   And Δθ₂   The difference between the two is exaggerated and actually Δθ₁   And Δθ₂   The difference is small.
[0018]
When collecting projection data, the control device outputs a control signal to the drive mechanism 3b so that the rotary arm 3 is slid and rotated. The two-dimensional detector 2 periodically captures and outputs an X-ray image at a constant time interval during the slide rotation.
[0019]
Consider a case in which projection data for 180 degrees corresponding to projection of 120 patterns on the two-dimensional detector 2, for example, is collected by X-ray exposure by the rotary arm mechanism as described above. Actually, it is not possible to obtain a complete reconstructed image only by projection data for 180 degrees. However, in such a device having a rotating arm as a rotating mechanism, it is often difficult to rotate 180 degrees or more. On the other hand, the goal of such reconstruction processing is to observe and measure the shape of a blood vessel or the like, and not to accurately determine the internal CT value distribution. Therefore, the occurrence of artifacts (false images) due to the influence of such insufficient projection data can be ignored by threshold processing, for example.
[0020]
The projection data for 120 patterns detected by the two-dimensional detector 2 is converted into an analog signal by a TV system including an image intensifier, sent to the data collection unit 4, and then digitally converted by the A / D converter 5. It is converted into a signal and stored in the projection data memory 6.
[0021]
The reconstruction unit 7 reads the projection data stored in the projection data memory 6 and performs reconstruction processing. Here, it is assumed that image reconstruction is performed by the filter-corrected back projection method. Note that the reconstruction area is defined as a cylinder or a sphere inscribed in the X-ray flux in all projection directions.
[0022]
The reconstruction unit 7 is a discrete element having a length d at the center of the reconstruction area projected onto the width of one detection element of the two-dimensional detector 2 in a plane perpendicular to the rotation axis of the X-ray tube 1. Then, a reconstructed image represented by a set of data of discrete points obtained is obtained. Note that the discrete intervals are not limited to a specific length d, and differ depending on the device or manufacturer. Therefore, in principle, discrete intervals defined for each device are used. Further, considering the width (in the rotation axis direction) of one element of the two-dimensional detector, the same discretization as in the tomographic image is performed in the rotation axis direction. Thus, the reconstruction area is discretized into a three-dimensional grid. Each element of the data discretized in a lattice form is generally called a voxel.
[0023]
Here, the reconstruction process in the reconstruction unit 7 will be described in more detail. The reconstruction unit 7 convolves the 120-frame projection data obtained from the two-dimensional detector 2 by applying an appropriate filter such as Shepp & Logan or Ramachandran for the filtered back projection method. Reconstruction data is obtained by performing a back projection operation on the convolution result. More specifically, distortion correction of the image intensifier by the distortion correction unit 71 and logarithmic conversion of the projection data by the logarithmic (log) conversion unit 72 are performed, and the filtering unit 73 applies the filter correction back projection method to the projection data. Convolution is performed using such a correction filter, and the coefficient correction unit 74 multiplies the convolution result by a correction coefficient for power correction described later. The back projection unit 75 reprojects the projection data multiplied by the correction coefficient to reconstruct an image. If the TV system does not include an image intensifier, the distortion correction unit 71 is unnecessary.
[0024]
In particular, the reconstruction unit 7 of the present embodiment suppresses degradation of a reconstructed image caused by nonuniformity in the collection angle interval of projection data when performing a reconstruction operation on a discretized reconstruction area. It has.
[0025]
The suppression means performs a back projection by applying a correction coefficient for correcting a power difference due to the nonuniformity of the projection angle interval after the projection data is subjected to the convolution filter according to the filter-corrected back projection method. . As a result, artifacts due to power differences can be reduced, and deterioration of the reconstructed image can be prevented.
[0026]
Depending on the projection angle, the power of backprojected data varies. For example, to simplify the description, as shown in FIG.₁   = Δθ (Δθ 0) interval projection data and Δθ₂   Consider a case in which projection data with an interval of (= Δθ / 2) coexist. If all 360 degrees are data collected at Δθ intervals, no artifacts will occur. However, in the vicinity of an arbitrary direction N · Δθ (N is an integer) collected at intervals of Δθ / 2, in addition to data (N + 1) · Δθ in the next direction when originally collected at intervals of Δθ, (N + 1 / 2) · Δθ exists. Since Δθ 0 here, the data of (N + 1/2) · Δθ is considered to be substantially equal to the data of N · Δθ. From this, back projection of N · Δθ and (N + 1/2) · Δθ data is equivalent to double projection of N · Δθ projection data, in other words, at the projection angle N · Δθ. Data twice as large as the projection data is collected. In this way, the projection directions collected at unequal angular intervals are equivalent to the fact that projection data having different powers are collected according to the projection angular intervals, and thus artifacts occur.
[0027]
Therefore, in this embodiment, in order to correct the change in power due to the difference in the projection angle interval dθ during the backprojection calculation, the backprojection calculation is performed after multiplying the backprojection data by the following correction coefficient. Yes.
[0028]
W (dθ) = dθ / dθ₀(Dθ₀Is a reference projection angle interval) (1)
By applying the above equation (1), the projection angle interval becomes Δθ.₀The projection data collected at / 2 is multiplied by a correction coefficient corresponding to the ratio to the reference projection angle, that is, 1/2, in the back projection calculation. Therefore, the doubled power due to the non-uniform projection collection angle can be corrected to an appropriate value.
[0029]
The projection angle interval dθ may be obtained by measurement, for example, by detecting the collection angle from the slide rotation amount of the rotary arm 3 every time projection data is collected, or each projection based on the collection angle measured in advance. Values between data may be stored as fixed values in a table or the like and used as known values.
[0030]
Next, the reconstruction data obtained by the back projection operation is normalized with a contrast scale. In general, reconstruction data obtained by backprojection is divided by the number N of projection directions (120 as an example). In this embodiment, in particular, N represented by the following equation (2) is used._dDivide by.
[0031]
N_d= Θ_a/ Dθ₀  ... (2)
Where θ_aIs an angle at which projection data is collected, and is 180 degrees in this embodiment, for example.
[0032]
An image obtained by image reconstruction processing in the reconstruction unit 7, that is, a tomographic image of the subject is sent to the image memory 8 and stored therein. This tomographic image can be displayed on the monitor 10, or a blood vessel region is extracted by threshold processing in another image processing apparatus (not shown), and shading processing for shading the blood vessel surface image is performed. The surface image of the blood vessel observed from an arbitrary direction can be displayed.
[0033]
As described above, according to the present embodiment, the reconstruction process related to the filter-corrected back projection method is performed while correcting the power difference according to the non-uniformity of the projection angle interval. For example, the algebraic reconstruction method Compared with a successive approximation method such as (ART: algebraic reconstruction technique), the degradation of the reconstructed image can be effectively suppressed in a short time without significantly increasing the processing time.
[0034]
(Second Embodiment)
Next, a second embodiment of the present invention will be described. The second embodiment relates to a modification of the first embodiment described above, and implements an imaging technique called digital subtraction angiography (DSA) in the circulatory organ X-ray diagnostic system. FIG. 4 is a block diagram showing the configuration of the present embodiment.
[0035]
In the present embodiment, X-ray imaging is performed while injecting a contrast medium into the subject. For example, when reconstructing an image representing only a blood vessel, the projection data is collected twice before and after the injection of the contrast agent, and the first (mask image) projection data memory 111, the second ( Live image) projection data memory 112. When the required data is accumulated in both memories, the subtraction processing unit 12 performs a subtraction (difference) process. That is, the mask image projection data and the live image projection data are each log (logarithmically) converted, and then the log conversion results are subtracted.
[0036]
The subtraction processing result is sent to the reconstruction unit 7. The image reconstruction process in the reconstruction unit 7 is substantially the same as that in the first embodiment, but is slightly different. That is, since the log conversion has already been performed in the subtraction processing unit 12, the reconstruction unit 7 of this embodiment does not include the log conversion unit 72.
[0037]
Then, as in the first embodiment, this embodiment suppresses degradation of the reconstructed image caused by nonuniformity in the collection angle interval of the projection data when performing the reconstruction calculation of the discretized reconstruction area. And a convolution unit that convolves the projection data with the correction filter according to the filter-corrected backprojection method by the filtering unit 73, and a coefficient correction unit 74 that determines the collection angle interval of the projection data with respect to the convolution result. Multiply the correction coefficient to correct the power difference due to non-uniformity. The back projection unit 75 reprojects the projection data multiplied by the correction coefficient to reconstruct an image.
[0038]
Therefore, according to the second embodiment, an image (subtraction image) from which only blood vessels are extracted can be obtained, artifacts due to power differences can be reduced, and deterioration of the reconstructed image can be prevented.
[0039]
(Third embodiment)
Next, a third embodiment of the present invention will be described. In the first and second embodiments, in order to correct the change in power due to the difference in the projection angle interval with respect to the convolution result in order to suppress the deterioration of the reconstructed image due to the nonuniformity of the projection angle interval. The configuration of the reconstruction processing in which the projection is performed by multiplying the correction coefficient is described. On the other hand, the image reconstruction apparatus of the third embodiment obtains projection data with a uniform projection angle interval by interpolation calculation, and convolves the projection data obtained by this interpolation and backprojects. About another structure, it is the same as that of the 1st Embodiment, The detailed description is abbreviate | omitted.
[0040]
FIG. 6 is a block diagram showing a schematic configuration of the third embodiment of the present invention. As shown in FIG. 6A, the configuration is different from that of the first embodiment in that an interpolation processing unit 20 is provided between the projection data 20 and the reconstruction unit 21. A specific configuration and operation of the interpolation processing unit 20 will be described later. Note that the reconstruction unit 21 of the present embodiment does not include the coefficient correction unit 74 of the reconstruction unit 7 according to the first embodiment, as illustrated in FIG. 6B.
[0041]
The backprojection filter used in the filter-corrected backprojection method is a high-frequency enhancement filter, and considering the case of reconstructing Dirac's δ function, all directions of the high-frequency enhancement filter except for the position where the δ function exists The back projection from cancels out the positive and negative values to zero. For example, a positive value is back-projected from one direction, and a negative value is back-projected from another direction.
[0042]
However, in the back projection with respect to projection data with unequal angular intervals, the positive / negative balance added during back projection is lost, resulting in artifacts. That is, the balance on which the back projection filter is applied is lost.
[0043]
Therefore, in the present embodiment, projection data at equiangular intervals is created by interpolation calculation from projection data collected at unequal angular intervals in order to optimize the balance to which the back projection filter is applied.
[0044]
In particular, in the present embodiment, a case where the interpolation processing unit 20 performs primary interpolation will be described. Where θ = θ₀The data of the Nth frame is collected at the time of θ = θ₁Assume that the data of the (N + 1) th frame is collected at the time of. And now θ = θ₂(Θ₀≦ θ₂≦ θ₁), When the projection data at the point of time is created by interpolation, θ = θ₂Projection data p (x, y, θ₂) Is calculated according to the following equation (3).
[0045]
p (x, y, θ₂) = M × p (x, y, θ₀) + N × p (x, y, θ₁) ... (3)
Here, (x, y) represents a coordinate system indicating an arbitrary pixel on the image, and m and n are given by the following equations (4) and (5).
[0046]
m = (θ₁−θ₂) / (Θ₁−θ₀(4)
n = (θ₂−θ₀) / (Θ₁−θ₀(5)
The projection data of equiangular intervals newly created by such interpolation is sent to the reconstruction unit 21. In this embodiment, an example in which a linear interpolation function is used as an interpolation function has been introduced. However, the present invention is not limited to only linear interpolation. For example, a high-order function such as a cubic B-Spline function is used. Also good.
[0047]
In the reconstruction process in the reconstruction unit 21, the reconstructed discretized reconstruction area is reconstructed. The reconstruction process here is the same as the reconstruction process of the first embodiment, except that the process does not perform the coefficient correction process of the first embodiment.
[0048]
As described above, in the third embodiment, projection data at equiangular intervals is created from the projection data collected at unequal angular intervals by interpolation, and the balance to which the back projection filter is applied is optimized. Can do. For this reason, the non-uniformity of the projection angle interval can be improved before the reconstruction calculation. Therefore, the degradation of the reconstructed image can be effectively suppressed in a short time as in the first embodiment.
[0049]
(Fourth embodiment)
Next, a fourth embodiment of the present invention will be described. The fourth embodiment relates to a modification of the above-described third embodiment and performs the above-described DSA. FIG. 7 is a block diagram showing the configuration of this embodiment.
[0050]
In the present embodiment, X-ray imaging is performed while injecting a contrast medium into the subject. For example, when reconstructing an image representing only a blood vessel, the projection data is collected twice before and after the injection of the contrast agent, and the first (mask image) projection data memory 111, the second ( Live image) projection data memory 112. When the required data is accumulated in both memories, the subtraction processing unit 12 performs a subtraction (difference) process. That is, the mask image projection data and the live image projection data are each log (logarithmically) converted, and then the log conversion results are subtracted.
[0051]
The subtraction processing result is sent to the interpolation processing unit 20. The contents of the interpolation processing in the interpolation processing unit 20 and the image reconstruction processing in the reconstruction unit 21 are almost the same as those in the third embodiment, but are slightly different. That is, since log conversion has already been performed in the subtraction processing unit 12, the reconstruction unit 7 of this embodiment does not include the log conversion unit 72 of the third embodiment.
[0052]
Then, as in the third embodiment, the interpolation processing unit 20 interpolates the projection data at equiangular intervals from the projection data collected at unequal angular intervals in order to optimize the balance to which the back projection filter is applied. create. For this reason, the non-uniformity of the projection angle interval can be improved before the reconstruction calculation. Therefore, a subtraction image can be obtained in a short time and effectively suppressing deterioration of the reconstructed image.
[0053]
(Fifth embodiment)
In the above-described first to fourth embodiments, after collecting data using a circulatory X-ray imaging apparatus including a rotating arm (C-arm, etc.) as an imaging system in the three-dimensional CT apparatus currently in the development stage. The case where the image is reconstructed has been described. The two-dimensional CT (which is taken for each cross section of the subject) that is widely used, and the subject is spiraled as shown in FIG. 8 by moving the bed while rotating the X-ray tube around the subject. The present invention can also be implemented for helical scan CT (also referred to as spiral scan CT or spiral scan CT) that scans in the shape of a line.
[0054]
In the case of the former, that is, CT that does not perform helical scanning, the reconstruction area is defined as a circle inscribed in the X-ray bundle in all projection directions, and discretization is performed in the cross section as in the first embodiment. Also in the latter case, the reconstruction space in the cross section perpendicular to the moving direction of the couch top is defined in the same manner as in the two-dimensional CT, and discretization is performed in the same manner. The helical scan CT can basically reconstruct a cross-sectional image at an arbitrary slice position.
[0055]
In any case, similarly to the first to fourth embodiments, it is possible to suppress the degradation of the reconstructed image caused by the unevenness of the projection angle interval.
As a specific example of the reconstruction method, the reference in the helical scan CT (Yusuke Higashiki et al., “Examination of interpolation reconstruction method in helical scan”, Med.Imag.Tech., Vol.8, No.3) 253-254 (1990)), the projection data may be interpolated by the opposed beam interpolation method or 360 degree interpolation method, and reconstruction may be performed by the same method as the fan beam reconstruction method.
[0056]
The present invention is not limited to the above-described embodiment, and can be implemented with various modifications. For example, the present invention is not limited only to the circulatory X-ray diagnostic system shown in FIG. 1, and various circulatory X-ray diagnostic systems capable of collecting X-ray images while rotating around the subject. It is applicable to.
[0057]
【The invention's effect】
According to the present invention, deterioration of a reconstructed image due to a false image (artifact) generated due to non-uniformity of an angular pitch at the time of projection data collection is effectively suppressed without significantly increasing the processing time. An obtained image reconstruction device can be provided.
[0058]
Therefore, the calculation time of the artifacts generated in the filtered back projection method (including the three-dimensional reconstruction method such as Feldkamp method) used in the actual reconstruction device or the experimental device at the practical stage is almost increased. Can be deterred.
[Brief description of the drawings]
FIG. 1 is a diagram showing an external appearance of a rotary arm of a circulatory X-ray imaging system according to a first embodiment of an image reconstruction apparatus of the present invention.
FIG. 2 is a block diagram showing a configuration of the first embodiment including an image reconstruction unit.
FIG. 3 is a diagram showing a state in which the projection data collection angle interval changes according to the first embodiment;
FIG. 4 is a schematic diagram for explaining a difference in power of projection data according to the first embodiment.
FIG. 5 is a block diagram showing a configuration when DSA is performed according to a second embodiment of the image reconstruction apparatus of the present invention.
FIG. 6 is a block diagram showing the configuration of a third embodiment of the image reconstruction device of the present invention.
FIG. 7 is a block diagram showing a configuration when DSA is performed according to a fourth embodiment of the image reconstruction device of the present invention.
FIG. 8 is a view for explaining a helical scan method according to a fifth embodiment of the invention.
[Explanation of symbols]
1 ... X-ray tube
2 ... Two-dimensional detector
3 ... Rotating arm
4. Data collection unit
5 ... A / D converter
6. Projection data memory
7 ... Reconstruction part
8 ... Image memory
9 ... D / A converter
10 ... Monitor
71: Distortion correction unit
72: Log conversion unit
73 ... Filtering section
74: Coefficient correction unit
75 ... Back projection unit

Claims (2)

A collecting means for collecting a plurality of projection data at a predetermined projection angle interval by irradiating the subject with X-rays from multiple directions;
Reconstructing means for reconstructing an image representing the X-ray absorption coefficient distribution inside the subject based on the projection data;
In the backprojection calculation performed in the reconstruction means, the reconstructed image is deteriorated by multiplying each backprojection data by a correction coefficient for correcting the power difference corresponding to the nonuniformity of the projection angle interval. Suppression means for suppressing,
An image reconstruction device comprising: The image reconstruction apparatus according to claim 1 , wherein the correction coefficient is calculated based on a ratio of a projection angle interval at a predetermined projection angle to a reference projection angle interval.

Images