JP5723256B2 - Tomographic image creation method and radiation imaging apparatus - Google Patents

Description

  The present invention relates to a tomographic image creation method and a radiation imaging apparatus.

  A gamma camera and a single-photon emission computed tomography apparatus (hereinafter referred to as a SPECT (Single Photon Emission Computed Tomography) apparatus) using the gamma camera (radiation imaging apparatus) ) In general, a gamma camera measures a distribution of a compound containing a radioisotope and provides a transmission image (planar image), whereas a SPECT apparatus rotates a gamma camera to measure a tomographic image (tomographic image). ).

  Conventional radiation detectors used in gamma cameras have mainly been a combination of a single large crystal scintillator and a plurality of photomultiplier tubes. This gamma camera determines the incident position of radiation by calculating the center of gravity from the output signals of a plurality of photomultiplier tubes. However, this method has a limit of about 10 mm in spatial resolution and is insufficient for clinical use, so a gamma camera with higher spatial resolution is required.

  In recent years, pixel-type radiation detectors (hereinafter referred to as detectors) have been developed as having higher spatial resolution. Pixel-type detectors include those composed of scintillators and photodiodes, and those composed of semiconductors that convert radiation into electrical signals. In either case, the position signal is acquired in small detector units, that is, in pixel units. Therefore, the intrinsic resolution of the detector is determined by the pixel size and performs spatially discrete measurements. In addition, pixel-type detectors having a pixel size of about 1 to 2 mm have been developed, and the spatial resolution has achieved 10 mm or less, which has been greatly improved.

  In the following description of the pixel-type detectors, the terms “detector” and “detector group” are used. The detector means one pixel having an arbitrary shape, and the detector group is a detection. An assembly in which vessels are arranged.

  In addition, a method for reconstructing a tomographic plane in a SPECT apparatus has been developed and improved, which greatly contributes to an improvement in spatial resolution. Up to now, filtered backprojection method (FBP method: filtered back-projection method), successive approximation method without resolution correction (maximum likelihood estimation expectation maximization method (MLEM method: Maximum Likelihood Expectation Maximization method)), subset The expected value maximization method (OSEM method: Ordered Subset Expectation Maximization method) has been used. In recent years, successive approximation methods with resolution correction have been developed. By this method, it is possible to reconstruct an image in consideration of physical factors such as a geometric shape of a collimator and a detector and scattered radiation, and it is possible to provide a more accurate image.

  In general, the detector has a rectangular shape, and when the detector group is viewed from the radiation incident side, the rectangle is densely packed. In order to make the sensitivity uniform in all the detectors constituting the detector group, the through-holes of the collimator and the detector are often arranged in a one-to-one correspondence. From the viewpoint of ease of handling, the shape of the through hole of the collimator is generally rectangular according to the shape of the detector.

  Currently, a SPECT apparatus with high spatial resolution and high sensitivity is in clinical demand. Factors that determine the spatial resolution and sensitivity include many factors such as the distance between the radiation source and the detector, the thickness of the septa, the energy of radiation, scattering, and absorption. Among these factors, the height of the collimator's septa and the size of the collimator's through hole are greatly involved in determining spatial resolution and sensitivity.

  That is, in order to obtain a high spatial resolution, it is necessary to limit the arrival direction of the radiation incident on the detector with a collimator. For this purpose, the field of view in which the detector looks at the measurement object may be narrowed by a collimator. As such a collimator, a LEHR (Low Energy High Resolution) collimator is known. However, this limitation sacrifices sensitivity.

  On the other hand, in order to obtain high sensitivity, it is necessary to increase the size of the through hole of the collimator. As such a collimator, a LEGP (Low Energy General Purpose) collimator and a LEHS (Low Energy High Sensitivity) collimator are known. However, increasing the size of the through hole deteriorates the spatial resolution. Thus, in the conventional SPECT apparatus, since high spatial resolution and high sensitivity are not compatible, it is necessary to replace the collimator according to the application, which is a burden on the clinical site.

  Therefore, a SPECT apparatus in which a plurality of detectors are included in one through hole has been invented as a SPECT apparatus that achieves both sensitivity and spatial resolution. In this SPECT apparatus, when the size of the through hole is the same, it has been demonstrated that a higher spatial resolution can be obtained than a conventional SPECT apparatus in which the through hole and the detector have a one-to-one correspondence (Patent Document 1, Non-Patent Document 1). Patent Document 1).

Special table 2010-507090 gazette Special table 2010-523965 gazette

C. Robert et al. (2008) 2008 IEEE Nuclear Science Symposium Conference Record Vol6 pp.4246-4251

  Due to recent technological innovations in SPECT devices, the spatial resolution of SPECT images has been greatly improved. The main techniques for improving the spatial resolution include the above-described pixel-type semiconductor detector technique and the image reconstruction technique using the successive approximation method. Pixel-type semiconductor detectors greatly improve their inherent resolution itself. In addition, image reconstruction technology has greatly advanced along with recent significant improvements in computer technology (for example, improvement in the number of memories and calculation speed), and in particular, the image quality improvement effect by the successive approximation method with resolution correction. There is a big (improving spatial resolution).

  However, some new problems have emerged due to the improved spatial resolution. The three main issues are described below.

  The first problem is sensitivity. As the spatial resolution is improved, the necessary number of counts is increased, and in order to obtain sufficient image quality in actual clinical application, it is essential to improve sensitivity. In a general collimator, spatial resolution and sensitivity are in a trade-off relationship, and it is difficult to achieve both.

  As a solution to this first problem, a technology has been developed that combines a collimator with a shape in which a plurality of detectors are included in one rectangular through hole and a successive approximation method with resolution correction in a pixelated semiconductor detector. Thus, both high sensitivity and high spatial resolution are realized (see Patent Document 1 and Non-Patent Document 1).

  A second problem is anisotropy of spatial resolution. The collimator is highly distance dependent, and the spatial resolution in the circumferential direction (the direction in which the gamma camera rotates) is high in the vicinity of the surface of the subject, but the space in the radial direction (the direction of the rotation center of the gamma camera) is high. The resolution is low. Due to the anisotropy of the spatial resolution, a circular RI (Radioisotope) distribution near the surface of the subject is distorted into an ellipse and imaged. In particular, in a collimator with a high spatial resolution, image distortion tends to be significant.

  In addition, when imaging the heart of the subject, an image with higher spatial resolution is required, so that the gamma camera is imaged as close as possible to the subject so that the distance from the detector to the heart is as short as possible. For this reason, the gamma camera may perform elliptical trajectory imaging that captures an image while moving along the subject to the elliptical trajectory. In this way, when the gamma camera moves in an elliptical orbit, since the distance from the detector to the heart differs depending on the projection direction, anisotropy of spatial resolution occurs.

  As one solution to this second problem, isotropic properties that reduce anisotropy by performing different filtering processes corresponding to the distance between the detector (gamma camera) and the subject for each projection data. A resolution image reconstruction method is disclosed (see Patent Document 2).

  However, the image reconstruction method disclosed in Patent Document 2 is effective for removing distortion of an image at a peripheral part when the volume is large and the spatial resolution difference between the central part and the peripheral part is large, such as the head. is not.

  A third problem is an artifact caused by resolution correction. When high spatial resolution is used and the count is not sufficient, statistical noise is generated. However, the noise is locally collected by the resolution correction and is imaged as an artifact. Further, when the point response function used for the resolution correction is different from the actually measured value, noise may be generated due to the image reconstruction incorporating the resolution correction.

  Accordingly, an object of the present invention is to provide a tomographic image creation method and a radiation imaging apparatus that reduce anisotropy of spatial resolution and reduce artifacts.

  In order to solve such problems, the present invention provides a detector group having a plurality of detectors that measure radiation, a collimator that is disposed in front of the detector and restricts the incident direction of the radiation, and data processing A tomographic image creating method using a radiation imaging apparatus, wherein the data processing means collects a plurality of projection data from the detector group, and reconstructs a tomographic image from the collected projection data And a correction processing step of performing filtering processing of different cutoff frequencies in each axial direction of the reconstructed tomographic image.

  The present invention also provides a radiation imaging apparatus comprising: a detector group having a plurality of detectors for measuring radiation; a collimator that is disposed in front of the detector to limit the incident direction of the radiation; and data processing means. The data processing means collects a plurality of projection data from the detector group, reconstructs a tomographic image from the collected projection data, and differs in each axial direction of the reconstructed tomographic image. It is characterized in that filtering processing of the cut-off frequency is performed.

  ADVANTAGE OF THE INVENTION According to this invention, the tomographic image creation method and radiation imaging device which can obtain the suitable tomographic image which reduced the distortion by reducing the anisotropy of spatial resolution and reducing an artifact can be provided.

It is a block diagram of the SPECT apparatus which concerns on 1st Embodiment. It is a perspective view which shows the pixel type detector group incorporated in the camera used for the SPECT apparatus which concerns on 1st Embodiment. It is a perspective view which shows another example of a pixel type detector group. It is a perspective view which shows the entrance plane side of the 1st modification of another pixel type detector group. It is a perspective view which shows the opposite surface side of the entrance plane of the 1st modification of another pixel type detector group. It is a perspective view which shows the 2nd modification of another pixel type detector group. It is a perspective view showing a pixel type scintillator detector. It is a perspective view which shows the collimator incorporated in the camera used for the SPECT apparatus which concerns on 1st Embodiment. It is a graph which shows an example of the distance dependence from the collimator of the collimator resolution in the SPECT apparatus which concerns on 1st Embodiment. It is a conceptual diagram which shows the resolution of the tomographic image of the SPECT apparatus which concerns on 1st Embodiment, and a tomographic image. It is a conceptual diagram which shows the tomographic image after the filtering process of the SPECT apparatus which concerns on 1st Embodiment. It is a perspective view which shows the collimator incorporated in the camera used for the SPECT apparatus which concerns on 2nd Embodiment. It is the figure which looked at the arrangement | positioning of one through-hole and detector of a collimator used for the SPECT apparatus which concerns on 2nd Embodiment from the radiation irradiation direction. It is a one-dimensional system diagram for calculating a point response function of a collimator used in the SPECT apparatus according to the second embodiment. It is a graph which shows an example of the point response function of the collimator used for the SPECT apparatus which concerns on 2nd Embodiment. It is the graph which divided and showed an example of the point response function of the collimator used for the SPECT apparatus which concerns on 2nd Embodiment by the left-right detector group. (A) is the conceptual diagram which showed the viewing angle of the left detector group, (b) is the conceptual diagram which showed the viewing angle of the right detector group. It is the graph which showed the point response function of the integrated detector which concerns on a comparative example, and the point response function of the left-right detector group which concerns on 2nd Embodiment. It is a graph explaining the half value width and 1/10 width | variety of the point response function of the integrated detector which concerns on a comparative example, and the point response function of the right detector group which concerns on 2nd Embodiment. It is a graph which shows an example of the distance dependence from the collimator of the collimator resolution in the SPECT apparatus which concerns on 2nd Embodiment. It is a conceptual diagram which shows the resolution of the tomographic image of a SPECT apparatus which concerns on a modification, and a tomographic image.

  Hereinafter, modes for carrying out the present invention (hereinafter referred to as “embodiments”) will be described in detail with reference to the drawings as appropriate. In each figure, common portions are denoted by the same reference numerals, and redundant description is omitted.

<< First Embodiment >>
<SPECT device (radiation imaging device)>
The overall configuration of the SPECT apparatus (radiation imaging apparatus) 1 according to the first embodiment will be described with reference to FIG. FIG. 1 is a configuration diagram of a SPECT apparatus 1 according to the first embodiment.
The SPECT apparatus 1 includes a gantry 10, cameras 11 </ b> A and 11 </ b> B, a data processing device 12, a display device 13, and a bed 14.

The subject 15 is administered a radiopharmaceutical, for example, a drug containing ^99m Tc with a half-life of 6 hours. Gamma rays (radiation) emitted from ^99m Tc in the body of the subject 15 placed on the bed 14 are detected by the cameras 11A and 11B supported by the gantry 10 to take a planar image and a tomographic image. .

The cameras 11A and 11B have a built-in collimator 26 and a detector group 21A (see FIG. 2) composed of a large number of detectors 21. The collimator 26 has a through-hole 27 and a septa 28 that partitions the through-hole 27, selects gamma rays emitted from ^99m Tc in the body of the subject 15 (regulates the incident angle), and passes only gamma rays in a certain direction. Have the role of The detector 21 detects gamma rays that have passed through the collimator 26 (through hole 27).

  The cameras 11A and 11B include an application specific integrated circuit (hereinafter referred to as an ASIC (Application Specific Integrated Circuit)) 25 for measuring a gamma ray detection signal. A minute detection signal of gamma rays is input to the ASIC 25 via the detector board 23 and the ASIC board 24 and amplified. The detector signal amplified by the ASIC 25 is converted into a digital signal by an ADC (Analog Digital Converter) (not shown), and then by a digital circuit element such as an FPGA (Field-Programmable Gate Array) (not shown). The ID of the detector 21 that detected the gamma ray, the peak value of the detected gamma ray and the detection time are detected. These are surrounded by a light shielding / gamma ray / electromagnetic shield 29 made of aluminum, iron, lead or the like constituting the cameras 11A and 11B, and block light, gamma rays and electromagnetic waves.

  The cameras 11A and 11B can capture a planar image by capturing a fixed image. Further, the cameras 11A and 11B can be rotated with an attachment portion (not shown) with the gantry 10 as an axis, and two cameras 11A and 11B can be fixed side by side to capture a planar image. The cameras 11A and 11B can be moved in the radial direction and the circumferential direction of the central axis of the cylindrical opening provided in the central portion of the gantry 10. At the time of SPECT imaging, the cameras 11A and 11B draw an image by drawing a closest trajectory around the subject 15.

  The data processing device 12 includes a storage device (not shown) and a tomographic image information creation device (not shown). The data processing device 12 takes in packet data including the detected peak value of the gamma ray, detection time data, and detector (channel) ID from a digital circuit element such as an FPGA (not shown) to generate a planar image or sinogram data. Tomographic image information is generated and displayed on the display device 13.

The image reconstruction executed by the data processing device 12 will be described.
When the detector group 21A (see FIG. 2) is at an angle with respect to the measurement target, the count number y _i of the detector _i is expressed by equation (1), where λ _j is the count number of the detection reconstruction pixel j It is represented by Here, C _ij represents the probability of being detected by the detector i.
y _i = ΣC _ij λ _j (1)
From the above equation, an image is reconstructed using a successive approximation image reconstruction method or the like (MLEM method, OSEM method, MAP method, or the like). Note that the image reconstruction method is not limited to this, and for example, the FBP method may be used.
Then, the data processing device 12 performs a filtering process (anisotropic filtering process) described later on the tomographic image that is the reconstructed image.
In this way, the SPECT apparatus 1 can image the radiopharmaceutical (gamma rays released from the radiopharmaceutical) accumulated in the tumor or the like in the body of the subject 15 and identify the position of the tumor or the like. .

<Detector>
Next, the detector 21 used in the cameras 11A and 11B will be further described with reference to FIG. FIG. 2 is a perspective view showing a pixel-type detector group 21A built in the cameras 11A and 11B used in the SPECT apparatus 1 according to the first embodiment.
A detector group 21A is configured by two-dimensionally arranging detectors 21 using CdTe semiconductors on a detector substrate 23 (see FIG. 1). Further, each detector 21 constitutes one pixel. In FIG. 2, the upper surface side is an incident surface 21 f of the detector 21, and electrodes 22 a and 22 b for applying a voltage are disposed on the side surface of the detector 21. Thus, unlike a scintillator made of one large crystal, detection signals are collected in units of detectors 21, that is, in units of pixels.

  In order to easily obtain the point response function, the detector group 21A preferably has a periodic structure. Otherwise, a point response function is determined for each pixel.

  The detectors 21 (detector group 21A) used in the cameras 11A and 11B are not limited to those divided for each pixel as shown in FIG. 2, and the detectors (detector group) shown in FIGS. 21B, 21C, 21D, 21E) may be used.

FIG. 3 is a perspective view showing another example of a pixel-type detector group.
In the detector (detector group 21B) shown in FIG. 3, the common electrode 22c is disposed on one surface of the CdTe semiconductor substrate, that is, on the entire incident surface 21f side, with respect to one CdTe semiconductor substrate. An electrode 22d divided in units of pixels is arranged on the surface opposite to the incident surface 21f, and the CdTe semiconductor substrate and the common electrode 22c in an area corresponding to one of the electrodes 22d correspond to the pixels, respectively. This constitutes a detector.

Next, modified examples of the detector (detector group 21B) shown in FIG. 3 are shown in FIGS.
FIG. 4 is a perspective view showing an incident surface side of a first modification of another pixel type detector group, and FIG. 5 is a perspective view showing an opposite surface side of the incident surface of the detector group of FIG. It is.
In the detector (detector group 21C) shown in FIGS. 4 and 5, a common electrode 22c is arranged on the entire incident surface 21f side with respect to one CdTe semiconductor substrate, and the incident surface 21f of the CdTe semiconductor substrate. An electrode 22d divided in units of pixels is arranged on the side opposite to the side, and in addition, it has a structure divided into individual detectors by grooves formed by dicing.

FIG. 6 is a perspective view showing a second modification of another pixel type detector group.
In the detector (detector group 21D) shown in FIG. 6, a plurality of strip-like electrodes 22e and 22f are opposed to the top and bottom surfaces of the CdTe semiconductor substrate in a right-angled relationship with respect to one CdTe semiconductor substrate. Arranged. The belt-like electrode 22e on either the upper surface or the lower surface is used as an anode, and the belt-like electrode 22f on the other surface is used as a cathode. A crossed portion of the anode electrode 22e and the cathode electrode 22f forms one detector (see Japanese Patent Application Laid-Open No. 2004-125757).

FIG. 7 is a perspective view showing a pixel-type scintillator detector.
The structure of the detector may be a scintillator detector that is divided into pixel units each composed of a scintillator 21g and a photodiode 21h, as in the detector (detector group 21E) shown in FIG.
In this case, the side surface of each scintillator 21g is surrounded by a light shielding material (not shown). Further, as a modification of the scintillator detector shown in FIG. 7, the scintillator detector may be composed of a scintillator 21g divided for each pixel and a position-sensitive photomultiplier tube (PSPMT).

<Collimator>
Next, the collimator 26 used in the cameras 11A and 11B will be further described with reference to FIG. FIG. 8 is a perspective view showing a collimator 26 built in the cameras 11A and 11B used in the SPECT apparatus 1 according to the first embodiment.
The collimator 26 is made of lead, and has a through-hole 27 in a direction that can be seen when viewed in a plan view from a direction perpendicular to the incident surface 21 f of the detector 21, and the through-hole 27 is arranged in a grid pattern. Each through hole 27 is partitioned by a septa 28.
As shown in FIG. 8, the collimator 26 used in the SPECT apparatus 1 according to the first embodiment has a configuration in which one detector 21 is included for one through hole 27.

<Resolution and image quality>
Next, the resolution of the tomographic image of the SPECT apparatus 1 according to the first embodiment will be considered. FIG. 9 is a graph showing an example of the distance dependence of the collimator resolution from the collimator 26 in the SPECT apparatus according to the first embodiment, the horizontal axis shows the distance from the collimator 26 to the point source, and the vertical axis shows the spatial resolution. Represents FWHM (Full Width at Half Maximum). The full width at half maximum means the minimum distance between the source capable of distinguishing two point sources.
In the example of FIG. 9, the pitch of the detector 21 (the distance between the center of the detector 21 and the center of the detector 21 adjacent in the x direction or the y direction) is 1.4 mm, and the pitch of the septa 28 (the septa The distance between the center of 28 and the center of the septa 28 adjacent in parallel) was 1.4 mm, and the septa length (z-direction length of the septa 28) was 26 mm.
As shown in FIG. 9, the FWHM increases in proportion to the increase in the distance from the collimator 26 to the point source (in other words, the distance from the detector 21 (detector group 21A) to the point source). It can be seen that the spatial resolution decreases.

FIG. 10 is a conceptual diagram showing the tomographic image and the resolution of the tomographic image of the SPECT apparatus 1 according to the first embodiment.
FIG. 10 shows circular RI accumulation (RI distribution RI ₁ , RI distribution) having the same radius at two locations, the central portion of the subject 15 (rotation center of the cameras 11A and 11B) and the peripheral portion (near the surface of the subject 15). An example of a tomographic image having the normal direction in the body axis direction of the subject 15 when there is RI ₂ ) is shown by a thick solid line.

For RI distribution RI ₁ in the center, RI and the distance from the detector 21 (detector array 21A) at the detector position A to RI distribution RI _1, from the detector 21 (detector array 21A) at the detector position B The distance to the distribution RI ₁ is almost equal. Therefore, when the detector 21 (detector group 21A) is at the detector position A, the resolution (FWHM of the count number R _cir1 ) and when the detector 21 (detector group 21A) is at the detector position B The resolution (FWHM of the count number R _rad1 ) is approximately equal isotropic resolution, and the round RI distribution RI ₁ can be imaged in a tomographic image as shown in FIG.

For the RI distribution RI _{2 at the} periphery, the distance from the detector 21 (detector group 21A) to the RI distribution RI ₂ at the detector position A is from the detector 21 (detector group 21A) at the detector position B to RI. It is shorter than the distance to the distribution RI _2. Therefore, when the detector 21 (detector group 21A) is located at the detector position A, the resolution (FWHM of the count number R _cir2 ) (circumferential resolution) is high (FWHM is small). When 21 (detector group 21A) is at detector position B, the resolution (FWHM of count number R _rad2 ) (radial resolution) is low (FWHM is large).
As described above, the RI distribution RI ₂ in the peripheral portion has an anisotropic resolution in which the circumferential resolution (FWHM of the count number R _cir2 ) and the radial resolution (FWHM of the count number R _rad2 ) are different. In the tomographic image, the round RI distribution RI ₂ is distorted into an elliptical shape extending in the radial direction as shown in FIG.

<Anisotropic filtering process>
Therefore, the data processing device 12 (see FIG. 1) of the SPECT device 1 according to the first embodiment particularly performs filtering processing (differential) on the tomographic image so that the circumferential resolution is comparable to the radial resolution. (Directional filtering process).
Specifically, the low-frequency pass filtering process is performed so that the cutoff frequency in the circumferential direction (cutoff frequency) is approximately the same as the cutoff frequency in the radial direction.
Note that the filtering process may be performed only in the circumferential direction, or may be performed more strongly in the circumferential direction than in the radial direction.

The image after the filtering process is shown in FIG. FIG. 11 is a conceptual diagram showing a tomographic image after the filtering process of the SPECT apparatus 1 according to the first embodiment. As shown in FIG. 11, not only the RI distribution RI ₁ in the central portion but also the RI distribution RI ₂ in the peripheral portion can have isotropic resolution, and distortion on the tomographic image can be reduced. it can.

Patent Document 2 is a method for reducing the anisotropy of a tomographic image by performing different filtering processes corresponding to the distance between the collimator and the imaging target (RI distribution) of the subject for each projection data, This is effective when the imaging target is a highly integrated organ such as the heart and has a small volume.
However, when the imaging target is a large volume such as a brain and the difference in resolution between the central part and the peripheral part is large, the method of the patent document is not effective. That is, in the method of Patent Document 2, since the filtering process is performed on the projection data, not only the circumferential direction but also the radial resolution is deteriorated, and the overall resolution is lowered.

  On the other hand, in the anisotropic filtering process in the SPECT apparatus 1 according to the first embodiment, the circumferential resolution decreases in accordance with the radial direction, but the radial resolution does not deteriorate. For this reason, anisotropy can be removed without impairing the overall resolution. That is, even when the imaging target has a large volume, such as a brain, and a large difference in resolution between the central part and the peripheral part, a suitable tomographic image with reduced resolution anisotropy and reduced distortion can be obtained.

<< Second Embodiment >>
<SPECT device (radiation imaging device)>
Next, a SPECT apparatus (radiation imaging apparatus) 1 according to the second embodiment will be described. FIG. 12 is a perspective view showing a collimator 26A built in the cameras 11A and 11B used in the SPECT apparatus 1 according to the second embodiment. FIG. 13 is a view of the arrangement of one through-hole 27A and the detector 21 of the collimator 26A used in the SPECT apparatus 1 according to the second embodiment as seen from the radiation irradiation direction.

  The difference between the SPECT apparatus 1 according to the second embodiment and the SPECT apparatus 1 according to the first embodiment is that the collimator 26 of the first embodiment includes one detector 21 for one through hole 27. In contrast, the collimator 26A of the second embodiment includes four detectors 21 for one through hole 27A (see FIG. 12, FIG. 8). (See FIG. 13). That is, the collimator 26A used in the SPECT apparatus 1 according to the second embodiment has a configuration in which four detectors 21 are included for one through hole 27A. About another structure, it is the same as that of the SPECT apparatus 1 which concerns on 1st Embodiment, and abbreviate | omits description.

In addition, the image reconstruction executed by the data processing device 12 of the SPECT apparatus 1 according to the second embodiment is similar to the first embodiment, from the equation (1), the successive approximation image reconstruction method or the like (MLEM method, An image is reconstructed using an OSEM method, a MAP method, or the like.
In particular, in the second embodiment, the spatial resolution is corrected by incorporating the point response function of the detector i into the successive approximation image reconstruction. Here, the point response function is the probability that the detector 21 detects the radiation generated from the point source, and is equal to the detection probability C _ij in the equation (1). By using this point response function, it is possible to provide a higher resolution image using the successive approximation image reconstruction method.

<Point response function>
Next, the point response function in the collimator 26A of the SPECT apparatus 1 according to the second embodiment will be considered. FIG. 14 is a one-dimensional system diagram for calculating the point response function of the collimator 26A used in the SPECT apparatus 1 according to the second embodiment. Here, for the sake of simplicity, a one-dimensional system will be discussed as shown in FIG. The actual detector 21 (detector group 21A) and the collimator 26A are two-dimensionally arranged (see FIG. 12), but the consideration of this one-dimensional system may be extended to two dimensions. For the sake of simplicity, discussion will be made assuming that the septa 28A and the detector 21 completely absorb gamma rays. In the following discussion, the detector 21 pitch was 1.4 mm, the septa 28A pitch was 2.8 mm, the septa length l was 26 mm, and the distance L from the collimator 26A to the point source 30 was 100 mm.

FIG. 15 is a graph showing an example of a point response function of the collimator 26A used in the SPECT apparatus 1 according to the second embodiment.
FIG. 15 shows the solid angle (proportional to the sensitivity) for each detector 21 from the point source 30 in the one-dimensional system of FIG. “♦” on the graph corresponds to one detector 21. The horizontal axis indicates the position of the detector 21, and the vertical axis indicates the solid angle at which the detector 21 is desired from the point source 30.
As shown in FIG. 15, it can be seen that the detector 21 behind the septa 28 </ b> A has a small solid angle and the detector 21 that is not so alternately, and the response function is complicated.

  According to the inventors' consideration, the point response function shown in FIG. 15 is a superposition of two types of point response functions, and by considering them separately, the point response function of the collimator 26A according to the second embodiment is simplified. The following can be understood.

  As shown in FIG. 14, the detector 21 (detector group 21A) is divided into a right detector group (right detector group 21R) and a left detector group (left detector group) in the through hole 27 of the collimator 26A. 21L). FIG. 16 shows the point response function shown in FIG. 15 divided into the point response function of the right detector group 21R and the point response function of the left detector group 21L.

FIG. 16 is a graph showing an example of the point response function of the collimator 26A used in the SPECT apparatus 1 according to the second embodiment, divided into the right detector group 21R and the left detector group 21L. “□” on the graph corresponds to one detector 21 of the left detector group 21L, and “■” on the graph corresponds to one detector 21 of the left detector group 21L. The horizontal axis indicates the position of the detector 21, and the vertical axis indicates the solid angle at which the detector 21 is desired from the point source 30.
As shown in FIG. 16, it can be seen that the point response function shown in FIG. 15 is a superposition of two types of substantially trapezoidal point response functions.
15 is a curve display as one point response function in the detector group 21A, and FIG. 16 is a curve divided into a point response function of the right detector group 21R and a point response function of the left detector group 21L. These are the same point response functions. That is, the position of “♦” on the graph of FIG. 15 matches the position of “□” or “■” on the graph of FIG.

Figure 17 (a) is a conceptual diagram showing a viewing angle VA _L of the left detectors 21L, FIG. 17 (b) is a conceptual diagram showing a viewing angle VA _R of the right detectors 21R.
As shown in FIG. 17, left and right detectors 21L, viewing angle _VA L, the VA _R are different 21R. That is, as shown in FIG. 17A, the field of view by the left detection group 21L is tilted to the right with respect to the direction perpendicular to the detector surface, and as shown in FIG. 17B, by the right detector group 21R. The field of view is tilted to the left with respect to the direction perpendicular to the detector surface.
The two types of point response functions shown in FIG. 16 (the point response function of the right detector group 21R and the point response function of the left detector group 21L) are substantially trapezoidal point response functions, each of which has an origin (x = 0) to have shifted from this viewing angle _VA L, it is due to differences in VA _R.

<Resolution and image quality>
Next, the spatial resolution of each planar image when the point response function is separated into the point response function of the right detector group 21R and the point response function of the left detector group 21L will be considered.

Here, as a comparative example, in the arrangement of the collimator 26A and the detector 21 (detector group 21A) shown in FIG. 12 to FIG. 14, the four detectors 21 included in the through hole 27A are independent detectors. FIG. 18 shows a point response function when considered as one detector (hereinafter referred to as an integral detector) (indicated by “◯” on the graph). In FIG. 18, the point response function of the right detector group 21R and the point response function of the left detector group 21L shown in FIG. 16 are also shown for comparison.
As shown in FIG. 18, it can be seen that the point response function of the integrated detector according to the comparative example is the sum of the point response function of the right detector group 21R and the point response function of the left detector group 21L. .

  The point response function of the integrated detector according to the comparative example is standardized with the maximum value, and the point response function of the right detector group 21R is shifted in the x-axis direction so that the center of gravity is at the origin (x = 0) and is standardized with the maximum value. The converted graph is shown in FIG. The center of gravity of the point response function of the right detector group 21R is normalized by the maximum value so that the center of gravity of the point response function of the left detector group 21L is shifted to the origin (x = 0). Is the same except that it is shifted in the x-axis direction so that it is at the origin (x = 0) and normalized by the maximum value and the left and right are reversed.

  As shown in FIG. 19, compared with the point response function of the integrated detector, the point response function of the right detector group 21R has the same FWHM (Full Width at Half Maximum), but FWTM (Full Width at (Tenth Maximum, 1/10 width) is small. Equal FWHM means equal resolution. A small FWTM means that the spread of data to the periphery is small, and a sharp, high-contrast image can be provided.

  That is, the planar image for each of the left and right detector groups 21L and 21R is sharper than the integrated detector (the four detectors 21 included in the through-hole 27A are one detector) (boundary boundary). It is clear that the image has a high contrast and a clear contrast.

  In addition, the tomographic image obtained by reconstructing an image from these planar images is also a good image (sharp boundary) and a high-contrast image (sharp contrast). Moreover, since the spatial resolution is corrected by incorporating the point response function into the successive approximation image reconstruction, a higher resolution image can be provided.

  FIG. 20 is a graph showing an example of the distance dependence of the collimator resolution from the collimator 26A in the SPECT apparatus according to the second embodiment, the horizontal axis shows the distance from the collimator 26A to the point source, and the vertical axis shows the spatial resolution. Represents the FWHM (Full Width at Half Maximum) of the point response function of the right or left detector group. In the example of FIG. 20, the detector 21 has a pitch of 1.4 mm, the septa 28A has a pitch of 2.8 mm, and the septa length is 26 mm.

  In the second embodiment, higher resolution can be achieved by correcting the spatial resolution. However, as shown in comparison between FIG. 9 (first embodiment) and FIG. 20 (second embodiment), the resolution (FWHM) of the second embodiment is higher than that of the first embodiment. The distance dependence from the distance is high (the inclination of the graph is large), and the distortion of the image at the peripheral part becomes remarkable.

  On the other hand, it is effective to perform the anisotropic filtering process on the tomographic image, and the distortion of the image of the peripheral part can be reduced without reducing the overall resolution.

Next, artifacts due to statistical noise will be described.
The successive approximation image reconstruction incorporating the resolution correction is a method of converging from the projection data blurred by the collimator 26A and the detector 21 to the correct RI generation position distribution by iterative calculation while weighting by the point response function. However, when there are noise components in the image, the noise components are collected and emphasized, and artifacts are often generated.

In particular, in the second embodiment, an increase in FWHM (a decrease in spatial resolution) is significant with respect to an increase in the distance from the collimator 26A to the point source. For this reason, in the tomographic image, the distortion of the image of the peripheral part is remarkable, and the image of the peripheral part is stretched without being separated in the radial direction, but on the other hand, the image is strongly separated in the circumferential direction. .
For this reason, if there is noise, artifacts extending radially are generated in the peripheral portion. Statistical artifacts that are difficult to avoid in nuclear medicine diagnosis where sufficient counts cannot be obtained cause radial artifacts in the peripheral region of the image.

  On the other hand, it is possible to reduce artifacts by performing the above-described anisotropic filtering process on the tomographic image so that the circumferential resolution of the peripheral part is the same as the radial resolution.

Next, an artifact caused by a point response function used for resolution correction being different from an actual point response function will be described.
The point response function used in the second embodiment is based on simple calculation in order to explain the essence of the present invention, and actually measured values should be used. However, the point response function is distance dependent, and it is difficult to obtain a necessary response function by actual measurement. Therefore, a point response function calculated by a more strict ray tracing method or Monte Carlo simulation considering the thickness of the collimator 26A and the gap (dead zone) of the detector 21 is used.

  However, as a matter of course, the point response function by this simulation does not exactly match the actual point response function. This mismatch component causes artifacts. In particular, if there is a noise component (statistical noise), this artifact is emphasized, and radial artifacts due to resolution anisotropy occur.

  On the other hand, it is possible to reduce artifacts by performing the above-described anisotropic filtering process on the tomographic image so that the circumferential resolution of the peripheral part is the same as the radial resolution.

≪Modification≫
The radiation imaging apparatus according to the present embodiment is not limited to the configurations of the first embodiment and the second embodiment, and various modifications can be made without departing from the spirit of the invention.

  In this embodiment, taking the tomographic image (see FIG. 10) having the body axis direction of the subject 15 as a normal line as an example, the anisotropic filtering processing described above is performed with the anisotropy of resolution in the circumferential direction and the radial direction as a problem. However, the present invention may be applied to resolution anisotropy in other directions.

FIG. 21 is a conceptual diagram showing the tomographic image and the resolution of the tomographic image of the SPECT apparatus 1 according to the modification. FIG. 21 shows a circular RI having the same radius at two locations, a central portion of the subject 15 (rotation center of the cameras 11A and 11B rotating around the body axis direction) and a peripheral portion (near the surface of the subject 15). An example of a tomographic image sliced in the body axis direction of the subject 15 when there is accumulation (RI distribution RI ₁ , RI distribution RI ₂ ) is indicated by a thick solid line.

For the RI distribution RI _{2 at the} periphery, the distance from the detector 21 (detector group 21A) to the RI distribution RI ₂ at the detector position C is the detector position D (not shown) from the detector position C (cameras 11A and 11B). Is shorter than the distance from the detector 21 (detector group 21A) to the RI distribution RI ₂ at the position rotated 90 degrees. Therefore, when the detector 21 (detector group 21A) is at the detector position C, the resolution (FWHM of the count number _Raxi2 ) (body axis direction resolution) is high (FWHM is small), When the detector 21 (detector group 21A) is at a detector position D (not shown), the resolution (FWHM of the count number R _rad2 ) (radial resolution) is low (FWHM is large).
As described above, the RI distribution RI ₂ in the peripheral portion has anisotropic resolution in which the resolution in the body axis direction (FWHM with the count number R _axi2 ) and the radial resolution (FWHM with the count number R _rad2 ) are different. Therefore, the round RI distribution RI ₂ is imaged in a tomographic image as shown in FIG. 21 distorted in an elliptical shape extending in the radial direction.

Thus, in the modification shown in FIG. 21, the difference between the resolution in the body axis direction and the radial direction is the same as in the first embodiment in which the anisotropic filtering process is performed on the anisotropy in the resolution in the circumferential direction and the radial direction. An anisotropic filtering process is performed on the sex. That is, the filtering process (anisotropic filtering process) is performed on the tomographic image so that the resolution in the body axis direction is approximately the same as the resolution in the radial direction. Specifically, low-frequency-pass filtering processing is performed so that the cutoff frequency (cutoff frequency) in the body axis direction is approximately the same as the cutoff frequency in the radial direction.
The filtering process may be performed only in the body axis direction, or may be performed more strongly in the body axis direction than in the radial direction.

  In the present embodiment, no mention is made of filter types in the anisotropic filtering process. This may basically be a filter in a direction that lowers the resolution, that is, it may be a low-frequency band-pass filter, for example, a smoothing process or the like. The filter type is not limited.

In the present embodiment, the anisotropic filtering process is performed on the tomographic image after the image reconstruction. However, the present invention is not limited to this, and the projection data before the image reconstruction may be filtered. It is also possible to incorporate an anisotropic filtering process during the iterative process of successive approximation image reconstruction.
By incorporating an anisotropic filtering process during the iterative process, an effect of suppressing artifacts and image distortion can be obtained, and the number of iterations can be reduced. Alternatively, a higher quality image can be provided even with the same number of iterations.

  The radiation imaging apparatus according to the present embodiment has been described as being a SPECT apparatus, but a positron emission computed tomography apparatus (PET (Positron Emission Computed Tomography) apparatus) using positron emission tomography (Positron Emission Tomography). The anisotropic filtering process described above may be applied.

  In addition, the radiation imaging apparatus according to the second embodiment has been described as having a configuration in which four detectors are included in one through-hole 27A, but is not limited thereto, and one through-hole 27A. However, a configuration in which M detectors 21 are included may be used. Note that M may not be an integer.

1 SPECT equipment (radiation imaging equipment)
10 Gantry 11A, 11B Camera 12 Data processing device (data processing means)
13 Display device 14 Bed 15 Subject 21 Detector 21A Detector group 21f Incident surface 23 Detector substrate 24 ASIC substrate 25 ASIC
26, 26A Collimator 27, 27A Through hole 28, 28A Septa 29 Light shielding / gamma ray / electromagnetic shield 30 Point source

Claims (16)

A detector group having a plurality of detectors for measuring radiation;
A collimator disposed in front of the detector to limit the incident direction of the radiation;
A tomographic image creating method by a radiation imaging apparatus comprising: a data processing means;
The data processing means includes
A collecting step of collecting a plurality of projection data from the detector group;
A reconstruction step of reconstructing a tomographic image from the collected projection data;
A tomographic image creating method comprising: performing a correction processing step of performing filtering processing of different cutoff frequencies in each axial direction of the reconstructed tomographic image. The collimator is
Partitioned by a septa and having a through-hole in a direction that can be seen when viewed from above in a direction perpendicular to the radiation incident surface of the detector,
The tomographic image creation method according to claim 1 , wherein the plurality of detectors are arranged in the through hole when viewed in plan from the vertical direction. A detector group having a plurality of detectors for measuring radiation;
A collimator disposed in front of the detector to limit the incident direction of the radiation;
A tomographic image creating method by a radiation imaging apparatus comprising: a data processing means;
The collimator is
Partitioned by a septa and having a through-hole in a direction that can be seen when viewed from above in a direction perpendicular to the radiation incident surface of the detector,
When the planar view from the vertical direction, the plurality of detectors are arranged in the through hole,
The data processing means includes
A collecting step of collecting a plurality of projection data from the detector group;
A correction processing step of performing filtering processing of different cutoff frequencies in each axial direction of the collected projection data;
Reconstructing a tomographic image from the filtered projection data. A detector group having a plurality of detectors for measuring radiation;
A collimator disposed in front of the detector to limit the incident direction of the radiation;
A tomographic image creating method by a radiation imaging apparatus comprising: a data processing means;
The collimator is
Partitioned by a septa and having a through-hole in a direction that can be seen when viewed from above in a direction perpendicular to the radiation incident surface of the detector,
When the planar view from the vertical direction, the plurality of detectors are arranged in the through hole,
The data processing means includes
A collecting step of collecting a plurality of projection data from the detector group;
A tomographic image is reconstructed from the collected projection data by a successive approximation method with resolution correction, and a filtering process with different cutoff frequencies is performed in each axial direction of the data being repeatedly processed by the successive approximation image reconstruction. A tomographic image generation method comprising: performing a configuration step. The axial directions are a radial direction and a circumferential direction,
The filtering process, a tomographic image generating method according to any one of claims 1 to 4, characterized in that the strong filtering than said radial in the circumferential direction. The axial directions are a radial direction and a body axis direction,
The tomographic image creation method according to any one of claims 1 to 4 , wherein the filtering process performs a filtering process stronger in the body axis direction than in the radial direction. A detector group having a plurality of detectors for measuring radiation;
A collimator disposed in front of the detector to limit the incident direction of the radiation;
A radiographic imaging device comprising data processing means,
The data processing means includes
Collecting a plurality of projection data from the detector group;
Reconstructing a tomographic image from the collected projection data,
A radiation imaging apparatus, wherein filtering processing of different cutoff frequencies is performed in each axial direction of the reconstructed tomographic image. The collimator is
Partitioned by a septa and having a through-hole in a direction that can be seen when viewed from above in a direction perpendicular to the radiation incident surface of the detector,
The radiation imaging apparatus according to claim 7 , wherein a plurality of the detectors are disposed in the through hole when viewed in plan from the vertical direction. A detector group having a plurality of detectors for measuring radiation;
A collimator disposed in front of the detector to limit the incident direction of the radiation;
A radiographic imaging device comprising data processing means,
The collimator is
Partitioned by a septa and having a through-hole in a direction that can be seen when viewed from above in a direction perpendicular to the radiation incident surface of the detector,
When the planar view from the vertical direction, the plurality of detectors are arranged in the through hole,
The data processing means includes
Collecting a plurality of projection data from the detector group;
Perform filtering processing of different cutoff frequencies in each axial direction of the collected projection data,
Reconstructing a tomographic image from the filtered projection data;
A radiation imaging apparatus. A detector group having a plurality of detectors for measuring radiation;
A collimator disposed in front of the detector to limit the incident direction of the radiation;
A radiographic imaging device comprising data processing means,
The collimator is
Partitioned by a septa and having a through-hole in a direction that can be seen when viewed from above in a direction perpendicular to the radiation incident surface of the detector,
When the planar view from the vertical direction, the plurality of detectors are arranged in the through hole,
The data processing means includes
Collecting a plurality of projection data from the detector group;
A tomographic image is reconstructed from the collected projection data by a successive approximation method with resolution correction, and a filtering process with different cutoff frequencies is performed in each axial direction of the data being repeatedly processed by the successive approximation image reconstruction. A radiation imaging apparatus. The axial directions are a radial direction and a circumferential direction,
The radiation imaging apparatus according to claim 7 , wherein in the filtering process, the filtering process is performed more strongly in the circumferential direction than in the radial direction. The axial directions are a radial direction and a circumferential direction,
It said filtering processing, the radiation imaging apparatus according to any one of claims 8 to 10, characterized in that performing strongly filtering than said radial in the circumferential direction. The axial directions are a radial direction and a body axis direction,
The radiation imaging apparatus according to claim 7 , wherein in the filtering process, the filtering process is performed more strongly in the body axis direction than in the radial direction. The axial directions are a radial direction and a body axis direction,
The radiation imaging apparatus according to any one of claims 8 to 10 , wherein in the filtering process, the filtering process is performed more strongly in the body axis direction than in the radial direction. The radiation imaging apparatus includes:
The radiation imaging apparatus according to any one of claims 7 to 14 characterized in that it is a single-photon emission computed tomography apparatus. The radiation imaging apparatus includes:
The radiation imaging apparatus according to claim 7 , wherein the radiation imaging apparatus is a positron emission computed tomography apparatus.

Images