JP5601208B2 - Positron emission tomography system - Google Patents

Description

  The present invention relates to a positron emission tomography apparatus that detects paired annihilation photons from a subject to which a radiopharmaceutical has been administered and images the distribution of the radiopharmaceutical in the subject, and more particularly to a technique for reducing artifacts.

  A positron emission tomography (PET (Positron Emission Tomography) device: hereinafter also referred to as “PET device”) is a 511 KeV pair annihilation photon emitted in a 180 ° opposite direction from a subject to which a radiopharmaceutical is administered ( For example, γ rays are detected by a ring detector in which a plurality of detectors provided in the PET gantry are arranged in a ring shape. And the time which this photon detected is measured, and when the detection time difference in two detectors is less than a fixed time, it counts as a pair of annihilation photon (simultaneous counting). Further, the occurrence point of pair annihilation is specified as existing on the straight line of the two detectors. The data counted in this way is accumulated and reconstructed to obtain a tomographic image (see, for example, Patent Document 1). In addition, when using a radiopharmaceutical that easily accumulates in a specific lesion, the lesion appears as a high pixel value on the tomographic image and can be identified.

  Further, in the PET apparatus, a plurality of detectors are arranged in a ring shape (more precisely, a polygon) in order to acquire a tomographic image. This is to collect information about all angles with respect to the tomographic direction and reconstruct a tomographic image (complete projection).

JP 2008-245695 A

  However, as in Patent Document 1, if a part of information is lost due to a failure of a detector or the like, there is a problem that an artifact is generated in a tomographic image. That is, artifacts due to incomplete projection occur. Therefore, conventional devices reduce artifacts by avoiding the use of detector data that has lost some information due to a failure, or by using data from nearby detectors such as adjacent detectors as they are. Yes. Basically, these methods are temporary measures up to the replacement of the detector.

  However, in the case of, for example, a C-type detector arrangement in which some (for example, two) detectors are extracted from a plurality of detectors arranged in a ring shape, information of complete projection is never obtained. Also, the angle at which information is missing is large. Therefore, even if the reconstruction is performed without using the detector data or using the nearby detector data as it is, the artifacts are remarkably generated.

  The present invention has been made in view of such circumstances, and it is an object of the present invention to provide a positron emission tomography imaging apparatus capable of reducing artifacts even when a part of a detector is extracted. To do.

In order to achieve such an object, the present invention has the following configuration. That is, the positron emission tomography apparatus according to the present invention is configured by extracting a part of a plurality of detectors arranged in a ring shape, and detects pair annihilation photons emitted from a subject to which a radiopharmaceutical is administered. The number of interpolating LORs that are estimated to be collected if there is the extracted detector part, and a data collecting part that collects LOR data indicating a line connecting the detector pair in which the pair annihilation photon is detected. interpolation LOR data creation unit for creating interpolated LOR data out calculate the, LOR interpolation processing for adding processing the interpolated LOR data calculated in the LOR data collected by the data collection unit in the interpolation LOR data creation unit comprising a part, the said interpolation LOR data creation unit, based on the LOR data collected by the data collecting unit, minute inside the plurality of detectors in a grid The number of LORs that have passed through each grid in the space is counted, the number of LORs that have passed through each grid that has been counted, the angle of the measured detection surface of each grid, and the extracted detector portion in each grid The number of interpolation of each grid is calculated based on the angle of the interpolation detection surface, and the interpolation LOR data creation unit calculates each of the calculated interpolation LORs that pass through each grid of the space divided into the grid. The interpolation LOR data is created by calculating the number of interpolation LORs by accumulating the number of lattices to be interpolated .

According to the positron emission tomography apparatus according to the present invention, the data collection unit collects LOR data indicating a line connecting the detector pair in which the pair annihilation photon is detected, and the interpolation LOR data creation unit collects the extracted detector. are creating an interpolation LOR data out calculate the number of interpolation LOR be estimated that the portion is collected, if any. Since the interpolation by the interpolation LOR data issued calculate the LOR that missing of a detector portion withdrawn, it is possible to reduce artifacts that occurred on the tomographic images.

  That is, among the detection units configured by extracting a part of a plurality of detectors arranged in a ring shape, if there was a detector of the extracted part, the missing LOR data that would have been actually measured was measured. It can be estimated and interpolated from LOR data. Therefore, artifacts occurring in the tomographic image can be reduced. Thereby, this apparatus provided with the detection part comprised by extracting a part of detector arrange | positioned at a ring shape can be made effective.

  In addition, the detection unit is a detector part extracted from a plurality of detectors arranged in a ring shape, and various objects such as a blood collection tube and other objects such as an object placed on a top plate and light Can pass through. For example, when a plurality of detectors is a detection unit (detector ring) arranged in a complete ring shape, the subject in the detection unit feels a sense of pressure. However, since the detection unit is configured by extracting a part of a plurality of detectors arranged in a ring shape, light or air can pass through the extracted detector part, The feeling of pressure can be reduced. In addition, since the number of detectors constituting the detection unit can be reduced, the manufacturing cost of the apparatus can be reduced.

Further, in the positron emission tomography apparatus according to the present invention, the space inside the plurality of detectors arranged in a ring shape of the detection unit is divided into a lattice shape, and the number of LORs passing through each lattice is counted. Grid data creation unit for creating grid data, and the interpolation LOR data creation unit is configured to calculate each of the number of LORs that have passed through each grid of the grid data, the angle of the measurement detection surface of each grid, and the total angle. Calculate the total number of LORs that pass through the grid, create the interpolation LOR data by calculating the number of interpolated lines of each grid from the estimated total number, the angle of the interpolation detection surface of the extracted detector part, and the total angle. It is preferable to do. In other words, the grid data creation unit creates grid data by counting the number of LORs that have passed through each grid obtained by dividing the space into grids, and the interpolation LOR data creation unit passes through each grid of grid data. Interpolated LOR data is created by calculating the number of LORs to be interpolated from the number of LORs, the angle of the actual measurement detection surface, the angle of the interpolation detection surface, and the like. Thereby. Since the missing LOR can be interpolated, artifacts occurring in the tomographic image can be reduced.

The positron emission tomography apparatus according to the present invention is interpolated by a rotation mechanism that rotates the detection unit around the central axis of a plurality of detectors arranged in a ring shape and the LOR interpolation processing unit . A rotation LOR interpolation processing unit for performing interpolation processing for interpolating LOR data of the position of the extracted detector portion with LOR data corresponding to the position of the extracted detector portion rotated and collected by the rotation mechanism; It is preferable to provide. Thereby, since the LOR data interpolated by the LOR interpolation processing unit is further rotated by the rotation mechanism and is collected by the LOR data corresponding to the position of the extracted detector portion, the tomographic image is obtained. The artifact generated in the above can be further reduced.

Further, in the positron emission tomography apparatus according to the present invention, the interpolation LOR data creation unit calculates the number of interpolation LORs by accumulating the calculated number of interpolations of each grid every time LOR data is collected. Then, the interpolation LOR data is created, and the LOR interpolation processing unit preferably interpolates the interpolation LOR data when the calculated integrated value of the interpolation LOR data is equal to or greater than a preset value. In the case of data collection that fluctuates with time, if LOR interpolation is performed collectively after data collection, artifacts may not be reduced satisfactorily depending on the amount of fluctuation. However, since the LOR interpolation processing unit interpolates the interpolated LOR data when the integrated value of the interpolated LOR data calculated every time LOR data is collected is equal to or greater than a preset value, LOR interpolation can be performed in real time. it can.

  An example of the interpolation processing of the positron emission tomography apparatus according to the present invention is a weighted average. Thus, for example, the weight of the extracted detector portion is set to “0 (zero)”, rotated by the rotation mechanism, and the weight of the LOR data detected and collected at the position of the extracted detector portion is weighted. Average processing is performed as “2”. When LOR data is collected before and after the rotation, the weighting of the collected LOR data is averaged as “1”. By performing the weighted average processing in this way, it is possible to satisfactorily perform interpolation processing.

  An example of the detection unit of the positron emission tomography apparatus according to the present invention is a C-type. As a result, it is possible to pass various objects such as a blood collection tube or the like, an object placed on the top board, light, and the like, and the manufacturing cost of the apparatus can be reduced. Furthermore, when photographing breast images, it is possible to satisfactorily photograph the fistula of the subject.

  In addition, an example of the detection unit of the positron emission tomography apparatus according to the present invention is configured by extracting a part of a plurality of detectors arranged in a ring shape and in multiple layers. For example, when a detection unit (detector ring) in which a plurality of detectors are arranged in a ring shape is configured in multiple layers in the body axis direction of the subject M, a part of the detection unit configured in multiple layers is used. Even if the image is extracted, a tomographic image with reduced artifacts can be acquired.

  Moreover, the positron emission tomography apparatus according to the present invention preferably includes an image reconstruction unit that reconstructs an image using the interpolated LOR data. As a result, image reconstruction is performed using LOR data obtained by interpolating the missing LOR data, so that a tomographic image with reduced artifacts can be acquired.

  According to the positron emission tomography apparatus according to the present invention, it is possible to actually measure if there is the extracted detector part among the detection parts configured by extracting a part of the plurality of detectors arranged in a ring shape. The missing LOR data, which may be, can be estimated and interpolated from the measured LOR data. Therefore, artifacts occurring in the tomographic image can be reduced. Thereby, this apparatus provided with the detection part comprised by extracting a part of detector arrange | positioned at a ring shape can be made effective.

According to the present invention, the data collection unit collects LOR data indicating a line connecting the detector pair in which the pair annihilation photon is detected, and the interpolation LOR data creation unit collects the extracted detector part if there is one. are creating an interpolation LOR data out calculate the number of interpolation LOR to estimate that. Since the interpolation by the interpolation LOR data issued calculate the LOR that missing of a detector portion withdrawn, it is possible to reduce artifacts that occurred on the tomographic images.

1 is a schematic configuration diagram of a positron emission tomography apparatus (PET apparatus) according to Example 1. FIG. It is a figure which shows an example of arrangement | positioning of the detector seen from the A direction of FIG. It is a figure where it uses for description of lattice-like data. It is a figure where it uses for description of operation | movement of a grid | lattice-like data preparation part. It is a figure where it uses for description of operation | movement of the interpolation LOR data preparation part. 3 is a schematic configuration diagram of a positron emission tomography apparatus (PET apparatus) according to Example 2. FIG. It is a figure where it uses for description of operation | movement of a LOR interpolation part, (a) shows the 1st LOR data collection, (b) shows the 2nd LOR data collection. 6 is a schematic configuration diagram of a positron emission tomography apparatus (PET apparatus) according to Example 3. FIG. It is a figure where it uses for description of operation | movement of a LOR interpolation part, (a) shows the 1st LOR data collection, (b) shows the 2nd LOR data collection. 6 is a schematic configuration diagram of a positron emission tomography apparatus (PET apparatus) according to Example 4. FIG. It is a figure which uses for description of the interpolation process which interpolates LOR data in real time, (a) shows measured LOR and interpolation LOR, (b) is the operation | movement which determines by integrating | accumulating the number of the calculated interpolation LOR. (C) shows the interpolation LOR addition process. It is a schematic block diagram of the positron emission tomography apparatus (PET apparatus) which concerns on Example 5, (a) shows the multilayer detection part, (b) is the arrangement | positioning of the detector seen from the A direction of (a). An example is shown. It is a figure which shows an example of arrangement | positioning of the detector which concerns on a modification.

  Embodiment 1 of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic configuration diagram of a positron emission tomography apparatus (PET apparatus) according to the first embodiment, and FIG. 2 is a diagram illustrating an example of the arrangement of detectors as viewed from the direction A in FIG. FIG. 3 is a diagram for explaining the grid data, and FIG. 4 is a diagram for explaining the operation of the grid data creation unit. FIG. 5 is a diagram for explaining the operation of the interpolation LOR data creation unit.

  Please refer to FIG. The PET apparatus 1 includes a bed apparatus 3 having a top plate 2 on which a subject M is placed and a gantry 5 having an opening 4. The bed apparatus 3 is configured to move the top plate 2 up and down in the vertical direction and to move the top plate 2 in parallel along the body axis direction of the subject M. The gantry 5 includes a detection unit 7 that detects γ rays generated from the subject M.

  As shown in FIG. 2, the detection unit 7 is configured by extracting some of the detectors 9 out of a plurality of detectors 9 arranged in a ring shape around the body axis of the subject M ( D). That is, the detection unit 7 is configured as a C type. The detector 9 is composed of, for example, a detector block including a scintillator block, a light guide, and a photomultiplier tube (all not shown). The scintillator block is composed of a plurality of scintillators. The scintillator block converts γ-rays generated from the subject M to which the radiopharmaceutical is administered into light, and the light guide guides the converted light, and the photomultiplier tube photoelectrically converts and outputs an electrical signal. . In addition, the opening 4 of the gantry 5 is configured to be opened in a tunnel shape, and to be opened along the extracted detector portion of the detector 9 arranged in a C shape.

  Further, the detection unit 7 is configured to be stacked in multiple layers in the body axis direction of the subject M. The detection unit 7 stacked in multiple layers is referred to as a multilayer detection unit 11.

  The data collection unit 13 is a line that connects a pair of detectors in which two γ rays radiating in the opposite directions of 180 ° are detected based on the electrical signals (detection signals) output from the detectors 9 of the multilayer detection unit 11. LOR (Line of Response) data is collected. The data collection unit 13 includes, for example, an amplifier, an A / D converter, an energy window determination unit, a coincidence circuit, and the like (all not shown). The electrical signal of γ rays is amplified and digitized. The energy window determination unit determines whether the amplified and digitized γ-ray electric signal is within a preset energy window (for example, 400 KeV to 600 KeV). Only the gamma rays determined to be within the energy window are simultaneously counted by the coincidence counting circuit. The coincidence circuit measures the time when the γ rays are detected, and simultaneously counts as a pair of γ rays if the time difference between the two detectors 9 is within a preset time (within the time window). Then, the pair of γ rays counted simultaneously is collected as LOR data.

  The LOR interpolation unit 15 interpolates the missing LOR in the interpolation LOR that is estimated to be collected if there is a detector portion extracted by the C-type detection unit 7. The LOR interpolation unit 15 includes a grid data creation unit 17, an interpolation LOR data creation unit 19, an addition processing unit 21, and the like. The addition processing unit 21 corresponds to the probability LOR interpolation processing unit in the present invention.

  The grid data generation unit 17 divides the space inside the plurality of detectors 9 arranged in a ring shape of the detection unit 7 into a grid pattern, and counts the number of LORs that have passed through each grid to generate grid data. 23 is created. That is, the grid data 23 first divides the space inside the multilayer detector 11 (a plurality of detectors 9) into a three-dimensional grid as shown in FIG. This one grid is called a voxel (or pixel). Then, as shown in FIG. 4, when the LOR data is collected, 1 is counted in all the lattices through which the LOR connecting the pair of detectors 9 counted simultaneously passes. In the lattice in which the LOR indicated by the symbol e in FIG. Such count is performed on all the measured LOR data, and the grid data 23 is created.

  As the grid data 23, histogram data is used. This has the effect that it is not necessary to create extra data. Histogram data is data obtained by compressing huge list data for convenience of storage capacity. The list data is data every time one LOR data is collected (every event), and the individual data are arranged in time series. Note that once the list data is histogram data, it cannot be restored to the original list data for reasons of the process. Therefore, when reconstructing an image from histogram data, LOR data (lattice data) after interpolation is created from the grid data 23.

  The interpolated LOR data creation unit 19 probabilistically calculates the number of interpolated LORs that are estimated to be collected from the number of LORs (counts) that have passed through each grid of the grid-like data 23 if there are extracted detector parts. Thus, interpolation LOR data is created. Specifically, the interpolation LOR data creation unit 19 calculates the total number of LORs that pass through each grid from the number of LORs that have passed through each grid of the grid data 23, the solid angle of the measurement detection surface of each grid, and all solid angles. The number of interpolation LORs is calculated by calculating the number of lines and calculating the number of interpolation LORs to be interpolated from the estimated total number and the extracted solid angle of the defect detection surface of the detector portion and all solid angles. That is, the number of LORs to be interpolated is calculated from the following equations (1) and (2).

Estimated total number of LORs passing through a certain grid Pm = count number of a certain grid / solid angle of actual detection surface × total solid angle (1)
Number of LORs to be interpolated = Pm × solid angle of interpolation detection surface / total solid angle (2)

  Note that the solid angle of the measurement detection surface indicates the solid angle of the surface that detects γ rays in the measurement of a configuration in which some of the detectors 9 of the multilayer detection unit 11 are extracted. The solid angle of the interpolation detection surface indicates the solid angle of the surface to be interpolated of the extracted detector portion. To explain in two dimensions, for example, when there are no gaps between adjacent detectors and two detectors 9 out of twelve detectors 9 arranged in a ring shape are detected, The angle actually measured by the vessel 9 is 300 ° {= 360 ° − (30 ° × 2 pieces)}. Further, the angle to be interpolated is an angle of 60 ° (30 ° × 2) for two detectors. Note that the solid angle of the actually measured detection surface and the solid angle of the interpolation detection surface that are different depending on each grid are given to the above-described equations (1) and (2).

  In this way, the interpolation LOR data creation unit 19 interpolates each grid estimated from the number of LORs that have passed through each grid of the grid-like data 23 if the extracted portion of the detector 9 can be measured. The number of is calculated. The calculations of equations (1) and (2) are performed for all the grids through which the interpolating LOR passes. The integrated value of all the lattices through which the LOR to be interpolated passes is the number of LORs to be interpolated. This process is performed for all interpolating LORs. Thereby, interpolation LOR data including the number of LORs to be interpolated is created.

  For example, if all the grids through which the LOR to be interpolated indicated by the symbol f in FIG. 5 passes are f1, f2, f3,..., Fn, interpolation is performed in each grid calculated by the formulas (1) and (2). Assume that the number of LORs is (f1, f2, f3,..., Fn) = (g1, g2, g3,..., Gn). The number G of LORs to be interpolated indicated by the symbol f is obtained as G = (g1 + g2 + g3 +... + Gn).

  The addition processing unit 21 adds the interpolation LOR data calculated by the interpolation LOR data creation unit 19 to the LOR data collected by the data collection unit 13. That is, the addition processing unit 21 adds and adds the LOR to be interpolated by the interpolation LOR data to the lattice data 23 created by the lattice data creation unit 17. In this case, addition is performed by counting the number of LORs to be interpolated that have passed through each grid of the grid data 23. For example, when the number G of LORs to be interpolated is 2.5, two may be added. Further, the addition may be performed with 2.5 pieces including the decimal part, or may be added by multiplying by 10 to 25 pieces. Correction processing including sensitivity correction described later can be performed including data after the decimal point.

  Note that, for example, when the volume of a lattice (voxel) is small and the total number of counts of each lattice is small, the LOR interpolation accuracy decreases. In this case, the lattice volume is set large. That is, each grid obtained by dividing the space into a grid is set in advance by the input unit 35 described later so that the LOR interpolation accuracy is good.

  The correction processing unit 25 performs necessary correction processing such as absorption correction and sensitivity correction performed by collecting transmission data, for example. The image reconstruction unit 27 reconstructs an image using the interpolated LOR data. Thereby, a tomographic image can be acquired.

  The PET apparatus 1 includes a main control unit 29, a display unit 31, a memory unit 33, and an input unit 35. The main control unit 29 controls each component centrally. The display unit 31 is composed of a liquid crystal display panel or the like, and displays a tomographic image obtained by image reconstruction. The memory unit 33 stores tomographic images and the like, and is configured by a storage medium such as a ROM (Read-only Memory), a RAM (Random-Access Memory), or a hard disk. The input unit 35 is configured with input settings and various operations, and includes a keyboard and a mouse.

<Operation of PET apparatus 1>
Next, the operation of the PET apparatus 1 will be described. The subject M is administered with a radiopharmaceutical labeled with a positron radioisotope (radioisotope: RI). Then, the subject M placed on the top 2 by the bed apparatus 3 is moved to an arbitrary position in the opening 4 of the gantry 5.

  The γ rays emitted from the subject are emitted uniformly in the 4π direction (omnidirectional). Each detector 9 of the multilayer detector 11 detects two γ rays radiating from the subject M in the opposite direction by 180 °, and the detector 9 detecting the γ rays outputs an electrical signal (detection signal). If the time difference between the γ rays detected by the two detectors 9 is within a preset time, the data collection unit 13 simultaneously counts as a pair of γ rays. Then, LOR data connecting the detector pairs in which two γ rays radiating in the opposite direction of 180 ° are detected is collected.

  The LOR data collected by the data collection unit 13 is transmitted to the grid data creation unit 17. As shown in FIGS. 3 and 4, the grid data creation unit 17 creates grid data 23 by counting the number of LORs that have passed through each grid obtained by dividing the space into a three-dimensional grid. That is, 1 is counted for all grids through which the LOR has passed, and this is performed for all the LOR data actually measured.

  The interpolated LOR data creation unit 19 probabilistically calculates the number of interpolated LORs that are estimated to be collected if there are extracted detector parts from the actually measured number of LORs that have passed through each grid of the grid-like data 23. Interpolating LOR data. For example, in the case of calculating the number of LORs to be interpolated indicated by the symbol f in FIG. 5, the above formula is obtained from the actually measured number of LORs of all the grids (f1, f2, f3,..., Fn) through which the LOR to be interpolated passes. From (1) and Equation (2), the number of LORs to be interpolated that is estimated to be probabilistically measurable if the extracted portion of the detector 9 is present is calculated. Assume that the number of LORs to be interpolated is calculated as (f1, f2, f3,..., Fn) = (g1, g2, g3,..., Gn). The number G of LORs to be interpolated is calculated by an integrated value, that is, G = (g1 + g2 + g3 +... + Gn). This is performed for all interpolated LORs to create interpolated LOR data. In this way, since the interpolation LOR is calculated individually for all the interpolation LORs, an accurate LOR interpolation process can be performed.

  The addition processing unit 21 adds the LOR for interpolation of the interpolation LOR data to the lattice data 23 created by the lattice data creation unit 17 and adds the result. As a result, the LOR data collected by the data collection unit 13 is interpolated by the interpolation LOR data created by the interpolation LOR data creation unit 19. Then, the image reconstruction unit 27 reconstructs an image using the interpolated LOR data obtained via the correction processing unit 25. As a result, even when a part of the plurality of detectors 9 arranged in a ring shape is extracted, a tomographic image with reduced artifacts such as when the extracted detector part is included can be acquired. it can. The acquired tomographic image is displayed on the display unit 31 or stored in the memory unit 33.

  According to the PET apparatus 1 according to the present embodiment, the data collection unit 13 collects LOR data indicating a line connecting a detector pair in which γ rays are detected, and the grid data creation unit 17 The space inside the plurality of detectors 9 arranged in a ring shape is divided into a lattice shape, and the number of LORs that have passed through each lattice is counted to create lattice data 23. Then, the interpolation LOR data creation unit 19 probabilistically calculates the number of interpolation LORs that are estimated to be collected from the number of LORs that have passed through each grid of the grid-like data 23 if there is a detector of the extracted part. Interpolation LOR data is created. As described above, since the LOR data missing from the extracted detector portion is interpolated with the interpolated LOR data calculated from the number of LORs that have passed through each lattice of the lattice-like data 23, it is generated in the tomographic image. Artifacts can be reduced.

  That is, among the detection units 9 configured by extracting a part of a plurality of detectors 9 arranged in a ring shape, if there is the extracted detector part, the missing LOR data that would have been actually measured can be measured. Can be estimated and interpolated from the LOR data. Therefore, artifacts occurring in the tomographic image can be reduced. Thereby, the present apparatus 1 including the detection unit 7 configured by extracting a part of the detector 9 arranged in a ring shape can be effectively realized.

  The detection unit 7 is an extracted detector portion of the plurality of detectors 9 arranged in a ring shape, such as a blood collection tube or the like, an object M placed on the top plate 2, light, or the like. Various things can be passed. For example, when the plurality of detectors 9 is a detection unit (detector ring) arranged in a complete ring shape, the subject M in the ring of the detection unit feels a sense of pressure. However, since the detection unit 7 is configured by extracting a part of the plurality of detectors 9 arranged in a ring shape, light and air can pass through the extracted detector part, and the subject The feeling of pressure felt by M can be reduced. In addition, when a plurality of detectors 9 are detectors (detector rings) arranged in a complete ring shape, instruments such as blood collection tubes are arranged along an opening (tunnel) of the gantry. , Can be placed from the extracted detector portion. Further, since the number of detectors 9 constituting the detection unit 7 can be reduced, the manufacturing cost of the apparatus 1 can be reduced. Furthermore, when the detection unit 7 is C-type, when a PET image of a breast is imaged, it is possible to satisfactorily image the vagina and lymph regions of the subject.

  Next, Embodiment 2 of the present invention will be described with reference to the drawings. FIG. 6 is a schematic configuration diagram of a PET apparatus according to the second embodiment. FIGS. 7A and 7B are diagrams for explaining the operation of the LOR interpolation unit. FIG. 7A shows the first LOR data collection, and FIG. 7B shows the second LOR data collection. Note that the description of the same configuration as that of the above-described embodiment is omitted.

  In the second embodiment, in the detection unit 7 configured by extracting a part of the plurality of detectors 9 arranged in a ring shape, a missing LOR that is estimated to be collected if there is an extracted detector part is obtained. Then, the detection unit 7 is rotated, and interpolation is performed with the LOR actually measured at the position of the extracted detector portion.

  The PET 40 includes a rotating mechanism 41 that rotates the detection unit 7 around the central axis h of a plurality of detectors 9 that are arranged in a ring shape and that have the extracted detector portions. The central axis h is parallel to the body axis direction of the subject M. The rotation mechanism 41 is controlled by the main control unit 29. The rotation mechanism 41 rotates so that the detector 9 is arranged at the position of the extracted detector portion. For example, when a plurality of detectors 9 arranged in a ring shape including the extracted detector portions are arranged at equal intervals, the rotation mechanism 41 may be rotated in units of intervals of the detectors 9. Good.

  Further, the PET 40 includes an LOR interpolation unit 43 that interpolates the missing LOR of the extracted detector portion. The LOR interpolation unit 43 includes a first collection memory unit 45, a second collection memory unit 46, and a weighted average processing unit 47. The first collection memory unit 45 stores LOR data collected within a preset period for the first time. Similarly, the second collection memory unit 46 stores the LOR data collected for the second time. The second collection is performed at a position rotated by the rotation mechanism 41 from the position of the first LOR data collection.

  The weighted average processing unit 47 performs an interpolation process for interpolating LOR data having a missing position of the extracted detector portion with LOR data corresponding to the position of the extracted detector portion collected by rotating the rotation mechanism 41. Do. For example, assume that LOR is collected at the rotational position of the detector 9 as shown in FIGS. 7 (a) and 7 (b). At the position of the detector 9 before and after the rotation, where there is no missing LOR data and there are two measured LOR data, the weighting is set to “1” and the two measured LOR data are averaged. Further, at the position of the detector 9 before and after rotation, where there is one measured LOR data and where there is a missing LOR, the weight of the missing LOR is set to “0” and the measured LOR weight is set to “2”. As above, the missing LOR and the measured LOR data are averaged. In FIG. 7, the weight is indicated by a symbol S. For example, at the position indicated by the symbol P, weighting S = 1. The weighted average processing unit 47 corresponds to the rotation LOR interpolation processing unit in the present invention.

<Operation of PET apparatus 40>
Next, the operation of the PET apparatus 40, particularly the operation of the LOR interpolation unit 43 will be described. The data collection unit 13 collects the first LOR data shown in FIG. The collected LOR data is stored in the first collection memory unit 45. Collection of LOR data is performed in a preset period. When the first collection of LOR data is completed, the detection unit 7 is rotated about the central axis h by the rotation mechanism 41. Then, the data collection unit 13 collects the second LOR data shown in FIG. The collected LOR data is stored in the second collection memory unit 46. The second LOR data collection is performed by being rotated by the rotation mechanism 41 so that the position corresponding to the detector portion extracted in the first collection can be collected.

  The weighted average processing unit 47 performs weighted average processing of the first LOR data collection shown in FIG. 7A and the second LOR data collection shown in FIG. 7B. For example, as shown in FIGS. 7A and 7B, since LOR data is collected at the position indicated by the symbol P, the weighting is averaged as “1”. Further, at the position indicated by the symbol Q, there is no extracted LOR data in the first collection, and there is no LOR collection data. In the second collection, the detector 9 collects LOR data. Therefore, the first weighting is “0”, and the second weighting is “2”. Thereby, the missing LOR is interpolated.

  Note that the PET apparatus 40 interpolates the missing LOR by collecting the LOR data twice, but the missing LOR may be interpolated from the LOR data three times or more.

  According to the PET apparatus 40 according to the present embodiment, the weighted average processing unit 47 is configured to rotate the rotation mechanism 41 and collect the LOR data in which the position of the extracted detector part is lost. Interpolation is performed using LOR data corresponding to. That is, by interpolating with the LOR data collected by rotating by the rotation mechanism at the same position as the extracted detector portion, artifacts occurring in the tomographic image can be reduced.

  That is, out of the detection unit 7 configured by extracting a part of the plurality of detectors 9 arranged in a ring shape, if there is a detector of the extracted part, the missing LOR data that would have been actually measured can be obtained. It is possible to estimate and interpolate from actually measured LOR data. As a result, artifacts occurring in the tomographic image can be reduced. Thereby, this apparatus provided with the detection part comprised by extracting a part of detector 9 arrange | positioned at ring shape can be made effective.

  The detection unit 7 is an extracted detector portion of the plurality of detectors 9 arranged in a ring shape, such as a blood collection tube or the like, an object M placed on the top plate 2, light, or the like. Various things can be passed. For example, when the plurality of detectors 9 is a detection unit (detector ring) arranged in a complete ring shape, the subject M in the ring of the detection unit feels a sense of pressure. However, since the detection unit 9 is configured by extracting a part of the plurality of detectors 9 arranged in a ring shape, light and air can pass through the extracted detector part, and the subject The feeling of pressure felt by M can be reduced. In addition, when a plurality of detectors 9 are detectors (detector rings) arranged in a complete ring shape, blood sampling tubes and other instruments are arranged along the opening of the gantry. It can be arranged from the detector part. Further, since the number of detectors 9 constituting the detection unit 7 can be reduced, the manufacturing cost of the apparatus 1 can be reduced.

  Next, Embodiment 3 of the present invention will be described with reference to the drawings. FIG. 8 is a schematic configuration diagram of a PET apparatus according to the third embodiment. FIG. 9 is a diagram for explaining the operation of the LOR interpolation unit, where (a) shows the first LOR data collection, and (b) shows the second LOR data collection. Note that the description of the same components as those in the above-described embodiments is omitted.

  In the third embodiment, the missing LOR is interpolated by the combination of the first and second embodiments.

  In the PET apparatus 50, the LOR interpolation unit 51 includes a first grid data creation unit 53, a first interpolation LOR data creation unit 54, and a first addition processing unit 55 that are used in the first LOR data collection. . The LOR interpolation unit 51 includes a second grid data generation unit 57, a second interpolation LOR data generation unit 58, and a second addition processing unit 59 that are used in the second LOR data collection. Details of each configuration are omitted because they overlap with the description of the grid data creation unit 17, the interpolation LOR data creation unit 19 and the addition processing unit 21 of the first embodiment.

  Further, the LOR interpolation unit 51 corresponds to the extracted detector portion position collected by rotating the LOR data interpolated by the first addition processing unit 55 of the position of the extracted detector portion by the rotation mechanism 41. A weighted average processing unit 61 that performs an interpolation process for interpolation using LOR data. The first addition processing unit 55 and the second addition processing unit 59 correspond to the probability LOR interpolation processing unit in the present invention. Further, the weighted average processing unit 61 corresponds to a rotation LOR interpolation processing unit in the present invention.

<Operation of PET apparatus 50>
Next, the operation of the PET apparatus 50, particularly the operation of the LOR interpolation unit 51 will be described. The data collection unit 13 collects the first LOR data shown in FIG. The collected LOR data is transmitted to the first grid data creation unit 53. Collection of LOR data is performed in a preset period.

  As shown in FIG. 3, the first grid data creation unit 53 creates the grid data 23 by counting the number of LORs that have passed through each of the three-dimensional grids in which the space is divided. Then, the first interpolation LOR data creation unit 54 probabilistically calculates the number of interpolation LORs that are estimated to be collected if there are extracted detector parts from the number of actually measured LORs that have passed through each grid of the grid data 23. To generate interpolated LOR data.

  The first addition processing unit 55 adds and adds the LOR to be interpolated by the interpolation LOR data to the grid data 23 created by the grid data creation unit 17. As a result, the LOR data collected by the data collection unit 13 is interpolated with the interpolation LOR data created by the first interpolation LOR data creation unit 54. The first post-interpolation LOR data subjected to the interpolation process is transmitted to the weighted average processing unit 61.

  When the first collection of LOR data is completed, the detection unit 7 is rotated about the central axis h by the rotation mechanism 41. Then, the data collection unit 13 collects the second LOR data shown in FIG. The collected LOR data is transmitted to the second grid data creation unit 57. Collection of LOR data is performed in a preset period.

  The second grid data creation unit 57, the second interpolation LOR data creation unit 58, and the second addition processing unit 59 are the first grid data creation unit 53, the first interpolation LOR data creation unit 54, and the first addition processing unit 55. The same operation as is executed. As a result, the LOR data collected by the data collection unit 13 is interpolated by the interpolation LOR data created by the second interpolation LOR data creation unit 58. The interpolated second post-interpolation LOR data is transmitted to the weighted average processing unit 61.

  The weighted average processing unit 61 performs the weighted average of the first post-interpolation LOR data shown in FIG. 9A and the second post-interpolation LOR data shown in FIG. 9B. For example, as shown in FIGS. 9A and 9B, since LOR data is collected at the position indicated by the symbol P and is not subject to LOR to be interpolated, the weight is set to “1”. Average. Further, at the position indicated by the symbol Q, the first collection corresponds to the extracted detector portion, so that it has LOR data to be interpolated as an object of LOR to be interpolated. On the other hand, in the second collection, LOR data is collected by the detector 9 and is not a target of LOR to be interpolated. Then, taking into consideration the reliability of the interpolated LOR data, the first weighting is averaged as “0.5” and the second weighting as “1.5”. Thereby, the LOR data is interpolated.

  According to the PET apparatus 50 according to the present embodiment, the extracted detector which is further collected by rotating the LOR data interpolated by the first addition processing unit 55 and the second addition processing unit 59 by the rotation mechanism 41. Since interpolation processing is performed to interpolate with LOR data corresponding to the position of the portion, artifacts occurring in the tomographic image can be further reduced.

  Next, Embodiment 4 of the present invention will be described with reference to the drawings. FIG. 10 is a schematic configuration diagram of a PET apparatus according to the fourth embodiment. FIG. 11 is a diagram for explaining an interpolation process for interpolating LOR data in real time. (A) shows actually measured LOR and interpolation LOR, and (b) shows the total number of calculated interpolation LORs. An operation for determination is shown, and (c) shows an interpolation LOR addition process. Note that the description of the same components as those in the above-described embodiments is omitted.

  In the first and third embodiments described above, LOR data is interpolated collectively for LOR data including a certain number of LORs. However, in the fourth embodiment, LOR is interpolated in real time.

  The LOR interpolation unit 71 of the PET apparatus 70 includes an interpolation LOR data creation unit 73, a determination processing unit 75, and an interpolation processing unit 77. The determination processing unit 75 and the interpolation processing unit 77 correspond to the probability LOR interpolation processing unit in the present invention.

  The interpolated LOR data creation unit 73 creates and updates interpolated LOR data by probabilistically calculating the number of interpolated LORs estimated to be collected if there is a detector part extracted every time one LOR data is collected. . That is, the interpolation LOR data creation unit 73 calculates the number of interpolation LORs for each piece of LOR data (list data).

  Assume that the data collection unit 13 collects LOR data in the order of event A, event B, event C,..., Event Z as shown in FIG. In this case, every time LOR data is collected, the interpolation LOR data creation unit 73 probabilistically calculates the number of interpolation LORs. For example, when the event A and the interpolation LOR1 that are counted simultaneously intersect at a certain grid, the solid angle of the actually detected detection plane, the solid angle of the interpolation detection plane, and the fact that one event has been collected in the grid, etc. The number of LORs to be interpolated is calculated. Note that when one piece of LOR data is collected, the probability (number) of interpolation LORs in each lattice is known in advance. For this reason, a conversion table may be provided to convert the collected LOR data into the number of interpolation LORs.

  As shown in FIG. 11B, the interpolation LOR data includes, for example, 100 interpolation LORs (interpolation LOR1, interpolation LOR2, interpolation LOR3,..., Interpolation LOR100). Each time one piece of LOR data is collected, the number of interpolated LORs estimated to be collected if there is a detector of the extracted part is calculated probabilistically, and the number of interpolated LORs calculated stochastically is calculated. Accumulated. In this way, the interpolation LOR data is created / updated.

  The determination processing unit 75 determines whether the integrated value of the number of interpolation LORs of the generated interpolation LOR data is equal to or greater than a preset value. When the integrated value of the number of interpolation LORs is equal to or greater than a preset value, the interpolation processing unit 77 performs interpolation processing by adding the interpolation LOR to the LOR data collected by the data collection unit 13. That is, for example, when the number of interpolation LOR data obtained by integrating the interpolation LOR1 exceeds one when the event Y is collected, the determination processing unit 75 determines that there is one interpolation LOR1. Then, the determination processing unit 75 notifies the interpolation processing unit 75 that the interpolation LOR1 is to be interpolated. The interpolation processing unit 77 interpolates the LOR between the event Y and the event Z as shown in FIG. 11C based on the notification from the determination processing unit 75 that the interpolation LOR1 is to be interpolated.

<Operation of PET apparatus 70>
Next, the operation of the PET apparatus 70, particularly the operation of the LOR interpolation unit 71 will be described. The data collection unit 13 collects LOR data such as event A. For each piece of LOR data collected, the grid data creation unit 73 probabilistically calculates the number of interpolation LORs that are estimated to be collected if there are detectors of the extracted portions. To do. Then, the interpolation LOR data creation unit 73 accumulates the number of interpolation LORs calculated stochastically every time LOR data is collected in each grid through which the interpolation LORs 1 to 100 to be interpolated pass (FIG. 11 b).

  The determination processing unit 75 determines whether the integrated value of the number of interpolation LORs calculated stochastically is a preset value (for example, one) or more in each lattice through which the interpolation LORs 1 to 100 pass. When the integrated value exceeds one, the determination processing unit 75 notifies the interpolation processing unit 77 of that. Based on the notification, the interpolation processing unit 77 adds one interpolation LOR after the LOR data when the number exceeds one. Thereby, the missing LOR is interpolated. Note that the determination processing unit 75 may subtract 1 from the integrated value, for example, when the integrated value exceeds one and notifies the interpolation processing unit 77 of that. In addition, the reconstruction processing unit 27 reconstructs an image in real time and acquires a tomographic image every time LOR data is collected.

  According to the PET apparatus 70 according to the present embodiment, in the case of data collection that varies with time, if LOR interpolation is performed collectively after data collection, artifacts may not be reduced satisfactorily depending on the amount of variation. However, since the determination processing unit 75 and the interpolation processing unit 77 interpolate the interpolation LOR data when the integrated value of the interpolation LOR data calculated every time LOR data is collected is equal to or greater than a preset value, LOR interpolation can be performed.

  Next, Embodiment 5 of the present invention will be described with reference to the drawings. FIG. 12 is a schematic configuration diagram of a positron emission tomography apparatus (PET apparatus) according to the fifth embodiment, where (a) shows the multilayer detection unit, and (b) is a detection seen from the A direction of (a). An example of the arrangement of the vessel is shown. Note that the description of the same components as those in the above-described embodiments is omitted.

  In the first embodiment described above, the C-shaped detection unit 7 configured by extracting a part of the plurality of detectors 9 arranged in a ring shape is provided, and the missing LOR is interpolated. It is not limited to.

  That is, the PET apparatus 80 according to the present embodiment is configured by extracting a part of the detection units 81 from the multilayer detection unit (detector ring) 81 including a plurality of detectors 9 arranged in a ring shape. The multilayer detector 83 is provided. That is, the multi-layer detection unit 83 has no extracted detector portion in the direction perpendicular to the body axis of the subject M, and is a detection unit 81 configured in multiple layers in the body axis direction of the subject M. It is the structure where a part of was extracted.

  The PET apparatus 80 may include a therapeutic radiation irradiation unit 85 and irradiate the subject M with radiation through the detection unit 81 extracted from the therapeutic radiation irradiation unit 85 from the multilayer detection unit 83.

  According to the PET apparatus 80 according to the present embodiment, the detection unit (detector ring) 81 in which a plurality of detectors 9 are arranged in a ring shape is configured in multiple layers in the body axis direction of the subject M. (Multi-layer detection unit 83) Even if a part of the detection unit 81 configured in multiple layers is extracted, a tomographic image with reduced artifacts can be acquired.

  The present invention is not limited to the above embodiment, and can be modified as follows.

  (1) In the above-described first to fourth embodiments, the C-shaped detection unit 7 configured by extracting a part of the plurality of detectors 9 arranged in a ring shape is provided. It is not limited to the configuration of the mold. For example, the detection unit may have a configuration in which the O-type in FIG. 13A is divided into two or a configuration in which the O-type in FIG. 13B is divided into three.

  (2) In the fifth embodiment described above, a part of the detection units 81 are extracted from the multi-layered detection unit (detector ring) 81 including a plurality of detectors 9 arranged in a ring shape. However, it is not limited to this configuration. For example, the extracted detection part may be constituted by a C-type detection part or a detection part of the modification (1).

  (3) In each Example mentioned above, although the detector 9 was comprised by the detector block, it is not limited to this structure. For example, it may be a semiconductor camera in which the detection element is made of cadmium telluride (CdTe) or the like, which is a semiconductor, and detects charges generated by incidence of γ rays.

  (4) In the first and third embodiments described above, the grid data creation unit 17 creates the grid data 23, and the interpolation LOR data creation unit 19 interpolates from the number of LORs that have passed through each grid of the grid data 23. The number of LOR is calculated. However, as shown in FIG. 11B, the LORs may be interpolated collectively based on the number of integrated interpolation LORs.

  (5) In the above-described fourth embodiment, the image reconstruction unit 27 inserts interpolated LOR data created by calculation from list data (one LOR data) into the list data, and from the interpolated list data. A tomographic image is acquired by image reconstruction. However, in FIG. 10, the grid data creation unit of the first embodiment may be provided between the data collection unit 13 and the interpolation LOR data creation unit 73. In this case, the grid data creation unit creates grid data 23 every time one LOR data is collected, and the interpolation LOR data creation unit 73 determines the number of LORs that have passed through each grid of the grid data 23. From the above, the number of interpolated LORs estimated to be collected if there are extracted detector parts is calculated stochastically, and interpolated LOR data is created and updated. That is, the grid data 23 may be created every time one piece of LOR data is collected, or the LOR to be interpolated from the grid data 23 may be calculated.

DESCRIPTION OF SYMBOLS 1,40,50,70,80 ... PET apparatus 7,81 ... Detection part 9 ... Detector 11, 83 ... Multilayer detection part 13 ... Data collection part 15,43,51,71 ... LOR interpolation part 17 ... Lattice-like data Creation unit 19, 73 ... Interpolation LOR data creation unit 21 ... Addition processing unit 23 ... Grid-like data 29 ... Main control unit 41 ... Rotation mechanism 45 ... First collection memory unit 46 ... Second collection memory unit 47, 61 ... Weighted average Processing unit 53 ... 1st grid data creation unit 54 ... 1st interpolation LOR data creation unit 55 ... 1st addition processing unit 57 ... 2nd grid data creation unit 58 ... 2nd interpolation LOR data creation unit 59 ... 2nd addition Processing unit 75 ... Determination processing unit 77 ... Interpolation processing unit

Claims (8)

A detection unit configured to extract a part of a plurality of detectors arranged in a ring shape and detect pair annihilation photons emitted from a subject to which a radiopharmaceutical is administered;
A data collection unit for collecting LOR data indicating a line connecting a detector pair in which the pair annihilation photons are detected;
Interpolation LOR data creation unit for creating interpolated LOR data out calculate the number of interpolation LOR be estimated that the detectors portion withdrawn is collected if,
An LOR interpolation processing unit that adds the interpolation LOR data calculated by the interpolation LOR data creation unit to the LOR data collected by the data collection unit;
Equipped with a,
The interpolated LOR data creation unit counts the number of LORs that have passed through each grid in a space obtained by dividing the inside of the plurality of detectors into a grid based on the LOR data collected by the data collection unit. , Interpolating each grid based on the counted number of LORs passing through each grid, the angle of the measured detection surface of each grid, and the angle of the detection surface of the extracted detector portion in each grid Calculate the number,
The interpolation LOR data creation unit calculates the number of interpolation LORs by accumulating the calculated number of interpolations of each grid in each of the interpolation LORs passing through each grid of the space divided into the grid. A positron emission tomography apparatus characterized in that the interpolation LOR data is created . The positron emission tomography apparatus according to claim 1,
A grid data creation unit that creates a grid data by dividing the space inside the plurality of detectors arranged in a ring shape of the detection unit into a grid and counting the number of LORs that have passed through each grid. Prepared,
The interpolated LOR data creation unit calculates the estimated total number of LORs that pass through each grid from the number of LORs that have passed through each grid of the grid-like data, the angle of the measurement detection surface of each grid, and all angles, and this estimation A positron emission tomography apparatus, wherein interpolation LOR data is created by calculating the number of interpolation of each grid from the total number, the angle of the interpolation detection surface of the extracted detector portion, and the total angle. The positron emission tomography imaging apparatus according to claim 1 or 2,
A rotation mechanism for rotating the detection unit around the central axis of a plurality of detectors arranged in a ring shape;
Interpolation that interpolates LOR data of the extracted detector portion position interpolated by the LOR interpolation processing unit with LOR data corresponding to the position of the extracted detector portion collected by rotation by the rotating mechanism. A rotational LOR interpolation processing unit for processing;
A positron emission tomography apparatus comprising: The positron emission tomography imaging apparatus according to any one of claims 1 to 3,
The interpolation LOR data creation unit calculates the number of interpolation LORs every time the LOR data is collected , and calculates the interpolation LOR data by calculating the number of interpolations of the calculated grids .
The positron emission tomography apparatus according to claim 1, wherein the LOR interpolation processing unit performs interpolation processing on the interpolation LOR data when the calculated integrated value of the interpolation LOR data is equal to or greater than a preset value. In the positron emission tomography apparatus according to claim 3,
The positron emission tomography apparatus according to claim 1, wherein the interpolation process is a weighted average. In the positron emission tomography apparatus according to any one of claims 1 to 5 ,
2. The positron emission tomography apparatus according to claim 1, wherein the detection unit is a C type. The positron emission tomography imaging apparatus according to claim 1 or 2,
2. The positron emission tomography apparatus according to claim 1, wherein the detection unit is configured by extracting a part of a plurality of detectors arranged in a ring shape and in multiple layers. In the positron emission tomography apparatus according to any one of claims 1 to 7 ,
A positron emission tomography apparatus comprising an image reconstruction unit that reconstructs an image using interpolated LOR data.

Images