JP5585094B2 - Radiation position detector position calculation method and apparatus - Google Patents

Description

  The present invention relates to a radiation position detector in which a light receiving element is optically coupled to two or more adjacent surfaces of a scintillator block having a substantially rectangular parallelepiped outer shape, which emits light when absorbing radiation. In particular, the present invention relates to a position calculation method and apparatus for a radiation position detector for calculating a radiation absorption position by calculating an output signal of each light receiving element, and in particular, a cubic or cuboid scintillator element that emits light when absorbing radiation is three-dimensionally arranged. The present invention relates to a position calculation method and apparatus for a radiation position detector suitable for use in a depth-of-interaction (DOI) detector in which a plurality of light receiving elements are optically coupled to the surface of a scintillator block. Here, the optically discontinuous region is, for example, a region where the refractive index is different from the surroundings, a region that scatters light, as described in JP2009-270971, in addition to the joint portion between the divided scintillator elements. It means a surface-like or dot-like region in which light changes its traveling direction or speed, such as a region constituting a diffractive lens.

  Conventionally, a photomultiplier tube (PMT) has been used as a light receiving element for a PET detector. When the PMT is located on the subject side (upper surface of the scintillator) when incorporated in the PET apparatus, it becomes a scatterer for radiation detection, and when combined with the side surface of the scintillator, the area where radiation cannot be detected increases and the sensitivity of the PET apparatus Therefore, as shown in FIG. 1A, the PMT 12 is coupled only to the bottom surface of the scintillator (scintillator block in which fine scintillator elements are arranged in the figure) 10. Two-dimensional positioning of the place where the radiation in the scintillator block 10 is absorbed is performed by connecting a plurality of PMTs or position-discriminating type PMTs (PS-PMT) 12 to the bottom surface and performing an anger calculation of the signals. As illustrated in FIG. 1B, a response corresponding to the absorbed position appears on the two-dimensional (2D) position histogram representing the result of the anger calculation, but does not have an optically discontinuous region inside. When using a scintillator block in which fine scintillator elements are arranged instead of a single scintillator block, a response of each scintillator element appears discontinuously, and it is convenient that the position of each scintillator element can be easily determined.

  Under the condition that the PMT is bonded only to the bottom surface of the scintillator, various measures have been made on the scintillator block in order to obtain a position (DOI information) in the depth direction with respect to the light receiving element. However, in recent years, semiconductor light-receiving elements such as avalanche photodiodes (APD) and Geiger mode APDs (product names are also called Si-PM, MPPC (Multi-Pixel Photon Counter), etc.) have made rapid development. Research has been made on PET detectors in which PS-PMT is replaced with semiconductor light-receiving elements. A small and thin semiconductor light-receiving element can have a new detector structure, and a semiconductor light-receiving element having a small volume does not become a scatterer even when arranged on the upper surface of the detector. 2 (a) and 2 (b), the light receiving elements (in FIG. 2A, the photodiode 14 on the upper surface side and the PS-PMT 12 on the lower surface (bottom surface) side are provided on the upper and lower surfaces of the scintillator block 10. In FIG. 2B, the position discrimination type APD 16) is combined on both the upper and lower surfaces, and the DOI detection method (see Non-Patent Documents 1 and 2) for obtaining DOI information by the ratio of those signals, or FIG. As shown, a DOI detector that couples a light receiving element (APD 16) to the side surface and identifies the position in the DOI direction from the signal has also been studied (see Non-Patent Document 3). In the method of connecting the light receiving element to the side surface as shown in FIG. 2C, not only the detection position in the APD 16 becomes the DOI information as it is, but also the scintillator light is efficiently obtained because the wide surface of the scintillator is coupled to the light receiving element. Although the loss of the light amount is small, the packing fraction when the PET apparatus is made by the amount of the light receiving element is reduced.

Further, as shown in FIG. 2D, the inventors have studied a DOI detector in which a plurality of semiconductor light receiving elements 18 are optically coupled to the surface of a scintillator block in which cubic or cuboid scintillator elements are three-dimensionally arranged. (See Patent Documents 1 and 2 and Non-Patent Document 4). That is, assuming that the three directions along the side of the scintillator block 10 are the x-axis, y-axis, and z-axis, the light receiving elements are arranged on the xy plane, the xz plane, and the yz plane, respectively. An x-axis component, a y-axis component, and a z-axis component are determined. A detector structure in which the scintillator block is composed of a single scintillator element that does not have one optically discontinuous region inside, and a light receiving element that is not a position discrimination type is arranged on each of the xy plane, the xz plane, and the yz plane. For this, a method for specifying a radiation detection position has been considered by another group (see Non-Patent Document 5), but no experimental data has been published yet.

  When the scintillator block in which the light receiving element is arranged is configured by a three-dimensional array of small scintillator elements, as shown in FIG. 3A, the scintillation light emitted by a scintillator element absorbing radiation is Due to the optical discontinuity characteristic with the scintillator element, there is a tendency that the scintillator element spreads widely in a row including the scintillator element (in the case where the scintillator element is a cube, three rows in the front and rear, left and right, and up and down directions). In addition, the amount of light obtained from the scintillator elements at both ends of the column largely depends on the position of the light emitting point, but when the surface of the scintillator element is in a mirror state, the attenuation of the amount of light during propagation is small. Even in the case of scintillation light emitted from an element, the difference between a signal obtained from a nearby light receiving element and a signal obtained from the opposite light receiving element is not large (see Non-Patent Documents 6 and 7). FIG. 3B shows an example of the scintillator block 10 'arranged in a line. If the light receiving elements 18 optically coupled to both ends of the scintillator are A and B, the radiation irradiation position is received from the light receiving element B side of the scintillator. When shifted to the element A side, as shown in FIG. 3C, the received light amounts of A and B increase and decrease in proportion to the distance to the irradiation position, but the scintillator surface shown on the left side of FIG. In the case of the mirror surface, unlike the case of the rough surface shown on the right side, the output difference between the light receiving elements A and B on both sides is small even when the end of the scintillator is irradiated. Since the radiation detection position is specified by the ratio of the signal output of the light receiving elements A and B, if the amount of change in the signal due to the irradiation position is small, the accuracy of position identification deteriorates. That is, the position resolution of the detector is deteriorated. Here, the substance between the scintillator elements is air, an optical adhesive, or the scintillator element itself, but a region having a different refractive index from the surroundings, a region that scatters light, as described in JP-A-2009-270971. An area constituting a diffractive lens, etc., refers to an optically discontinuous area in the form of a plane or a spot where light travels in a different direction or changes its speed.

JP 2004-279057 A JP 2009-121929 A

JSHuber, WWMoses, MS Andreaco, and O. Petterson, “An LSO scintillator array for a PET detector module with depth of interaction measurement,” IEEE Trans. Nucl. Sci., Vol. 48, No. 3, pp. 684 -688, June 2001. Y. Shao, RW Silverman, R. Farrell, L. Cirignano, R. Grazioso, KS Shah, G. Visser, M. Clajus, TO Tumer, and SR Cherry, “Design strudies of a high resolution PET detector using APD array, ”IEEE Trans. Nucl. Sci., Vol. 47, No. 3, pp. 1051-1057, June 2000. CS Levin, “Design of a high-resolution and high-sensitivity scintillation crystal array for PET with nearly complete light collection,” IEEE Trans. Nucl. Sci., Vol. 49, No. 5, pp. 2236-2243, October 2002 . Y. Yazaki, H. Murayama, N. Inadama, A. Ohmura, H. Osada, F. Nishikido, K. Shibuya, T. Yamaya, E. Yoshida, T. Moriya, T. Yamashita, H. Kawai, “Preliminary study on a new DOI PET detector with limited number of photo-detectors, ”The 5th Korea-Japan Joint Meeting on Medical Physics, Sept 10-12, 2008, Jeju, Korea, YI-R2-3, 2008. J. W. LeBlanc and R..A. Thompson, “A novel PET detector block with three dimensional hit position encoding” IEEE Nuclear Science Symposium Conference Record, J1-2, Portland, Oregon, 2003. T. Umehara, H. Murayama, T. Omura, H. Ishibashi, H. Kawai, N. Inadama, T. Kasahara, N. Orita, and T. Tsuda, “Basic study on pulse height of distribution of DOI detectors constructed of stucked crystal element, ”IEEE Nuclear Science Symposium Conference Record, M10-29, Norfolk, Virginia, 2002. Y. Shao, K. Meadors, RW Silverman, R. Farrell, L. Cirignano, R. Grazioso, KS Shah, and SR Cherry, “Dual APD array readout of LSO crystals: optimization of crystal surface treatment,” IEEE Trans. Nucl Sci., Vol. 49, No. 3, pp. 649-654, June 2002.

As illustrated in FIG. 1, in the conventional detector in which the light receiving element 12 is coupled to one surface of the scintillator block 10, the radiation detection position is specified by a plurality of light receiving elements or position discriminating type PMT (PS-PMT ) Is performed on a 2D position histogram representing the result of position calculation. As shown in FIG. 4A, when scintillation light is detected by a plurality of light receiving elements, position calculation is usually performed by anger calculation for each direction. For example, position calculation in the x-axis direction is expressed by the following expression. It is. Here, it is assumed that the light receiving points (the position of the light receiving element or the position of the anode of the PS-PMT) are n columns (n = 4 in FIG. 4) in the x-axis direction.
Wxi: Coefficient applied to the light receiving point signal in the i-th column when calculating the position in the x direction Sxi: Sum of signals from the light receiving points in the i-th row in the x direction

  When the weighting is proportional to the distance from the origin of x of the light receiving point, the calculation result has a distribution corresponding to the position of the light receiving point. As shown in FIG. 4 (b), the same calculation is performed for the y-axis direction, and the result is expressed as x-axis and y-axis coordinates, which is a 2D position histogram. Therefore, when the scintillator block is composed of a small scintillator element array, a distribution corresponding to each scintillator element appears on the 2D position histogram as shown in FIG. This individual distribution is called an element response. Here, if the element responses overlap, it becomes impossible to distinguish between the corresponding elements, and the detection position resolution of the element size cannot be maintained. The discriminating ability of the detected element, that is, the discriminating ability of the detection position is good. Therefore, it is preferable that the element responses are arranged uniformly on the 2D position histogram.

  As described above, in the conventional detector, the position of the light receiving point is two-dimensional (x, y), but the light receiving point is three-dimensional by the arrangement of the light receiving element as in the DOI detector being developed by the inventors. Consider a distributed detector. As shown in FIG. 5 (a), when the scintillator block 10 is composed of a lump scintillator that does not have an optically discontinuous region inside, it includes light receiving elements on both end surfaces orthogonal to the x-axis as in the conventional method. When all the obtained light receiving element signals x1 to x4 are used and an anger calculation corresponding to the position of the light receiving point is performed, the light receiving elements (x1 and x4) are removed from both end surfaces orthogonal to the x axis, and black paper is used instead. However, in the case of a scintillator block composed of scintillator elements, if all signals are used, separation of element responses on the 2D position histogram may be worsened. I understood.

  When the light receiving points are distributed three-dimensionally, in a detector composed of a single scintillator in which the scintillator block does not have an optically discontinuous region inside, the radiation detection position is specified in the x-axis direction and is orthogonal to the x-axis. As shown in FIG. 5B, the light receiving element on the surface may be considered to be equivalent to receiving light that should be received when arranged on only one surface (bottom surface) of the scintillator 10.

Hereinafter, the configuration of the scintillator block and the light receiving element will be examined. Here, it is assumed as follows and this is referred to as an ideal condition.
(1) The refractive index of a scintillator is large like a BGO scintillator.
(2) The scintillator surface is a mirror surface.
(3) There is no light absorption or scattering in the scintillator.

  Now, as shown in FIG. 6A, a single scintillator block (also referred to as a monolithic) having no optically discontinuous region inside, and a small scintillator element as shown in FIG. Consider the case of a three-dimensional array in which a block is configured by three-dimensionally arraying on a plane discontinuous surface. Here, the six surfaces of the block 10 are covered with the position sensitive light receiving element 20.

  The light emission state in the scintillator is shown in comparison with FIGS. 7 (a) and 7 (b). In the case of the monolithic shown in FIG. 7A, since the scintillation light spreads widely, the D2 signal from the light receiver is insensitive to the position information in the x-axis direction, particularly when the light emitting point is away from the light receiver, and the x-axis The coordinates are determined using the intensity ratio of the D1 signal and the D3 signal in addition to the D2 signal. On the other hand, in the case of the three-dimensional array shown in FIG. 7B, under ideal conditions, along a column including internal scintillator elements (when the scintillator element is a cube, three columns in the front, rear, left, and right directions). Thus, the scintillation light is transmitted linearly without spreading. That is, since the D1 signal and the D3 signal are insensitive to position information in the x-axis direction, the x-axis coordinates are determined using the D2 signal and the D4 signal.

That is, under ideal conditions, there are the following differences between monolithic and three-dimensional arrays (see Non-Patent Document 5).
(1) Monolithic calculation is performed mainly by light receiving element signals orthogonal to each coordinate.
(2) In a three-dimensional array, calculation is performed by the light receiving element in the axial direction for each coordinate.
(3) That is, the monolithic and the three-dimensional array are both extreme position calculations.

In contrast, the three-dimensional array has the following differences as shown in FIG.
(1) Light is distributed to light receiving elements other than the column to which the scintillator elements belong.
(2) How light is distributed depends on the arrangement of the scintillator elements.
(3) A mixture of a three-dimensional array and monolithic under ideal conditions.

The advantages of the three-dimensional array are as follows.
(1) Since discontinuous counting peaks corresponding to individual scintillator elements can be obtained by uniform radiation irradiation, position calibration is easy and maintenance is easy.
(2) The count of the count peak is physically standardized by the volume of the scintillator element.
(3) Deterioration of performance against fluctuations in the electronic circuit system can be suppressed to a relatively low level.
(4) When the light receiver covering the six surfaces is changed to a mixed two-dimensional array of a large number of small light receiving elements and reflectors, the center position discrimination performance is greatly increased due to the scattered light from the reflectors in the monolithic case. Although it is considered to decrease, in the three-dimensional arrangement method, the direction of reflected light is limited by the optical discontinuity of the three-dimensional arrangement, so that the discrimination performance of the light emitting element is not greatly affected.

  For simplicity, it is assumed that one light receiving element is coupled to each crystal element in a 6 × 6 × 6 three-dimensional crystal element array. When performing the anger calculation with respect to the x direction, as shown in FIG. 9A, the light receiving elements on the xy plane and the xz plane are classified into x1 to x6 as the positions of the light receiving points in the x axis direction. The output signals of the light receiving elements on the yz left surface and the yz right surface at both ends of the yz plane orthogonal to the x axis direction are positional information corresponding to x1 or x6, respectively, and are therefore included in x1 or x6, respectively. I think. As shown in FIG. 3A, for example, the scintillation light emitted from the scintillator element A in FIG. 9A tends to spread along the scintillator array. If the scintillation light spreads by 1/3 each in the x, y, and z axis directions, as shown in FIG. 9 (b), in addition to x5 where the scintillator element A is located, A signal is also output to irrelevant x1. The result of the position calculation in that state is close to the center and deteriorates the element discrimination ability in the x-axis direction. That is, the position resolution corresponding to the scintillator element size by the light propagating from the scintillator element A in the y-axis direction (vertical direction in the figure) and the z-axis direction (front-back direction in the figure) by controlling the light spread by the substance between the scintillator elements. However, the position resolution obtained by the ratio of the light propagating in the x-axis direction (left and right direction in the figure) tends to be inferior as described above, so that the position calculation using all signals is inferior in resolution. It will be.

  Therefore, as shown in FIG. 10, in the position calculation, by deliberately excluding the signals of the light receiving elements coupled to the two yz planes at both ends of the x axis direction that receive the light propagating in the x axis direction, Deterioration can be prevented. More clearly, the detection is performed by the scintillator element B at the end as shown in FIG. 9C. When the signal of x1 is used for position calculation, the response is drawn to the x1 side, and the scintillator element A of the adjacent column Separation from response is poor. Therefore, when there are only output signals from both end faces x1 and x6 in this way, the signal of the light receiving element on the left side of the yz with a small output signal is excluded.

  Conventionally, in order to accurately determine the radiation detection position, it has been common knowledge to use all of the obtained signals, but in the method of the present invention, the three-dimensional positioning is performed at the time of axial positioning. It is characterized in that the calculation is performed without using the signal of the part. However, since signals that are not used differ depending on the calculation in each axis direction, there is no unnecessary light receiving element signal, and all signals are used for position calculation in two directions. Depending on the energy of the radiation to be detected, the type of scintillator, the optical conditions of the scintillator and the entire scintillator block, etc., it is possible to leave a signal value by multiplying a low value as a weighting factor without completely excluding it.

The present invention has been made on the basis of the above knowledge, and the outer shape having at least one optically discontinuous region that emits light when absorbing radiation is formed on two or more adjacent surfaces of a substantially rectangular parallelepiped scintillator block. In the position calculation method of the radiation position detector that specifies the radiation absorption position by calculating the output signal of each light receiving element of the radiation position detector optically coupled to the light receiving element, the light receiving element is located at the end in the axial direction where the position is to be specified. In some cases, the output signal of the light receiving element on the surface intersecting the axial direction is excluded .

Further, when position calculation, the x-axis direction orthogonal to each other, y-axis, the z-axis direction, during the positioning of the x-axis direction, the output signal of the light receiving element on the yz plane in at least one of x-axis direction end When the positioning in the y-axis direction is excluded, the output signals of the light receiving elements on the xz plane at at least one of both ends in the y-axis direction are excluded, and in the positioning in the z-axis direction, the output signals at at least one of both ends in the z-axis direction are excluded. The output signal of the light receiving element on the xy plane can be excluded.

The present invention also provides a radiation position in which a light receiving element is optically coupled to two or more adjacent surfaces of a scintillator block having a substantially rectangular parallelepiped shape that emits light when absorbing radiation. In the position calculation device of the radiation position detector that calculates the output signal of each light receiving element of the detector and specifies the radiation absorption position, if there is a light receiving element at the end in the axial direction for which the position is to be specified, there is provided a position calculation apparatus of the radiation position detector with means to exclude the output signal of the light receiving element on the intersecting plane.

Hereinafter, the six surfaces of the scintillator block that are parallel to the xy surface are the xy upper surface, the xy lower surface, the two surfaces parallel to the xz surface are the xz rear surface, the xz front surface, the two surfaces parallel to the yz surface are the yz right surface, and the yz left surface. Here, the description will be made on the assumption that the light receiving elements are arranged on the six surfaces. The light receiving points (the positions of the light receiving elements or the positions of the anodes of the PS-PMT) on each surface are assumed to be in n _A j columns with respect to the j (j = x, y, z) direction (A = 1 (xy upper surface). ), 2 (xy lower surface), 3 (xz rear surface), 4 (xz front surface), 5 (yz right surface), 6 (yz left surface)).

In the present invention, the radiation position is obtained from the light receiving element signal groups S _A on the surface A j (j = x, y , z) , when the function representing the position calculation of the j direction and f _j (),
_{j = f j (C j,} 1 × S 1, C j, 2 × S 2, C j, 3 × S 3, C j, 4 × S 4, C j, 5 × S 5, C j, 6 × S ₆₎
It becomes.

Here, C _{j, A} is a weighting coefficient for the A plane when position calculation in the j direction is performed, and the greatest feature is that the weighting coefficient for each plane differs depending on the position direction j to be obtained.

  Examples of weighting factors for each surface are shown below.

For example, when the position calculation in the x direction is performed with the light receiving element arrangement shown in FIG. 11, if the signal of the surface orthogonal to the x axis is not used, the weighting coefficient for each surface is
C _{x, 1} = 1, C _{x, 2} = 1, C _{x, 3} = 1, C _{x, 4} = 1, C _{x, 5} = 0, C _{x, 6} = 0
It becomes. Similarly, if the position calculation in the y direction does not use the signal of the plane orthogonal to the y axis, and the position calculation in the z direction does not use the signal of the plane orthogonal to the z axis,
C _{y, 1} = 1, C _{y, 2} = 1, C _{y, 3} = 0, C _{y, 4} = 0, C _{y, 5} = 1, C _{y, 6} = 1
C _{z, 1} = 0, C _{z, 2} = 0, C _{z, 3} = 1, C _{z, 4} = 1, C _{z, 5} = 1, C _{z, 6} = 1
It becomes.

Although f _j () does not matter, when a general anger calculation method is used, the position calculation according to the present invention is expressed as follows.
W _A ji: Coefficient applied to the light receiving point signal in the i-th column when calculating the position in the j-direction on the surface A S _A ji: Sum of signals of light receiving points in the i-th row in the j-direction on the surface A

For the four-sided light receiving element to be used, the number n of rows is
n ₁ x = 5, n ₂ x = 4, n ₃ x = (4) (not shown in the figure), n ₄ x = 4
It becomes.

Therefore, for example, when W is a coefficient that proportionally represents the position of the light receiving element, the position x in the x direction is calculated by the following equation.

Here, S _N is the output of the light receiving element number N shown in FIG. In FIG. 12, it is assumed that the invisible xz rear surface (A = 3) and the yz left surface (A = 6) have light receiving element arrangements of 4 × 4 and 2 × 2, respectively.

  In the case where the scintillator block is composed of a lump of scintillators that do not have optically discontinuous regions inside, the light receiving element signal that has spread the scintillation light affects the position calculation. As the distance from the light emitting position increases, the light intensity decreases and the statistical error increases. By configuring the scintillator block with a scintillator element or equivalent optical conditions, the spread of light is suppressed, and a signal with little statistical error is obtained. Conventionally, even if the scintillator block does not have an optically discontinuous region in the three-dimensional arrangement of the light receiving elements, the same calculation as in the case of a lump scintillator has not been utilized, but the method of the present invention As a result, the radiation detection position identification performance was improved.

  The present invention is only based on a new idea of differently weighting or selecting an output signal of a light receiving element depending on a desired position and direction, and without introducing a new technique or complicating processing. The discrimination ability can be improved.

The technique of the present invention can also be applied to a radiation detector in which light receiving elements other than semiconductor light receiving elements such as PS-PMT are arranged in a three-dimensional array.
Further, the effect of the method of the present invention is not limited to position calculation by anger calculation, but can be applied to various radiation position calculation methods such as position calculation method by maximum likelihood estimation.

(A) A perspective view of an example of a conventional detector using a scintillator block, and (b) a diagram showing its two-dimensional (2D) position histogram. Perspective views showing various examples of conventional DOI detectors (A) A perspective view showing the light spread in a three-dimensional array of scintillators, (b) a diagram showing light propagation in a series of scintillators, and (c) a case where the surface state of the scintillator is a mirror surface or a rough surface The figure which compares and shows the light reception quantity of the light receiving element of both ends Illustration of anger calculation Explanatory drawing about light reception when a scintillator block is constituted by one scintillator The figure which compares and shows the detector in case the scintillator block is (a) monolithic and (b) three-dimensional arrangement Similarly, the figure which compares the light emission inside the scintillator Similarly, the figure which shows the state of the light of an actual three-dimensional arrangement The figure which shows the position calculation at the time of making a light receiving element couple | bond with the whole surface of the scintillator block comprised by the three-dimensional arrangement | sequence of a fine scintillator element The figure which similarly shows the example of the position calculation by this invention Explanatory drawing which similarly shows an example of the position calculation method by this invention Explanatory drawing which similarly shows the specific example of the position calculation method by this invention The figure which shows the (a) structure of a detector and (b) position calculation method in 1st Embodiment of this invention. The figure which shows the structure of the detector in 2nd Embodiment of this invention. (A) Comparison of 2D position histograms (simulations) when using all signals for position calculation (conventional) and (b) using selected signals (present invention) (A) When using all signals for position calculation (conventional) and (b) using selected signals (present invention), response profiles and peak-to-valley values at the center and end scintillator element arrays (C) and (c) are diagrams showing the peak positions of the central response profiles shown in (a) and (b) combined with each other. Diagram showing experimental setup FIG. 8 is a diagram showing a comparison of 2D position histograms obtained by experiments in which (a) all signals are used for position calculation (conventional) and (b) a selected signal is used (invention).

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  As shown in FIG. 13A, the first embodiment of the present invention is a DOI detector in which a plurality of light receiving elements 20 are optically coupled to the entire surface of a scintillator block 10 in which cubic scintillator elements are three-dimensionally arranged. As shown in the flow in (b), when calculating the position in the x-axis direction, data on the yz right surface and yz left surface, which are yz planes at both ends of the x-axis direction, are excluded (step 100), and the x-axis direction is excluded. Is calculated (step 102), and when calculating the position in the y-axis direction, data on the xz rear surface and the xz front surface, which are xz planes at both ends in the y-axis direction, are excluded (step 110), and the y-axis is calculated. When calculating the position y in the direction (step 112) and calculating the position in the z-axis direction, the data on the xy upper surface and the xy lower surface, which are both end surfaces in the z-axis direction, are excluded (step 120), and the z-axis direction By calculating (step 122), it is obtained by the position (x, y, z) to determine (step 130) as.

  The light receiving elements may be sparsely arranged as in the second embodiment shown in FIG.

The method of the present invention was implemented with a DOI detector having light receiving elements disposed on the entire surface, and verified by both computer simulation and experiment. LGSO crystals of 3.0 × 3.0 × 3.0 mm ³ (Lu _1.8 Gd _0.2 SiO ₅ : Ce, manufactured by Hitachi Chemical Co., Ltd.) are arranged in 6 × 6 × 6, and FIG. Light receiving portions were provided at the nine positions shown. Radiation from a ¹³⁷ Cs radiation source is uniformly irradiated from three directions of x, y, and z, and three-dimensional position calculation is performed using the obtained light receiving element signals from a total of 54 locations to obtain 2D on a specific cross section. The element discrimination ability was evaluated on the position histogram. Except for the light receiving element, it was covered with a reflective material. The space between the scintillator elements in the array was air, and a specular reflection film having a reflectance of 98% was used as the reflector.

  FIGS. 15A and 15B show the results of computer simulation. In the case of (a) calculation using all light receiving element signals (conventional) and (b) x-axis direction positioning, x When the signals of the light receiving elements on the yz plane at both ends in the axial direction are excluded, and the signals of the light receiving elements on the xz plane at both ends in the y axis direction are excluded during positioning in the y axis direction (the present invention) 2D is a 2D position histogram for the xy plane. By excluding the selected signal, it can be seen that the element discrimination ability is restored as shown in FIG.

  FIGS. 16A and 16B are response profiles of the scintillator element arrays at the center and end of FIG. 15A (using all signals) and FIG. 15B (selecting signals), respectively. It is also shown that the peak-to-valley value is lowered by selecting a signal, and the discrimination ability of the scintillator element is improved. FIG. 16 (c) is obtained by superposing the central response profiles shown in FIGS. 16 (a) and 16 (b) after adjusting the scale so that the peak positions match each other. In this figure, when the signal is selected, the overlapping of responses is improved, so that the effect of improving the position discrimination ability according to the present invention is quantitatively shown.

  In the experiment, PS-PMT12 was used in place of the semiconductor light receiving element, and was arranged on each surface of the scintillator block 10 as shown in FIG. By limiting the light receiving surface of the light receiving element by the reflecting material 22, a state equivalent to the case where the semiconductor light receiving elements are arranged in a tile shape is realized. Specifically, holes were made in the reflector 22 at locations corresponding to nine light receiving portions, and light was received only from the holes. Since the PS-PMT 12 is large and collides with the adjacent PS-PMT, as shown in FIG. 17B, the scintillator surface was indirectly coupled using the light guide 24. The obtained result is shown in FIG. Although the loss of light quantity is inevitable due to the introduction of the light guide, the element can be identified on the 2D position histogram, and therefore the influence of the light guide may be ignored as a comparative experiment. FIG. 18 (a) shows the result obtained by the conventional method in which all signals are calculated, and FIG. 18 (b) shows the result obtained by the method of the present invention in which the signal was selected. It was.

  In the above embodiment, the DOI detector in which the light receiving elements are disposed on the entire surface is used as the radiation position detector. However, the type of the radiation position detector is not limited to this, and the light receiving element is not limited to this. , Two orthogonal surfaces, or 3 to 5 surfaces may be provided. The positioning direction is not limited to the triaxial direction, and may be, for example, an xy biaxial direction.

  Also, the scintillator block is not limited to a cubic scintillator element arranged in a three-dimensional array, and is described in a three-dimensional array of rectangular scintillator elements, or in JP-A-2001-28371 and JP-A-2009-270971. As described above, an optical discontinuity may be formed by another method using a crystal growth method, laser light, or the like. Of course, the rectangular parallelepiped need not be a complete rectangular parallelepiped, and may be a substantially rectangular parallelepiped. The number of divisions by the optical discontinuity is not limited.

  Further, for the position calculation, other calculation methods such as a maximum likelihood estimation method may be used in addition to the anger calculation.

  The application target is not limited to the PET detector, and may be a SPECT detector or a gamma camera.

  Further, it can be similarly applied to detectors other than the DOI that do not detect the position in the depth direction.

  The position calculation method of the radiation position detector according to the present invention can be used for calculation of the radiation absorption position of a PET detector, a SPECT detector, a gamma camera or the like.

10 ... Scintillator block 12 ... PS-PMT
20 ... Position sensitive light receiving element 22 ... Reflector 24 ... Light guide

Claims (4)

Each light receiving element of a radiation position detector that emits light when it absorbs radiation and optically couples the light receiving elements to two or more adjacent surfaces of a scintillator block having an approximately rectangular parallelepiped outer shape. In the position calculation method of the radiation position detector for calculating the output signal of and specifying the radiation absorption position,
A position calculation method for a radiation position detector, characterized in that, when a light receiving element is present at an end in an axial direction whose position is to be specified, an output signal of the light receiving element on a surface intersecting the axial direction is excluded. In position calculation, the x-axis direction, y-axis, z-axis directions orthogonal to each other, during the positioning of the x-axis direction, excludes the output signal of the light receiving element on the yz plane in at least one of x-axis direction end When positioning in the y-axis direction, the output signals of the light receiving elements on the xz plane at at least one of both ends in the y-axis direction are excluded, and when positioning in the z-axis direction, the xy plane at at least one of both ends in the z-axis direction is excluded. The position calculation method of the radiation position detector according to claim 1, wherein an output signal of the upper light receiving element is excluded. Each light receiving element of a radiation position detector that emits light when it absorbs radiation and optically couples the light receiving elements to two or more adjacent surfaces of a scintillator block having an approximately rectangular parallelepiped outer shape. In the position calculation device of the radiation position detector that specifies the radiation absorption position by calculating the output signal of
Position calculation of a radiation position detector comprising means for excluding an output signal of a light receiving element on a surface intersecting with the axial direction when there is a light receiving element at an end of the axial direction whose position is to be specified apparatus. In position calculation, the x-axis direction, y-axis, z-axis directions orthogonal to each other, during the positioning of the x-axis direction, excludes the output signal of the light receiving element on the yz plane in at least one of x-axis direction end When positioning in the y-axis direction, the output signals of the light receiving elements on the xz plane at at least one of both ends in the y-axis direction are excluded, and when positioning in the z-axis direction, the xy plane at at least one of both ends in the z-axis direction is excluded. 4. The position calculation device for a radiation position detector according to claim 3, wherein an output signal of the upper light receiving element is excluded.