JP2010169542A - Tomographic image information generation method of positron emission tomographic device, and positron emission tomographic device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a positron emission tomographic (PET) device capable of acquiring clearer emission image information in a short time, wherein an influence of a body motion of portions easily affected by the body motion, and a portion having especially useful information among them, is corrected. <P>SOLUTION: A first γ-ray generated in a body caused by a PET drug, and a second γ-ray radiated from a γ-ray source and transmitted through the body are detected by a radiation detector. Each emission image information (E-image information) E<SB>0</SB>, E<SB>1</SB>, E<SB>2</SB>in each body motion topological section 0, 1, 2 classifying a concerned domain and a breathing period is generated by using each information acquired from the detected first γ-ray. Each transmission image information (T-image information) T<SB>0</SB>, T<SB>1</SB>, T<SB>2</SB>in each body motion topological section 0, 1, 2 acquired from the detected second γ-ray is generated. Each relative displacement [F<SB>10</SB>], [F<SB>20</SB>] determined by superposing on the T-image information T<SB>0</SB>, other T-image information T<SB>1</SB>, T<SB>2</SB>is calculated. Each E-image information E<SB>1</SB>, E<SB>2</SB>is superposed on the E-image information E<SB>0</SB>by using the relative displacement. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a tomographic image information generation method and a positron emission tomography apparatus for a positron emission tomography (Positron Emission Tomography, hereinafter referred to as PET) apparatus, and more particularly, a positron suitable for performing gate imaging. The present invention relates to a method for creating tomographic image information of an emission tomography apparatus and a positron emission tomography apparatus.

  A method called gate imaging is known as a method for compensating for blurring of image information accompanying periodic body movement such as respiration. Gate imaging means that E measurement data measured over a plurality of body motion periods is divided into data of each body motion phase, and E image information for each body motion phase section is respectively divided using these divided data. It is a method of reconfiguration. For example, in gate imaging for respiratory motion, as described in Non-Patent Document 1, respiratory phase interval information is obtained by capturing minute changes in the temperature of the breath, and E measurement is performed based on this phase interval information. The obtained data is divided into body motion phase sections.

The Journal of Nuclear Medicine, Vol. 45, No.2, pages 214-219

  By performing gate imaging as described above during E measurement, it is possible in principle to compensate for the effects associated with periodic body movements. However, in order to obtain data with sufficient statistical accuracy for each body motion phase, the measurement time required is long. Since a long PET examination is painful for the subject (often a patient with a disease), the PET examination is usually completed in about 20 to 30 minutes. In this case, only measurement data of several minutes or less is acquired for each body motion phase (see Non-Patent Document 1). For this reason, the E image information obtained by reconstructing the respective data will probably contain statistical noise. When filtering is performed in the imaging process to suppress statistical noise, the image information is blurred. The statistical accuracy per single body motion phase can be improved by reducing the number of body motion phases to be divided. However, the body motion within one body motion phase time increases, and it is difficult to obtain the effect of compensating the body motion in the first place. That is, effective body motion compensation cannot be effectively performed in a short time with simple gate imaging, and there are problems of (1) blurring of E image information and (2) deterioration of quantitativeness of E image information. It is not fundamentally resolved.

  Non-Patent Document 1 solves the problems (1) and (2) by non-linearly transforming and superimposing E image information obtained by gate measurement into image information of a certain phase and superimposing pixel values. It is stated that it is possible to improve the statistical accuracy. However, Non-Patent Document 3 does not recognize the difficulty of deforming and superimposing E image information. Unlike morphological images such as image information reconstructed by X-ray CT and T image information, E image information is not intended to depict the structure of the body, so E image information for each body motion phase is transformed. It is extremely difficult to overlap each other. An object of the present invention is to provide a tomographic image information creation method and a positron emission tomography apparatus for a positron emission tomography apparatus capable of obtaining clearer emission image information in a short time for a part affected by body movement. It is to provide.

A bed on which the subject is placed, a plurality of radiation detectors arranged around the bed and outputting a signal corresponding to the energy of the detected radiation, and rotated around the bed and emitted from the subject Second radiation generating means for irradiating the subject with second radiation having energy different from the first radiation, means for obtaining body motion phase information of the subject, the first radiation and the second radiation And a processing device for creating an image based on
The processing device includes a phase section information processing unit that classifies a body motion phase at each time from the body motion phase information, and creates a first radiation tomographic image for each phase section from the first radiation information and the phase section information. One image creating means; means for setting a region of interest in the first radiation tomographic image for each phase interval; and a relative relationship between the intra-region of interest image and the first radiation tomographic image for each phase interval from the body motion phase interval Relative displacement information calculating means for obtaining displacement information, second image creating means for creating a second radiation tomographic image for each phase section from the second radiation information and the phase section information, and a second radiation tomographic image for each phase section Relative displacement information calculation means for obtaining relative displacement information between the body motion phase section and the relative displacement information between the body motion phase section obtained from the first radiation tomographic image and the second radiation tomography image First radiation tomographic image mapping calculating means for calculating a mapping from the first radiation tomographic image for each phase interval to one phase interval image using relative displacement information between body motion phase intervals obtained from A positron emission tomography apparatus comprising: an image addition unit that adds a plurality of first radiation tomographic images that are maps from a phase section.

  According to the present invention, clearer emission image information in which the influence of body movement of a part affected by body movement and a part having particularly useful information among them is corrected can be obtained in a short time.

Outline of respiratory motion compensation method. The outline | summary of the relative displacement information acquisition method which used the region of interest for E image information. The detailed block diagram of the positron emission tomography apparatus shown in FIG. The detailed block diagram of the positron emission tomography apparatus shown in FIG.

  Hereinafter, a positron emission tomography apparatus according to an embodiment of the present invention will be described with reference to the drawings as appropriate.

  In recent years, PET apparatuses have been heavily used mainly for the purpose of tumor diagnosis in the medical field. The PET device measures radiation (γ rays) derived from the radiopharmaceutical injected into the subject who is the subject and released from the subject's body, and the radiopharmaceutical in the subject's body is measured from the measurement data. The distribution of is imaged. Such a PET apparatus is used for diagnosis of metabolic function and physiological function. An X-ray CT apparatus is a representative example of a radiological image diagnosis apparatus that applies a radiation measurement technique and obtains an image inside a subject in a non-invasive manner.

Diagnosis of a malignant tumor or the like using a PET apparatus is performed as follows. First dose, labeled with positron emitting nuclide (such as ^{15 O, 13 N, 11 C} , 18 F), radiopharmaceutical accumulated specifically in the particular site in the body (hereinafter, referred to as PET agent) is the subject Is done. The positrons released from the PET drug in the body of the subject combine with the electrons in nearby cells and disappear. At the time of annihilation, a pair of γ rays having energy of 511 KeV (hereinafter referred to as pair γ rays) are emitted. Since each pair of γ-rays is emitted in directions almost opposite to each other, it is possible to determine on which straight line in the body the positron annihilation event has occurred by simultaneously measuring both γ-rays. These gamma rays are detected by a radiation detector. After detecting a statistically sufficient number of pairs of γ-rays, an image reconstruction algorithm such as a filtered back projection method (see IEEE Transactions on Nuclear Science, NS-21, pages 228 to 229) should be used. Thus, the occurrence frequency distribution of γ rays, that is, the distribution of the PET drug in the subject can be imaged. The measurement of γ-rays caused by the PET drug in the body is emission measurement (hereinafter referred to as E measurement), and the image information reconstructed based on the γ-ray detection signal obtained by the emission measurement is emission image information ( Hereinafter referred to as E image information). The E image information is generally simply referred to as a PET image, but is referred to as E image information in this specification in order to distinguish it from transmission image information (hereinafter referred to as T image information) described later. A series of processes from E measurement to reconstruction is collectively referred to as emission imaging.

  In an examination using such a PET apparatus, for example, when a PET drug called FDG (Fluoro-2-deoxyglucose), which is an analog of sugar (glucose), is administered to a subject, it is compared with a normal site. PET drugs accumulate in malignant tumors (cancer) with large glucose metabolism. For this reason, it becomes possible to diagnose the position and shape of the malignant tumor.

  By the way, in PET inspection requiring quantitativeness, measurement called transmission (transmission measurement, hereinafter referred to as T measurement) using a γ-ray source which is a transmission source provided in the PET apparatus is performed separately from E measurement. The attenuation of γ rays in PET measurement is a phenomenon that is not detected as simultaneous measurement data effective for imaging as a result of interaction with substances in the body before γ rays derived from radiopharmaceuticals go out of the subject's body. Point to. This process of correcting the attenuation of γ rays is called attenuation correction and is currently performed in most PET examinations.

The attenuation correction is usually performed using data obtained by T measurement. That is, the subject is irradiated with γ-rays (radiation) emitted from the γ-ray source by turning the γ-ray source around the subject on the bed. The respective radiation transmittances in various directions through which the γ rays pass through the subject are obtained. Data obtained by the E measurement is corrected using these radiation transmittance data. A radioisotope (Radio-Isotope; hereinafter referred to as “RI”) such as ⁶⁸ Ge— ⁶⁸ Ga or ¹³⁷ Cs is usually used as the γ-ray source. It is also possible to use an X-ray source instead of the γ-ray source. If necessary, the tomographic image information of the subject is reconstructed based on the data obtained by T measurement. This tomographic image information is morphological image information and is hereinafter referred to as T image information. The T image information represents the distribution of radiation attenuation in the body of the subject. If necessary, the attenuation rate for each projection direction can be obtained again based on the image information, and these attenuation rates can be used for attenuation correction.

  In recent years, a combined PET / CT apparatus in which an X-ray CT apparatus is arranged in parallel with a PET apparatus has become widespread. In these composite PET / CT apparatuses, attenuation correction is performed using tomographic image information obtained by an X-ray CT apparatus. A PET apparatus having a structure in which the X-ray source is swiveled inside a plurality of radiation detectors arranged in an annular shape is also reconfigured using the X-ray detection signal emitted from the X-ray source and transmitted through the subject. Attenuation correction is performed using tomographic image information.

  As a major factor that degrades the image quality of E image information, there is an influence of the movement of the subject (hereinafter referred to as body movement). Body movement includes involuntary breathing and periodic movements associated with the heartbeat, and voluntary posture changes. In the PET examination, the measurement time usually takes as long as several minutes to several tens of minutes. Therefore, it is difficult to suppress body movement without applying stress to the subject. In particular, the movement due to breathing (hereinafter referred to as breathing motion) reaches 2 to 3 cm even during resting breathing, so the influence of body motion on E image information is large in a PET examination near the lung field.

  A method called gate imaging is known as a method for compensating for blurring of image information accompanying periodic body movement such as respiration. Gate imaging means that E measurement data measured over a plurality of body motion periods is divided into data of each body motion phase, and E image information for each body motion phase section is respectively divided using these divided data. It is a method of reconfiguration. For example, in gate imaging for respiratory motion, as described in Non-Patent Document 1, a slight change in the temperature of breath is captured, or pages 876-881 of The Journal of Nuclear Medicine, Vol. 43, No. 7. Respiratory phase interval information is obtained, for example, by tracking the movement of the chest body surface with an infrared stereo camera as described in, and data obtained by E measurement based on this phase interval information is obtained for each body movement phase interval. Divide into

  As another method for compensating for respiratory motion, there is a method using a combined PET / CT apparatus (see pages 1287 to 1292 of The Journal of Nuclear Medicine, Vol. 45, No. 8). On pages 1287 to 1292 of The Journal of Nuclear Medicine, Vol. 45, No. 8, X-ray CT imaging is performed in a cine mode in a combined PET / CT apparatus to obtain T image information for each respiratory phase, and gate measurement When creating each phase image information of the E image information, attenuation correction is performed using an X-ray CT image of the corresponding respiratory phase. Note that pages 712 to 720 of IEEE Transactions on Medical Imaging, Vol. 18, No. 8 show a non-linear image registration method (non-rigid image registration method) of two image information, and IEEE Transactions on Medical Imaging, Vol. 21, No. 5, pages 450 to 461 disclose alignment techniques using reference points and pixel values.

There are the following three effects of respiratory motion on E image information.
(1) E image information is blurred.
(2) The quantitativeness of E image information is reduced.
(3) Around the lung field, the position when the X-ray CT image obtained by breath-holding imaging and the PET image are superimposed is shifted.

  Among these, (1) and (2) are the major factors that impair the image quality at the present time when the spatial resolution of the PET apparatus has dramatically increased. As for (3), superimposition of image information has become familiar with the composite PET / CT apparatus that has recently appeared (see pages 4S-14S of The Journal of Nuclear Medicine, Vol. 45, No. 1). In addition, there is a high demand for solutions mainly in fields that require the location of malignant tumors with higher accuracy, such as radiotherapy and biopsy.

  The problem of (1) often makes it difficult to evaluate the radiation treatment plan and the effect of the treatment because the area of the organ and malignant tumor or other diseased region becomes unclear. In addition, a malignant tumor that is small and does not have a relatively high degree of accumulation is clearly visible in the displayed E image information when there is no body movement. However, when there is a body motion, the malignant tumor is buried in the statistical noise included in the E image information, and it is difficult to even recognize its presence.

  The problem of (2) is that, for example, in malignant tumors with large movements, the accumulation degree is generally estimated to be low. In the E image information, the integration degree shown in a certain pixel at a position where the body movement displacement is large becomes the time-average integration degree of the peripheral part, and as a result, the integration degree deviates from the actual value. For this reason, as described above, the accumulation degree is lowered, and the diagnosis of the tumor may be inaccurate.

  The problem (3) is a problem that occurs when E image information obtained by imaging for a few minutes while breathing naturally and an X-ray CT image that is captured in a short time by holding the breath. This superimposed image information is useful for diagnosis, but it appears that there is a shift on the image up to about 1 cm around the lung field. This is because the handling of respiration is different between the PET examination and the normal X-ray CT examination, so that the positional correspondence is not necessarily taken. This problem is pointed out also in the PET / CT apparatus, and a solution is desired.

  By performing gate imaging as described above during E measurement, it is possible in principle to compensate for the effects associated with periodic body movements. However, in order to obtain data with sufficient statistical accuracy for each body motion phase, the measurement time required is long. Since a long PET examination is painful for the subject (often a patient with a disease), the PET examination is usually completed in about 20 to 30 minutes. In this case, only measurement data of several minutes or less is acquired for each body movement phase (see pages 214 to 219 of The Journal of Nuclear Medicine, Vol. 45, No. 2). For this reason, the E image information obtained by reconstructing the respective data will probably contain statistical noise. When filtering is performed in the imaging process to suppress statistical noise, the image information is blurred. The statistical accuracy per single body motion phase can be improved by reducing the number of body motion phases to be divided. However, the body motion within one body motion phase time increases, and it is difficult to obtain the effect of compensating the body motion in the first place. In other words, effective body motion compensation cannot be performed in a short time with simple gate imaging, and the problems (1) and (2) are not fundamentally solved.

  The Journal of Nuclear Medicine, Vol. 45, No. 2, pages 214 to 219 shows the E image information obtained by gate measurement for the problems (1) and (2). It states that statistical accuracy can be improved by nonlinearly deforming and superimposing information and superimposing pixel values. However, pages 214 to 219 of The Journal of Nuclear Medicine, Vol. 45, No. 2 do not mention a specific method. Unlike morphological images such as image information reconstructed by X-ray CT and T image information, E image information is not intended to depict the internal structure. For this reason, it is very difficult to deform and superimpose E image information for each body motion phase.

  As a method of removing the influence of respiratory motion in E measurement, there is a method of taking an image (breath-holding imaging) in a state where the subject has stopped breathing for a predetermined period. This is a method for reconstructing an image using data measured while the subject stopped breathing. The subject can stop breathing at a time for at most 30 seconds, so repeat the breath-holding imaging as many times as necessary, and superimpose multiple images so that statistical accuracy is improved. compensate. According to this method, the problems (1) to (3) can be solved in principle. However, since the subject who is subjected to the PET examination is forced to hold his / her breath, it will cause great pain to the subject and is not realistic.

  The Journal of Nuclear Medicine, Vol. 45, No. 8, pages 1287 to 1292. The CT image information for each body motion cycle obtained from the combined PET / CT apparatus described in pages 1287 to 1292 and the body motion cycle obtained from transmission imaging. From the T image information, the contour information is corrected using the non-rigid image registration method shown in pages 712 to 720 of IEEE Transactions on Medical Imaging, Vol. A method for obtaining image information is conceivable. However, since the composite PET / CT apparatus images two gantry images side by side with respect to the subject, the target imaging region cannot be imaged simultaneously. On the transmission image, for example, if there is a part that does not appear as information on the image, such as an invasive tumor, the part itself cannot be aligned. Cannot be aligned properly. For this reason, the above method is not a sufficient method for obtaining high-quality E image information excluding the influence of body movement.

  In the PET examination around the lung field, it is indispensable to remove the influence of body movement, particularly respiratory movement, in order to obtain high quality E image information. For this purpose, various approaches described above have been considered, but none of them has led to a fundamental solution.

  In order to obtain measurement data with high statistical accuracy for each body motion phase interval of periodic body motion (respiratory motion or heart beat) in a subject, a PET test for a long time is required. For this reason, it is difficult to obtain the measurement data within a measurement time that a general subject can withstand without causing pain. Therefore, the present inventors obtain E image information for each body motion phase section obtained by gate imaging (hereinafter referred to as E gate measurement) in emission measurement in a shorter time, and that image of a certain body motion phase section. We focused on the fact that statistical accuracy can be improved by superimposing information and superimposing each pixel value. As described in pages 214 to 219 of The Journal of Nuclear Medicine, Vol. 45, No. 2, the E image information is gated and the image information of each body motion phase section is stored in a certain way. It is possible to obtain image information with high statistical accuracy and compensated for breathing motion by adding the pixel values of the corresponding pixels by deforming and superimposing the image information in one body motion phase interval while nonlinearly distorting them. Although it has been suggested to do so, it is generally impossible to superimpose non-linear image information using only E image information with poor morphological information as a clue. Further, the E image information for each body motion phase section that can be obtained based on data obtained by gate imaging in a short time is insufficient in statistical accuracy. For this reason, it is extremely difficult to perform nonlinear superimposition of E image information directly between E image information for each body motion phase section.

  On the other hand, in parallel with E gate measurement, gate imaging in transmission measurement (hereinafter referred to as T gate measurement) is performed to reconstruct T image information for each body motion phase section. As a result, by using transmission superimposed image information (hereinafter referred to as T superimposed image information) obtained by nonlinearly superimposing T image information for each body motion phase section, these body motions are indirectly detected. It is possible to obtain emission superimposed image information (hereinafter referred to as E superimposed image information) by nonlinearly superimposing E image information for each phase interval.

  When periodic body movement is guaranteed, CT image information for each body movement phase section is collected using the CT apparatus of the PET / CT apparatus in the same manner as the T-gate measurement. Furthermore, it is possible to obtain E superimposed image information by obtaining relative displacement information from CT image information for each body motion phase section and superimposing E image information non-linearly from them.

  However, it was considered that a more effective E-superimposed image can be obtained by effectively using information that is not included in the T image information and CT image information but only included in the E image information and the superimposed image information. In other words, an invasive tumor or the like cannot be recognized in T image information or CT image information because it looks the same as a normal part, but on the E image information, a part where FDG is accumulated is compared with a normal part. Therefore, it is possible to obtain relative displacement information from the movement of the high part even in each body motion phase section. By combining this relative displacement information with the T image information and CT image information to obtain E superimposed image information, clearer E image information can be obtained in a short time.

  The basic concept of the new method found by the inventors will be described in detail below with reference to FIGS. FIG. 1 shows, as an example, a case where a respiratory cycle, which is a body motion cycle, is divided into three body motion phase sections for easy understanding. In FIG. 1, B represents the spine, C represents the malignant tumor, and L represents the lung.

T ₀ , T ₁ and T _{2 in} the upper part of FIG. 1 are T image information corresponding to body motion phase sections 0, 1 and 2, respectively. Such image information can be obtained by reconstructing information obtained by gate imaging in transmission measurement. E ₀ , E ₁ and E _{2 in} the lower part of FIG. 1 are E image information corresponding to body motion phase sections 0, 1 and 2, respectively. These pieces of image information can be obtained by reconstructing information obtained by E-gate measurement. Since the T image information T ₀ , T ₁ , T ₂ is morphological image information, the contours of bones and organs appear clearly, whereas the E image information E ₀ , E ₁ , E ₂ is functional image information. The outline is unclear or does not appear in the first place. In FIG. 1, the outline of the T image information is indicated by a solid line, and the outline of the E image information is indicated by a broken line.

As the γ-ray source, a radiation source that emits γ-rays having an energy different from that of γ-rays generated in the body of the subject by positron annihilation is used, and a radiation detector with good energy resolution such as a semiconductor radiation detector is used. Since the radiation detection signal output from the radiation detector is discriminated based on energy, E measurement and T measurement can be performed in parallel. Each data (first packet information and second packet information) obtained and discriminated by these measurements is divided into body motion phase sections based on body motion phase information acquired in the same manner as known gate imaging. To do. The E image information and the T image information for each body motion phase section are reconstructed using the divided data. Thus, T image information T ₀ , T ₁ , T ₂ and E image information E ₀ , E ₁ , E ₂ can be obtained. In addition, complete simultaneous imaging cannot be performed, but if the body motion is periodic, an image with clear and high-resolution contour information obtained by the CT apparatus of the PET / CT apparatus is displayed as T image information T ₀ , T It can be used in place of ₁ and T ₂ .

Each E image information and T image information obtained for each body motion phase section by E gate measurement, T gate measurement, or CT gate measurement is the same part of the same subject (for example, part affected by body motion). Is image information obtained when these measurements are performed in parallel (simultaneously). There is essentially no difference in the form of the subject between E and T measurements performed in parallel. That is, the spatial relative displacement between the two E image information corresponding in the two body motion phase sections in the E gate measurement is the two T image information corresponding in the body motion phase space in the T gate measurement. Will be identical to that between. For example, the spatial relative displacement between the E image information E ₁ E image information E ₀ and motion phases 1 motion phase 0 (see phase section) in the E gate measurement [F _10], see phase it is the same as the displacement between T image information T ₀ and the body movement of the phase interval 1 T image information T ₁ of the section. Therefore, information on the relative displacement (for example, [F ₁₀ ] and [F ₂₀ ]) is obtained between the two body motion phase sections in the T-gate measurement, and the information on the relative displacement is obtained as the two body motion phases. By applying to two pieces of E image information in space, E image information of one body motion phase section can be non-linearly superimposed on E image information of the other body motion phase section (for example, reference phase space). It is. As a result, non-linearly superimposed emission superimposed image information can be obtained.

  Since T image information can acquire morphological information such as body contours, lung fields and bones, nonlinear overlays described in pages 712 to 720 of IEEE Transactions on Medical Imaging, Vol. 18, No. 8, for example. By using the matching technique, it is possible to superimpose the T image information of each body motion phase section on the image of the reference phase section while nonlinearly distorting the T image information.

  However, when a malignant tumor (C in the figure) cannot be recognized in the T image information as shown in FIG. 2, the alignment information must be performed using recognizable parts such as a body contour, a lung contour, and a bone. The resulting relative displacement amount does not necessarily match the relative displacement amount of the malignant tumor in the E image information. Rather, more accurate information may be obtained by calculating the amount of displacement from the malignant tumor that can be recognized as a landmark in the E image information. Therefore, it is possible to calculate the displacement amount of the malignant tumor site by using a SME (Square Mean Error) minimizing method or a mutual information maximizing method within a certain region including the malignant tumor. The displacement obtained from these is obtained, for example, by using a non-linear superposition technique described in pages 450 to 461 of IEEE Transactions on Medical Imaging, Vol. 21, No. 5, and further the T image. From the morphological information obtained from the information, the relative displacement amount is obtained by using the non-linear superposition technique described in pages 712 to 720 of IEEE Transactions on Medical Imaging, Vol. 18, No.8. By obtaining the total displacement amount in consideration of the weight (β), an E-superimposed image can be obtained as described above.

  A positron emission tomography apparatus (PET apparatus) which is a preferred embodiment of the present invention will be described with reference to FIG. FIG. 3 is a detailed block diagram of the positron emission tomography apparatus shown in FIG. FIG. 3 shows the PET apparatus, data processing, and data flow of this embodiment. The process of obtaining E image data and T image data or CT image data from the PET or PET / CT apparatus 1 is the same as that of a general PET apparatus or PET / CT apparatus.

  The PET apparatus 1 includes a bed on which the subject sleeps, a detector ring 3 that detects radiation, an external source 4 that rotates around the bed and emits radiation, and a respiratory monitor device 5 that detects body movement of the subject. It has a processing device 6 that processes a signal that detects radiation from a detector and generates an image, and an image display device 7 that displays an image generated by the processing device 6. The processing device 6 has a CPU, a memory, and a hard disk. In order to generate an image from the radiation detection signal from the radiation detector, the processing device 6 reads a plurality of operation program modules into the memory and performs an operation process by the CPU. The arithmetic program module may be a circuit module.

  The processing device 6 includes a phase section information processing calculation module 8, a T-gate image creation calculation module 9, an E-gate image creation calculation module 10, a T-gate image relative displacement information calculation module 11, an E-gate image mapping calculation module 12, and an image addition. It has a module 14.

  The phase interval information processing operation module 8 receives the radiation detection signal of T image information and E image information from the detector ring 3, receives body motion phase information from the respiration monitor device 5, and receives the T image information and E image information. The radiation detection signal is divided into body motion phases at each time (gate time). In order to superimpose E image information in different body motion phase sections, it is necessary to associate the same body motion phase information with E image information and T image information.

The T gate image creation computation module 9 creates an image of each body motion phase section from the T image information T ₀ , T ₁ , T ₂ divided into the body motion phases received from the phase section information processing computation module 8. The CT gate image may be used instead of the T gate image.

The E-gate image creation computation module 10 creates an image of each body motion phase section from the E image information E ₀ , E ₁ , E ₂ divided into body motion phases received from the phase section information processing computation module 8.

The relative displacement information calculation module 11 between T-gate images is based on the image of each body movement phase section from the T-gate image creation calculation module 9 and a reference phase section which is a certain body movement phase section and other reference phase sections. Conversion matrices [F ₁₀ ] and [F ₂₀ ] representing the relative displacement in the body motion phase section are calculated. In FIG. 1, the respiration phase interval corresponding to the state in which the subject exhales is set as a reference phase interval, and T image information in the interval is set as reference image information T ₀ in FIG. When the T image information T _j of this reference image information T ₀ other respiratory phase interval (a phase interval j) was spatially superimposed reference image information T ₀ T image information T _j that is superimposed on the (1) It can be expressed as a map like the equation (the number of phases to be divided is 3 for the sake of explanation).

T _jo = [F _jo ] · T _j (j = 1, 2) (1)
Here, [F _j0 ] is a transformation matrix representing a relative displacement from the T image information T _j corresponding to the respiratory phase j to the reference image information T ₀ . T _{j0 on the} left side is T image information obtained by mapping the T image information T _j of the phase _j on the form of the subject (reference image information T ₀ ) in the reference phase space. Each of these T image information is expressed as a vector. Note that it is easy to extract and save a transformation matrix representing a spatial mapping from the image superposition process.

The E-gate image mapping calculation module 12 receives a conversion matrix [F _jo that represents the relative displacement from the T image information T _j corresponding to the respiratory phase to the reference image information T ₀ received from the T-gate relative relative displacement information calculation module 11. ] And the E gate image of each phase interval received from the E gate image creation calculation module 10 are used to calculate the superposition (mapping) of the reference phase interval on the reference image information E ₀ . In other words, superposition (mapping) of the reference image information E ₀ of E image information E _j corresponding to respiratory phase j obtained in PET apparatus 1, compared E image information E _j, superposed above the T image information This is performed using the transformation matrix F _j0 obtained from the _combination . The E image information E _j superimposed on the reference image information E ₀ is a mapping represented by the equation (2). The reference image information E ₀ is E image information corresponding to the reference phase section.

E _j0 = [F _j0 ] · E _j (j = 1, 2) (2)
The image addition module 14 superimposes a plurality of first radiation tomographic images, which are maps from each phase interval received from the E-gate image mapping operation module 12, on one phase interval image.

  Based on the relative displacement between the corresponding T image information in the two body motion phase sections described above, the corresponding E image information in the body motion phase section is superimposed non-linearly, and thus the above-described problem Can be improved.

  In this embodiment, relative displacement information obtained by superimposing T image information of another body motion phase section on T image information of one body motion phase section (for example, a reference phase section), that is, conversion matrix information is obtained. Since the E image information of the other body motion phase section is superimposed on the E image information of the one body motion phase section, the E image information can be created from the information for the entire measurement time. For this reason, clearer E image information (final E superimposed image information) targeted for a part affected by body movement (for example, respiratory movement) can be obtained in a significantly shorter time than normal gate imaging. According to the clearer E image information (final E superimposed image information) obtained by the present embodiment, it is possible to accurately diagnose a tumor present in a region affected by body movement.

  As described above, a bed on which the subject is placed, a plurality of radiation detectors arranged around the bed and outputting a signal corresponding to the energy of the detected radiation, and rotated around the bed, from the subject A second radiation generating means for irradiating the subject with a second radiation having an energy different from the emitted first radiation, a means for obtaining the body motion phase information of the subject, and the first radiation and the second radiation. A processing device for creating an image based on the phase information processing means for dividing the body motion phase at each time from the body motion phase information, and each phase section from the first radiation information and the phase interval information. Separately, a first image creating means for creating a first radiation tomographic image, a second image creating means for creating a second radiation tomographic image for each phase section from the second radiation information and phase section information, and a second for each phase section Radiation tomographic image Relative displacement information calculating means for obtaining relative displacement information between body motion phase sections, and first radiation for calculating a mapping from the first radiation tomographic image for each phase section to one phase section image using the relative displacement information. By using a positron emission tomography apparatus equipped with an image mapping calculation means and an image addition means for adding a plurality of first radiation images that are maps from each phase interval, the region affected by body motion is targeted. Clear emission image information can be obtained in a short time.

  Further, as described above, the processing device classifies the body motion phase at each time from the body motion phase information, creates a first radiation tomographic image for each phase section from the first radiation information and the phase section information, Create a second radiation tomographic image for each phase interval from the two radiation information and phase interval information, calculate the relative displacement information between the body motion phase interval from the second radiation tomographic image for each phase interval, and use the relative displacement information A positron emission tomography characterized by calculating a mapping from a first radiation tomographic image for each phase interval to a single phase interval image and adding a plurality of first radiation tomographic images that are mapping from each phase interval According to the tomographic image information creation method of the imaging apparatus, clearer emission image information for a part affected by body motion can be obtained in a short time.

  A positron emission tomography apparatus (PET apparatus) which is a preferred embodiment of the present invention will be described with reference to FIG. FIG. 4 is a detailed configuration diagram of the positron emission tomography apparatus shown in FIG. 4, the description of the same configuration as in FIG. 3 is omitted.

  In addition to the configuration of FIG. 3, the processing device 6 includes an E-gate image ROI relative displacement information calculation module 13. ROI (Region of Interest) is a region of interest and is a partial region of E image information including a tumor site and the like.

The E-gate image ROI relative displacement information calculation module 13 receives the region-of-interest information extracted from the E image displayed on the image display device 7, and is the relative position information of the body motion phase sections 0, 1, and 2. The transformation matrices [F ₁₀ ] _e and [F ₂₀ ] _e are calculated. More specifically, when obtaining the relative displacement information using the region of interest for the E image information, the E image information is once sent to the image display device 7. Region-of-interest information including a tumor site and the like is input on the image displayed on the image display device 7. Based on the E image information and the region of interest information, a relative displacement is obtained by the relative displacement information calculation module 13 in the E gate image ROI. In FIG. 2, this process is obtained from the transformation matrix [F _j0 ] _t representing the relative displacement from the T image information T _j corresponding to the respiratory phase j obtained from the T image information to the reference image information T ₀ and the E image information. From the transformation matrix [F _j0 ] _e representing the relative displacement from the in-region information E _{roij of} the E image information corresponding to the respiratory phase j to the reference image information E _roi0 , E corresponding to the respiratory phase j obtained from the E image information The transformation matrix [F _j0 ] _e representing the relative displacement from the in-region information E _roij of the image information to the reference image information E _roi0 is a mapping represented by the equation (3).

[F _j0 ] = (1−β) · [F _j0 ] _t + β · [F _j0 ] _e (j = 1, 2) (3)
Here, β is a weighting coefficient, and is a coefficient representing how much displacement information of E image information is considered in the transformation matrix. This coefficient can be determined by whether or not it includes a statistical error included in the E image information, for example. By using E image information and the region-of-interest information, clearer emission image information in which the influence of body movement of a part affected by body movement and a part having particularly useful information among them is corrected is obtained in a short time. be able to.

As described above, after each E image information of all body motion phase sections other than the reference phase section is superimposed on the reference E image information of the reference phase section by the image addition module 14, each pixel of each E image information is superposed. The value is added to the pixel value of the reference E image information. By this processing, it is possible to obtain E image information E ₀ ′ with high statistical accuracy for the form of the subject in the reference phase section. This is expressed by equation (3).

E ₀ ′ = (E ₀ + E ₁₀ + E ₂₀ ) / 3 (4)
(4) E ₁₀ of a calculation value when the 1 j in equation (2), E ₂₀ is the calculated value when the 2 j in equation (2).

  As described above, the processing apparatus includes means for setting a region of interest in the first radiation tomographic image for each phase interval, and the relative between the intra-region of interest image and the first radiation image for each phase interval from the body motion phase interval. Relative displacement information calculation means for obtaining displacement information is provided. In addition, the processing apparatus weights the relative displacement information between the body motion phase sections obtained from the second radiation tomographic image and the relative displacement information between the body motion phase sections obtained from the first radiation tomographic image. Has setting means for setting.

  According to the present embodiment, it is possible to obtain clearer emission image information in a short time for a part affected by body movement and a part having particularly useful information among them.

  Each embodiment described above is intended for gate measurement corresponding to respiratory motion, but can also be applied to gate measurement corresponding to heart pulsation. The PET apparatus may be an open type PET apparatus in which a flat plate detector is sandwiched between a bed and a circular detector.

DESCRIPTION OF SYMBOLS 1 PET apparatus or PET / CT apparatus 2 Bed 3 Detector ring 4 External radiation source 5 Respiration monitor apparatus 6 Processing apparatus 7 Image display apparatus 8 Phase section information processing calculation module 9 T gate image creation calculation module 10 E gate image creation calculation module 11 T-gate relative image displacement calculation module 12 E-gate image mapping calculation module 13 E-gate image ROI relative displacement information calculation module 14 Image addition module

Claims (4)

A bed on which the subject is placed, a plurality of radiation detectors arranged around the bed and outputting a signal corresponding to the energy of the detected radiation, and rotated around the bed and emitted from the subject Second radiation generating means for irradiating the subject with second radiation having energy different from the first radiation, means for obtaining body motion phase information of the subject, the first radiation and the second radiation And a processing device for creating an image based on
The processing device includes a phase section information processing unit that classifies a body motion phase at each time from the body motion phase information, and creates a first radiation tomographic image for each phase section from the first radiation information and the phase section information. One image creating means; means for setting a region of interest in the first radiation tomographic image for each phase interval; and a relative relationship between the intra-region of interest image and the first radiation tomographic image for each phase interval from the body motion phase interval Relative displacement information calculating means for obtaining displacement information, second image creating means for creating a second radiation tomographic image for each phase section from the second radiation information and the phase section information, and a second radiation tomographic image for each phase section Relative displacement information calculation means for obtaining relative displacement information between the body motion phase section and the relative displacement information between the body motion phase section obtained from the first radiation tomographic image and the second radiation tomography image First radiation tomographic image mapping calculating means for calculating a mapping from the first radiation tomographic image for each phase interval to one phase interval image using relative displacement information between body motion phase intervals obtained from A positron emission tomography apparatus comprising: an image addition unit that adds a plurality of first radiation tomographic images that are maps from a phase section.   The processing device calculates a coefficient for weighting the relative displacement information between the body motion phase sections obtained from the second radiation tomographic image and the relative displacement information between the body motion phase sections obtained from the first radiation tomographic image. The positron emission tomography apparatus according to claim 2, further comprising setting means for setting. A bed on which the subject is placed, a plurality of radiation detectors arranged around the bed and outputting a signal corresponding to the energy of the detected radiation, and rotated around the bed and emitted from the subject Second radiation generating means for irradiating the subject with second radiation having energy different from the first radiation, means for obtaining body motion phase information of the subject, the first radiation and the second radiation In the tomographic image information creating method of a positron emission tomography apparatus having a processing device for creating an image based on
The processing device classifies the body motion phase at each time from the body motion phase information, creates a first radiation tomographic image for each phase section from the first radiation information and the phase section information, A region of interest is set in one radiation tomographic image, relative displacement information between the body motion phase interval is calculated from the image in the region of interest and the first radiation tomographic image for each phase interval, and the second radiation information and the phase interval A second radiation tomographic image is created for each phase section from the information, the relative displacement information between the body motion phase sections is calculated from the second radiation tomographic image for each phase section, and the body determined from the first radiation tomographic image Using the relative displacement information during the dynamic phase interval and the relative displacement information during the body motion phase interval obtained from the second radiation tomographic image, the first radiation tomographic image for each phase interval to one phase interval image Mapping Calculated by the tomographic image information preparation method comprising adding a plurality of first radiological image is a mapping from each of the phase sections.   The processing device calculates a coefficient for weighting the relative displacement information between the body motion phase sections obtained from the second radiation tomographic image and the relative displacement information between the body motion phase sections obtained from the first radiation tomographic image. The tomographic image information creation method according to claim 3, wherein setting is performed.

Images