Image configuration device, image configuration method, image configuration program, and storage medium

Abstract

An image configuration device 1 acquires measurement data captured by a PET device 2 according to a time series, distinguishes body movement data relating to body movement of a subject obtained from the PET device 2 according to the time series for each prescribed unit time, acquires the body movement data as a unit time measurement value, sets a frame division point when the amount of variation in the unit time measurement value exceeds a prescribed threshold value, generates correction data in which measurement data is corrected in order to cancel out body movement of the subject in each frame unit divided at the division point, and reconstructs a medical image on the basis of the correction data for each frame. This makes it possible to obtain a highly accurate PET image without fixing a subject in PET inspection.

Description

Image construction device, image construction method, image construction program, and storage medium

[0001] The present invention relates to an image construction device, an image construction method, an image construction program, and a storage medium for non-invasively imaging a state inside a subject, for example.

[0002] Positron emission tomography (PET) has been used as a medical imaging technique for identifying lesions, closely examining the affected areas, and determining diagnostic and treatment strategies without harming the patient (subject). Images obtained by PET imaging are used as diagnostic images (medical images).

[0003] Here, a PET examination requires an imaging time of several minutes to several tens of minutes. The subject's movement (body movement) during the imaging time causes image blurring, which deteriorates the accuracy of the PET image. To address this problem, in conventional PET examinations, body movement is suppressed by fixing the subject's imaging target area to an examination bed or the like. However, in a PET examination targeting the head, for example, fixing the head places a burden on the subject. Furthermore, there are some subjects for whom it is difficult to fix the head.

[0004] Therefore, a method has been proposed in which, instead of immobilizing the subject, the subject is allowed to move and appropriate image correction is performed to obtain images that eliminate the effects of the subject's movement during the PET examination.

[0005] For example, a data-driven frame method has been proposed that analyzes measurement data obtained by a PET examination to estimate the amount of body movement at each time and extract frame images with little body movement. This method acquires data indicating body movement from the measurement data, registers a predetermined continuous period of time in which no body movement occurs as a static period, and generates a PET image using only images captured in the static period (frame images). Since the data-driven frame method uses only frame images with little influence from body movement, it is possible to obtain an image in which the influence of body movement is eliminated.

[0006] However, since the data-driven frame method generates frame images using only imaging data from static sections, frame images from sections with high frequency of body movement (dynamic sections) are discarded, resulting in a problem of small data statistics and large noise content in the PET images. In other words, there is room for improvement in terms of the accuracy of the PET images.

[0007] Enette Mae Revilla, Jean-Dominique Gallezot, Mika Naganawa, Takuya Toyonaga, Kathryn Fontaine, Tim Mulnix, John A Onofrey, Richard E Carson, Yihuan Lu. Adaptive data-driven motion detection and optimized correction for brain PET. NeuroImage. 2022. 252. 119031-119031.

[0008] In view of the above-mentioned problems, an object of the present invention is to provide an imaging device, an imaging method, an imaging program, and a storage medium that are capable of obtaining high-precision PET images in a PET examination without immobilizing a subject.

[0009] The present invention is characterized by an imaging device, an imaging method, an imaging program used therefor, and a storage medium on which this imaging program is recorded, which include an image acquisition unit that acquires measurement data measured by a PET device in time series, a measurement value acquisition unit that divides body movement data related to the body movement of the subject obtained in time series from the PET device into predetermined unit times and acquires them as unit time measurement values, a division point setting unit that sets a frame division point when the amount of fluctuation in the unit time measurement value exceeds a predetermined threshold, a correction data generation unit that generates correction data to correct the measurement data so as to cancel out the body movement of the subject in frame units divided by the division point in accordance with the amount of fluctuation in the unit time measurement value, and a reconstruction unit that reconstructs a PET image from the measurement data based on the correction data.

[0010] The present invention provides an imaging device, an imaging method, an imaging program, and a storage medium that can obtain high-precision PET images in a PET examination without immobilizing the subject.

[0011] 1 is a block diagram showing an example of the configuration of an image construction device of the present invention. FIG. 2 is a flowchart showing the operation of image construction processing. FIG. 3 is a flowchart showing the operation of frame division processing. FIG. 4 is a diagram showing the LoR of coincidence events. FIG. 5 is a graph showing the relationship between measurement data and frame division points. FIG. 6 is a graph showing an example of frame division processing. FIG. 7 is a flowchart showing the operation of registration. FIG. 8 is a flowchart showing the operation of reconstruction processing with body motion correction. FIG. 9 is a graph showing the operation of frame division processing in the prior art. FIG. 10 is a graph showing the operation of frame division processing of the present invention. SUVR images for comparing the present invention with the prior art.

[0012] Fig. 1 is a block diagram showing the configuration of an image construction device 1 of the present invention. As shown in Fig. 1, the image construction device 1 is communicably connected to an imaging device (PET device) 2 for positron emission tomography (PET) examinations, and has a function of constructing (reconstructing) a PET image based on measurement data output from the PET device 2.

[0013] PET is a technology in which a radiopharmaceutical labeled with a positron-emitting nuclide is administered to a subject and the radiation emitted from the body is measured using a PET device 2. The acquired data is reconstructed into a tomographic image (PET image) after various corrections. PET images are used as medical images, and analysis of the PET images is used for evaluation of physiological and pathological functions and image diagnosis. In other words, examinations (PET examinations) using the PET device 2 can identify lesions and closely examine the relevant areas without harming the patient (subject).

[0014] In PET, a diagnostic agent containing a positron-emitting radionuclide is introduced into a subject's body by injection, inhalation, or other means. The introduced diagnostic agent accumulates in specific locations within the subject's body that function according to the characteristics of the diagnostic agent. For example, when a sugar-based diagnostic agent is used, it selectively accumulates in metabolically active areas such as cancer cells. At this time, a positron is emitted from the radionuclide contained in the diagnostic agent. When the emitted positron combines with surrounding electrons and annihilates, two gamma rays (so-called annihilation gamma rays) are emitted in directions approximately 180 degrees from each other. These two gamma rays are detected by radiation detectors positioned around the subject, and the detected detection data (measurement data) is reconstructed to acquire (image) image data of the subject. In this embodiment, the PET device 2 transmits the measurement data to the image construction device 1, which then reconstructs the measurement data.

[0015] The image construction device 1 includes a control unit 11, an input unit 12, a display unit 13, a communication unit 14, and an auxiliary storage unit 15. The input unit 12, the display unit 13, the communication unit 14, and the auxiliary storage unit 15 are each connected to the control unit 11.

[0016] The control unit 11 includes a calculation unit 16 and a main memory unit 17, and executes various calculations and control operations in the image construction device 1. The calculation unit 16 is an arithmetic processing unit having a CPU or an MPU, etc. The main memory unit 17 is a so-called primary memory having a RAM (DRAM) and a ROM, etc. The RAM is used as a work area and buffer area for the calculation unit 16. Program data stored in the auxiliary memory unit 15 and data required for program execution are appropriately expanded in the RAM. The ROM stores the startup program for the image construction device 1 and default values ​​for various information.

[0017] The input unit 12 has input components that accept operational inputs from a user (e.g., a medical professional) of the image construction device 1, and an input detection circuit that is interposed between the input components and the control unit 11. The input components are, for example, a touch panel (touch input means) and / or hardware operation keys. The input detection circuit outputs an operation signal or operation data to the control unit 11 in response to the operation (operation input) of each input component.

[0018] The display unit 13 has a display and a display control circuit interposed between the display and the calculation unit 16. The display may be, for example, an LCD (liquid crystal display) or an organic EL display. The display displays various medical images such as PET images as well as an appropriate operation screen.

[0019] If the input unit 12 has a touch panel, the touch panel may be provided so as to overlap the display surface of the display of the display unit 13. In this case, the touch panel cooperates with the display of the display unit 13 to form a display with a touch panel (touch panel display). The display of the display unit 13 may also be configured to display a GUI (Graphical User Interface) having software keys (operation keys or operation buttons). In this case, operation input can be accepted via the GUI (operation screen).

[0020] The communication unit 14 has a communication circuit for connecting to a communication line. The communication circuit is a wired communication circuit or a wireless communication circuit, and is connected to other electronic devices (e.g., the PET device 2) via the communication line so as to be able to communicate with each other in accordance with instructions from the control unit 11.

[0021] Auxiliary storage unit 15 is configured with other non-volatile memories such as HDD, SSD, flash memory, and EEPROM, and stores programs and various data used by calculation unit 16 to control the operation of image construction device 1. Auxiliary storage unit 15 is a so-called secondary memory.

[0022] The auxiliary storage unit 15 of the image construction device 1 stores (registers) various data used by the image construction device 1, such as data for executing image reconstruction processing. The auxiliary storage unit 15 also stores an image construction processing program 18 for automatically executing image reconstruction processing in response to a user's operation input or when measurement data is input, and image construction processing data 19 required for executing the image reconstruction processing.

[0023] The image construction processing program 18 and the image construction processing data 19 are read out from the auxiliary storage unit 15 as needed and stored (expanded) in the main storage unit 17 (RAM). The operation of the image construction device 1 is realized by the calculation unit 16 executing the image construction processing program 18 expanded in the main storage unit 17 (RAM).

[0024] The image construction processing program 18 has a plurality of programs for executing at least the processing of each step of the image construction processing described below. For example, the image construction processing program 18 has a main processing program for selecting and executing various functions of the image construction device 1, a measurement data acquisition program for acquiring measurement data from the PET device 2, a body movement data acquisition program for acquiring body movement data related to the body movement of the subject (object) from the measurement data, a division point setting program for setting frame divisions (frame division points), a correction data generation program for generating correction data for correcting the measurement data so as to cancel out the body movement of the subject, and a reconstruction program for reconstructing a PET image from the measurement data based on the correction data.

[0025] The image construction processing data 19 includes measurement data acquired from the PET device 2, body movement data relating to the body movement of the subject, data on one or more frame division points (frame division point data) constructed based on the measurement data, correction data for correcting the measurement data to cancel out the body movement of the subject on a frame-by-frame basis, and data on a PET image (PET image data) reconstructed from the measurement data based on the correction data.

[0026] 1 is merely an example, and the configuration of the image construction device 1 is not limited to this. For example, although the image construction device 1 is formed as a single integrated device in this embodiment, the image construction device 1 may be formed by a group of multiple computers.

[0027] Fig. 2 is a flowchart showing the operation of the image construction process executed by the image construction device 1. As shown in Fig. 2, when the image construction process starts, the image construction device 1 acquires measurement data transmitted from the PET device 2 (measured by the PET device 2) in chronological order (step S1), analyzes the measurement data and performs frame division processing according to the degree of body movement of the subject (step S2), performs image (frame image) reconstruction processing for each divided frame (step S3), performs registration (step S4), performs body movement corrected reconstruction processing to correct the measurement data so as to cancel out the body movement of the subject (step S5), generates a body movement corrected PET image (step S6), and ends the image construction process.

[0028] 3 is a flowchart showing the operation of the frame division processing. The flow of the frame division processing shown in FIG. 3 is a subroutine executed in step S2 of the image construction processing described above.

[0029] 3, when the image construction device 1 starts the frame division process, it first converts the measurement data into body movement data indicating the body movement of the subject in a time series (step S21). Note that the measurement data converted into body movement data in step S21 may be data obtained by duplicating the measurement data acquired in step S1.

[0030] The type of body movement data is not particularly limited, but for example, a CoD (Center of Distribution) trace can be used as the body movement data. An outline of a method for acquiring a CoD trace, which is an example of body movement data, will be described below.

[0031] As shown in FIG. 4 , when an annihilation photon is incident on two opposing radiation detectors 2 a (a pair of opposing detectors) within a very short time interval, it can be recognized that a positron annihilation has occurred on the line of response (LoR) connecting the pair of opposing detectors. In PET, a positron annihilation event is detected by this coincidence detection. By arranging the radiation detectors 2 a in a ring shape, data can be obtained in all directions. In a PET examination, the incident direction of the annihilation photon on the radiation detector 2 a can be determined by coincidence detection.

[0032] In the coincidence counting event of annihilation photons on the LoR formed by the pair of opposing detectors, data (list mode data) of the event (list mode event) having LoR and time information is output as measurement data.

[0033] A CoD trace is dynamic data obtained by acquiring the center coordinates (center positions) of the LoR of coincidence counting events with a predetermined time resolution, for example, and averaging the coordinates for each predetermined unit time or each predetermined number of events.

[0034] An example of a method for deriving a CoD trace will be described below. For a certain list mode event, the coordinates of two crystals that make up the LoR are respectively (s ｘ , s ｙ , s ｚ ), (e ｘ , e ｙ , e ｚ ), the center coordinate of LoR (X ｔ , Y ｔ , Z ｔ ) is expressed by the following [Equation 1].

[0035]

[0036] In addition, in the above formula (1), (X ｔ , Y ｔ , Z ｔ ) indicates the midpoint between the two crystal coordinates, but correction can be made based on the difference in detection time between the two crystals. ｘ , s ｙ , s ｚ ) detection time ts and coordinate (e ｘ , e ｙ, e ｚ ) and the detection time te is Δ ｔ = t ｓ -t ｅ When this is the case, it can be expressed by, for example, [Equation 2], where c is the speed of light. That is, the center coordinate (X ｔ , Y ｔ , Z ｔ ) can also be determined using the difference in detection time between the two crystals.

[0037] If the total number of list mode events included in a certain unit time, or the number of predetermined events, is T, the value of the CoD trace in that unit time is expressed by the following [Equation 3].

[0038] Furthermore, continuous body movement data in one measurement by the PET device 2 can be obtained by the following [Equation 4].

[0039]

[0040] 5 is a graph showing an example of a CoD trace as an example of body movement data. As shown in FIG. 5, the CoD trace is converted in time series in conjunction with the measurement data, and the CoD trace value changes more when the body movement is large, and the CoD trace value changes less when the body movement is small.

[0041] Returning to FIG. 3, when the measurement data is converted into body movement data in step S21, the body movement data is classified for each predetermined unit time or for each predetermined count to obtain the unit time measurement value C ｎ (n=1 to N) (step S22). For example, the unit time is 0.1 sec to 1 sec, and the predetermined count is 10,000 events. Unit time measurement value C ｎ Each of these has data on the elapsed time from the start of measurement (imaging). The following description will be given taking an example where the unit time is 1 sec.

[0042] Next, at the reference time i ０ (Step S23). ０The reference time i is a time that defines the start of a certain frame. The "time" may also be expressed as the elapsed time from the start of measurement (imaging). For example, the reference time i ０ is the time (for example, i) of the first unit time measurement value (for example, unit time measurement value C1) for which the elapsed time from the start of measurement (imaging) is zero or the smallest. ０ = 1).

[0043] Next, a start time n is set (step S24). For example, a reference time i ０ is set as the start time n. Next, the processing time i is updated (step S25). For example, the reference time i ０ The processing time i is set to the time advanced by a unit time based on ０ +1).

[0044] Then, the determination time m is set (step S26). For example, the processing time i is set as the determination time m. Then, the reference time i ０ The reference unit time measurement value C corresponding to ｉ０ and the unit time measurement value C corresponding to the current determination time m. ｉ It is determined whether the difference in absolute value between the initial value and the current value exceeds a predetermined threshold value σ (step S27). That is, it is determined whether the amount of change in the unit time measurement value from the initial value to the current value exceeds the threshold value σ.

[0045] However, the threshold value σ is set to be equal to or less than the resolution (size of a single pixel) of an image (an image reconstructed from measurement data) captured by the PET device 2. Preferably, the threshold value σ is set to be equal to or less than half (½) the size of a single pixel of an image captured by the PET device 2, and more preferably, the threshold value σ is set to be equal to or less than ¼ the size of a single pixel of an image captured by the PET device 2.

[0046] In step S27, the reference unit time measurement value C ｉ０ and the unit time measurement value C corresponding to the current determination time m (processing time i). ｉ If it is determined that the difference in absolute value between the current determination time m and the next reference time (i=i０ ) (step S29), and then the process returns to step S24 (setting of start time n). That is, when the process returns to step S24 via steps S28 and S29, the end of the current frame (frame n) is set as the current determination time m, and the start of the next frame (frame n+1) is set as the current determination time m.

[0047] On the other hand, the reference unit time measurement value C ｉ０ and the unit time measurement value C corresponding to the current determination time m. ｉ If it is determined that the difference in absolute value between the current time and the current value does not exceed the threshold value σ (step S27: NO), it is determined whether the next determination time m (next processing time i) after a unit time has passed will not exceed the end time of the body movement data (measurement data) (i+1<N) (step S30). If it is determined that the next processing time i will not exceed the end time of the body movement data (measurement data) (step S30: YES), the processing time i is updated to the next time, and the process returns to step S26 (setting the next determination time m).

[0048] 6 is a graph showing an example of the frame division process (particularly the series of processes from step S24 to S31). As shown in FIG. 6, the unit time measurement value at the start of the frame is used as a reference (compared to the unit time measurement value at the start of the frame), and the amount of fluctuation in the unit time measurement value (reference unit time measurement value C ｉ０ and the unit time measurement value C corresponding to the determination time m ｉ A frame division point is set when the difference between the absolute value of the current frame and the absolute value of the frame (difference ... ｉ０ Then, in the next frame, the reference unit time measurement value C ｉ０ Compared with the unit time measurement C ｉ If the threshold value σ is exceeded, a frame division point is set. By repeating this process, frame division points are set.

[0049] 3, if it is determined that the next processing time i has exceeded the end time of the entire body movement data (measurement data) (the current unit time is the last unit time, or there is no next unit time) (step S30: NO), the frame division processing flow ends and returns to the image construction processing (step S2 ends and proceeds to step S3). Note that when the frame division processing flow ends, frame images are generated, with the frame division points set in the frame division processing as delimiters. Furthermore, in step S3, reconstruction processing is performed for each generated frame image.

[0050] As described above, the frame division process is a process in which the body movement data regarding the subject's body movement obtained in time series from the PET device 2 is divided into predetermined unit times to obtain unit time measurement values, and a frame division point is set when the amount of fluctuation in the unit time measurement value exceeds a predetermined threshold.

[0051] 7 is a flowchart showing the registration operation. The registration flow shown in FIG. 3 is a subroutine executed in step S5 of the image construction process described above.

[0052] As shown in FIG. 7, when the image construction device 1 starts registration, it creates a plurality of frame images I separated by frame division points set in the frame division process. ｆ (step S41), and a plurality of frame images I ｆ From the reference frame I ｒ (Step S42). For example, the generated frame image I ｆ (f=1 to F), and one of them is used as the reference frame I ｒ and the reference frame I ｒ Frame images other than the target frame I ｔ (1≦t≦F). Here, the frame image I ｆ Which of the reference frames I ｒ Although there is no particular limitation as to which frame to use, for example, the first frame I １ the reference frame I ｒ It can be said that:

[0053] Next, the parameter (registration parameter) R is initialized (step S43), and the target frame I ｔ (Target frame I ｔ (Step S44), and the target frame I ｔ , and the image transformation is performed on the correction frame I ｎ (I ｎ =RI ｔ ) is generated (step S45). Here, the target frame I set in step S44 ｔ is a set of multiple target frames I ｔ If there is a parameter R (for example, R ｔ ) is expressed by, for example, a 4×4 homogeneous transformation matrix.

[0054] Next, the reference frame I ｒ and correction frame I ｎ Similarity (sim(I ｒ , I ｎ )) is calculated (step S46), and the similarity (sim(I ｒ , I ｎ )) (step S47), and it is determined whether to end the registration (step S48). ｒ , I ｎ )) is calculated using, for example, mutual information. Also, a known technique such as the Nelder-Mead algorithm can be used for the optimization process in step S47. Furthermore, in step S48, if there is no unprocessed target frame (it is the last target frame), or if the difference (variation) in the absolute values ​​of the parameter R between consecutive frames exceeds a predetermined threshold b (|R ｔ＋１ -R ｔ |<threshold b) can be set as a registration termination condition.

[0055] If it is determined that the registration is not to be completed (the registration completion condition is not satisfied) (step S48: NO), the process proceeds to step S48 (next target frame I ｔ On the other hand, if it is determined that the registration is to be completed (the registration completion condition is satisfied) (step S48: YES), the body motion correction parameter M is derived, the registration flow is completed, and the process returns to the image construction process (step S4 is completed, and the process proceeds to step S5). ｔ The parameter M for correcting body movement in ｔ can be derived in the form of the following [Equation 5]. ｆ The motion correction parameter M corresponding to ｆ It is expressed as (f=1 to F).

[0056] Note that the reference frame I ｒ For the registration of ｒ This can be omitted because it is a process of comparing the values.

[0057] 8 is a flowchart showing the operation of the reconstruction processing with motion correction. The flow of the reconstruction processing with motion correction shown in FIG. 8 is a subroutine executed in step S6 of the image construction processing described above.

[0058] As shown in FIG. 8, when the image construction device 1 starts the reconstruction process including the body motion correction, it generates an initial image from the measurement data (step S51) and calculates the body motion correction parameter M ｆ (Step S52). Hereinafter, the body motion correction parameter M ｆUsing the parameter M, a series of processes including forward projection (step S53), back projection (step S54), sensitivity correction (step S55), attenuation correction (step S56), and scattering correction (step S57) are performed on the initial image, the image is updated (step S58), the flow of reconstruction processing including body motion correction is terminated, and the process returns to image construction processing (step S5 is terminated and the process proceeds to step S6). ｆ By applying the motion correction parameter M ｆ is a value for correcting the image movement in the opposite direction of the body movement, and the body movement correction parameter M ｆ The subject's body movement is cancelled out by adding the above. The updated image is then generated as a body movement corrected PET image (step S6). Note that the series of processing steps from the forward projection processing (step S53) to the scatter correction processing (step S57) can be performed using known techniques in the field of PET image correction processing, and therefore detailed description thereof will be omitted.

[0059] With the above configuration and operation, when the amount of fluctuation in the measurement value per unit time in the measurement data output from the PET device 2 exceeds the threshold value σ, a frame division point is set, correction data is generated for each divided frame image, and reconstruction processing is performed based on the correction data.As a result, a PET image with body movement correction can be generated that includes data from sections where no body movement occurs (static sections) and data from sections where body movement occurs frequently (dynamic sections), and high-precision PET images can be obtained during PET examinations without immobilizing the subject.

[0060] In the conventional method (data-driven frame method), frame images in periods with high frequency of body motion (dynamic periods) are discarded (see FIG. 9A), but according to the present invention, it is possible to generate a motion-corrected PET image using frame images from all periods, including dynamic periods (see FIG. 9B), thereby preventing a decrease in data statistics and reducing noise contained in the PET image.

[0061] Taking the case of Figure 9B as an example, according to the present invention, frames are set consecutively from the start position of the frame until a certain amount of movement occurs, which is regarded as one frame. This allows for smaller frames to be taken in sections with large movement (the section from 10 seconds to 20 seconds on the horizontal axis of Figure 9B) and larger frames to be taken in sections with small movement (the section from 20 seconds to 50 seconds on the horizontal axis of Figure 9B), thereby enabling adaptive frame division.

[0062] Fig. 10 shows a standardized uptake value ratio (SUVR) image generated from a PET image for comparing the present invention with the prior art. As shown in Fig. 10, the present invention can reconstruct an image that is more suitable as a medical image for diagnosis than the prior art.

[0063] Furthermore, in the present invention, the section from 10 seconds to 20 seconds on the horizontal axis of Figure 9B where intermittent movement occurs is divided into small frame units of approximately 0.5 seconds, and can be used to estimate the subject's body movement without discarding all the data.

[0064] Furthermore, when aligning each frame, tracking accuracy can be stabilized by setting appropriate initial parameters based on the body motion in the previous frame, filtering noise that occurs in images with low statistics, and suppressing noise using histogram processing. Essentially, only one parameter is required: the difference from the starting position, which serves as the basis for frame division. Furthermore, this parameter is independent of the PET data and is determined by the resolution of the PET system. Therefore, by trialing several values ​​in advance, it is possible to easily derive a universally optimal value. The frame-division data is imaged and then aligned, and corrections to counteract body motion are performed during image reconstruction. This allows for the generation of PET images that eliminate the effects of body motion.

[0065] Therefore, according to the present invention, highly accurate body motion correction can be performed on PET data acquired using a general clinical protocol, while eliminating all factors that make it difficult to handle, such as complicated preprocessing, sensor installation, or parameter adjustment. In particular, even in cases where intermittent motion occurs, which was difficult to handle using conventional methods, body motion can be estimated by performing appropriate frame division according to the motion. Implementation of body motion correction according to the present invention enables PET examinations that reduce the burden on elderly patients and patients with severe illnesses, whose heads are difficult to immobilize, without requiring additional protocols such as external sensors or pre-measurements.

[0066] Furthermore, according to the present invention, the unit time measurement value is the average value of the center positions of the LORs for each unit time, so that the frame division points can be set appropriately and the frame images can be corrected appropriately.

[0067] Furthermore, according to the present invention, the threshold value σ is set to a value that is equal to or less than the value indicating the resolution of the captured image in the PET device 2 (the size of a single pixel), so that body movements accompanied by rotational movements that tend to produce small amounts of fluctuation can be appropriately detected.

[0068] The PET image construction device in this invention corresponds to the image construction device 1, and similarly, the PET device corresponds to the PET device 2, the image acquisition unit corresponds to the measurement data acquisition program and the control unit 11 that operates in accordance with the program, the measurement value acquisition unit corresponds to the body movement data acquisition program and the control unit 11 that operates in accordance with the program, the division point setting unit corresponds to the division point setting program and the control unit 11 that operates in accordance with the program, the correction data generation unit corresponds to the correction data generation program and the control unit 11 that operates in accordance with the program, and the reconstruction unit corresponds to the reconstruction program and the control unit 11 that operates in accordance with the program, but this invention is not limited to this embodiment and can be embodied in various other ways. Furthermore, the specific configurations etc. given in the above-mentioned embodiments are merely examples and can be changed as appropriate depending on the actual product.

[0069] For example, in addition to the above-described embodiments, there are other PET image reconstruction methods that include motion correction to which the frame segmentation of the present invention can be applied. For example, there is a method called multiple acquisition frames (MAF), which superimposes registered frame images. There is also a method (MOLAR (Motion-compensation OSEM List-mode Algorithm for Resolution-recovery reconstruction)) in which the crystal coordinates at both ends of the LoR included in the list-mode event are coordinate-transformed using motion-corrected data at the corresponding time and used for reconstruction. By performing coordinate transformation according to the time information included in the list-mode event on the global sensitivity image used for sensitivity correction and the μ map used for attenuation correction and scatter correction, it is possible to obtain reconstructed images to which these corrections have been applied.

[0070] The invention of claim 1 can be an image construction device including an image acquisition unit that acquires measurement data measured by a PET device in time series, a measurement acquisition unit that divides body movement data related to the body movement of the subject obtained in time series from the PET device into predetermined unit times and acquires them as unit time measurement values, a division point setting unit that sets a frame division point when a fluctuation amount of the unit time measurement value exceeds a predetermined threshold, a correction data generation unit that generates correction data for correcting the measurement data so as to cancel out the body movement of the subject in units of frames divided by the division points in accordance with the fluctuation amount of the unit time measurement value, and a reconstruction unit that reconstructs a PET image from the measurement data based on the correction data.The invention of claim 2 can be an image construction device according to claim 1, wherein the unit time measurement value is a value obtained by averaging the center positions of the LORs for each unit time.The invention of claim 3 can be an image construction device according to claim 1 or 2, wherein the correction data generation unit sets a frame division point when a fluctuation amount of the unit time measurement value exceeds a predetermined threshold, based on the unit time measurement value at the start of a frame. The invention of claim 4 can be the image construction device of claim 1, 2, or 3, in which the threshold is set to be equal to or smaller than the size of a single pixel of an image reconstructed from the measurement data.The invention of claim 5 can be the image construction device of claim 4, in which the threshold is set to be equal to or smaller than half the size of the single pixel.

[0071] The present invention can be utilized in industries related to image construction devices for non-invasively imaging activity within an object.

[0072] REFERENCE SIGNS LIST 1... image construction device 2... PET device 11... control unit 15... auxiliary storage unit 18... image construction processing program 19... image construction processing data

Claims

1. An image acquisition unit that acquires measurement data measured by a PET device in chronological order, A measurement value acquisition unit that acquires body movement data relating to the subject's body movements obtained from the PET device in a time series, distinguishing it at predetermined unit time intervals and acquiring it as unit time measurement values, A division point setting unit sets a frame division point when the amount of variation of the unit time measurement exceeds a predetermined threshold, A correction data generation unit generates correction data for correcting the measurement data in a frame-by-frame manner divided at the division point according to the amount of variation of the unit time measurement, so as to cancel out the subject's body movement. The system includes a reconstruction unit that reconstructs a PET image from the measurement data based on the correction data. Image configuration device.

2. The aforementioned unit time measurement is the average value of the center position of LOR for each unit time. The image constructing apparatus according to claim 1.

3. The correction data generation unit sets a frame division point based on the unit time measurement at the start of the frame, when the amount of variation in the unit time measurement exceeds a predetermined threshold. The image configuration apparatus according to claim 2.

4. The threshold is set to be less than or equal to the size of a single pixel in the image reconstructed from the measurement data. The image configuration apparatus according to claim 1, 2, or 3.

5. The threshold is set to be less than or equal to half the size of the single pixel. The image constructing apparatus according to claim 4.

6. The measurement data obtained by the PET device is acquired in chronological order. Body movement data of the subject obtained in a time series from the PET device is separated into predetermined unit time intervals and acquired as unit time measurements. When the amount of variation of the unit time measurement exceeds a predetermined threshold, a frame division point is set. Correction data is generated to correct the measurement data in a way that cancels out the subject's body movement in a frame divided at the division point according to the amount of variation of the measured value per unit time. Based on the correction data, the PET image is reconstructed from the measurement data. Image composition method.

7. Computers, An image acquisition means that acquires measurement data measured by a PET device in a time series, A measurement value acquisition means that acquires body movement data relating to the subject's body movements obtained from the PET device in a time series, distinguishing it at predetermined unit time intervals and acquiring it as a unit time measurement value, A division point setting means for setting a frame division point when the amount of variation of the unit time measurement exceeds a predetermined threshold, Correction data generation means for generating correction data to correct the measurement data in a frame unit divided at the division point according to the amount of variation of the unit time measurement value, so as to cancel out the subject's body movement. This is configured to function as a reconstruction means for reconstructing a PET image from the measurement data based on the correction data. Image composition program.

8. Computers, An image acquisition means that acquires measurement data measured by a PET device in a time series, A measurement value acquisition means that acquires body movement data relating to the subject's body movements obtained from the PET device in a time series, distinguishing it at predetermined unit time intervals and acquiring it as a unit time measurement value, A division point setting means for setting a frame division point when the amount of variation of the unit time measurement exceeds a predetermined threshold, Correction data generation means for generating correction data to correct the measurement data in a frame unit divided at the division point according to the amount of variation of the unit time measurement value, so as to cancel out the subject's body movement. The system stores an image configuration program that functions as a reconstruction means for reconstructing a PET image from the measurement data based on the correction data. A computer-readable storage medium.