JP2011022119A - Method, device and program for processing nuclear medicine image - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for removing an influence of a cardiac muscle image by high accumulation other than cardiac muscle such as liver and gallbladder, a program for allowing a computer to execute the method, and a device for executing the method, in a cardiac muscle nuclear medicine image. <P>SOLUTION: On each projection image in the cardiac muscle nuclear medicine image, a processing is performed, wherein a concerned domain is set in the whole cardiac muscle, and a projection image in which a high accumulation portion other than a cardiac muscle part exists in the concerned domain is deleted, and a portion other than the concerned domain set in the cardiac muscle domain relative to a residual projection image is mask-processed and removed. According to the processing, effective removal of an influence of the high accumulation portion in the projection image in which a cardiac muscle portion is close to a high accumulation portion other than cardiac muscle, which has been difficult to be removed hitherto, becomes possible. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an image processing technique for nuclear medicine images. More specifically, the present invention relates to an image processing technique capable of effectively removing artifacts in a myocardial nuclear medicine image.

As one of the image diagnosis of heart disease, a nuclear medicine image diagnosis method represented by SPECT is used. In the nuclear medicine image diagnostic method, a diagnostic image can be obtained by administering a drug (radiopharmaceutical) labeled with a radioisotope to a subject, detecting the released γ-rays, and performing an imaging process. Radiopharmaceuticals used for diagnostic imaging of heart diseases include thallium chloride-201, hexakis (2-methoxyisobutylisonitrile) technetium ( ^99m Tc), tetrophosmin technetium ( ^99m Tc), 3-iodobenzylguanidine ( ¹²³ I), 15- (4-Iodophenyl) -3 (R, S) -methyldecanoic acid ( ¹²³ I) and the like have been developed and used in clinical practice.

Among these radiopharmaceuticals, a preparation using ^99m Tc as a radioisotope (hereinafter referred to as ^99m Tc preparation) is excellent in versatility because it can be made into a kit and has energy suitable for a SPECT apparatus. Therefore, there is an advantage that a good image can be obtained. However, ^99m Tc formulations are generally large in molecular size and have high accumulation in the hepatobiliary system. Such high integration other than in the myocardium causes streak artifacts on the reconstructed image and hinders diagnosis (Non-patent document 1, Non-patent document 2, Non-patent document 3, Non-patent document 4). ). In addition, radioiodine preparations have also been reported to accumulate highly in the liver and to have a deficient lower wall in myocardial SPECT (Non-patent Document 5).

  As a method for preventing the occurrence of artifacts on a reconstructed image caused by high accumulation other than the myocardium, a method for removing a high accumulation site by mask processing has been studied (Non-Patent Document 1, Non-Patent Document). 2, Non-Patent Document 3). According to previous reports, it has been disclosed that the influence on the myocardial image due to the high integration can be reduced by performing mask processing on the projection data before reconstruction processing (Non-Patent Document 1, Non-Patent Document 1). Document 2, Non-Patent Document 3). However, when a highly integrated part other than the myocardium is close to the myocardium, it is difficult to remove the highly integrated part on the image data, and it is impossible to completely remove the influence of the highly integrated part on the myocardial image. It is difficult (Non Patent Literature 2, Non Patent Literature 3).

  Haruo Kuno et al., “Effects of high accumulation of gallbladder on 99mTc-tetrafosmin on myocardial SPECT images and improvement method”, Journal of Japanese Society of Radiological Technology, August 2000, 56, 8, p. 1044-1051   Kamon Imai et al., “Regarding the Cause of MIBI Myocardial Scinching Artifact”, Nuclear Medicine, 1995, 32, 3, p. 307-310   Akihiro Takagi et al., “Reduction of the effect of high hepatic accumulation on the myocardium in SPECT imaging using 99mTc myocardial perfusion preparation; usefulness and problems of mask processing method”, Nuclear Medicine, 1999, 36, 5, p. 459-465   Onishi Hideo et al., “Effects of the liver on myocardial SPECT images using 99mTc-labeled myocardial blood flow preparation”, Nuclear Medicine, 1998, 35, 6, p. 375-383   Hideki Kobayashi et al., “Characteristics and countermeasures of 123I-MIBG myocardial SPECT that appear due to high accumulation of liver and its countermeasures—an examination using phantoms”, Nuclear Medicine, 1994, 31, 4, p. 359-366

  The present invention has been made in view of the above facts, and provides a method for removing the influence on highly integrated myocardial images other than the myocardium, a program for causing a computer to execute the method, and an apparatus for executing the method The purpose was to do.

  As a result of the study, the inventor has found that the image quality of a reconstructed image created using the remaining data is essential even if a projection image in which a highly integrated part other than the myocardium is close to the myocardial part is deleted in the projection data set. Found no negative impact. Based on this knowledge, a region of interest is set for the entire myocardium on each projection image, a projection image in which a highly integrated portion other than the myocardial region is present in the region of interest is deleted, and the region of interest for each remaining projection image The above-mentioned problems can be solved by performing a process such as removing a region other than the region (that is, a region other than the region of interest set in the myocardial region) by performing a mask process or the like, and the effect of the highly integrated region is effectively As a result, the present invention was completed.

  In this specification, the projection data set refers to a set of projection images at each projection angle obtained by one imaging process. For example, when a 360 ° projection is performed on a patient in 2 ° steps, a set of 180 projection images obtained at projection angles 2 °, 4 °, 6 °,... 358 °, 360 ° This is the projection data set in this specification.

  An image processing method according to an aspect of the present invention includes a projection data set acquisition step in which a computer acquires a projection data set of a myocardial nuclear medicine image, and the computer calculates a myocardial region on each projection image included in the projection data set. A region of interest setting step for setting a region of interest in a region to include, an image deleting step for deleting a projection image in which a highly integrated region other than the myocardial portion overlaps the region of interest, and a computer for the image deleting step For each projection image included in the projection data set after deleting the projection image in which the highly integrated region other than the myocardial region overlaps the region of interest, a portion other than the region of interest set in the region of interest setting step is The non-region of interest deletion step to be deleted and the computer uses the projection data set after the non-region of interest deletion step Comprising an image reconstruction step of performing reconstruction, the.

  An image processing program according to another aspect of the present invention includes, in a computer, a projection data acquisition step for acquiring a projection data set of a myocardial nuclear medicine image, and a myocardial region on each projection image included in the projection data set. A region of interest setting step for setting a region of interest in a region, an image deleting step for deleting a projection image in which a highly integrated portion other than the myocardial portion overlaps the region of interest, and a high integration other than the myocardial portion in the image deleting step Non-interest region deletion step of deleting a portion other than the region of interest set in the region of interest setting step for each projection image included in the projection data set after deleting the projection image whose part overlaps the region of interest And an image reconstruction step of performing image reconstruction using the projection data set after the non-interest region deletion step.

  An image processing apparatus according to yet another aspect of the present invention includes a projection data set acquisition unit that captures a projection data set of a myocardial nuclear medicine image, and an area including a myocardial region on each projection image included in the projection data set. A region-of-interest setting unit that sets a region of interest; an image deletion unit that deletes a projection image in which a highly integrated region other than a myocardial region overlaps the region of interest among the projection images included in the projection data set; A non-region of interest deletion unit that deletes a portion other than the region of interest set by the function of the region of interest setting unit on each projection image, and an image reconstruction unit that performs image reconstruction using the projection data set .

As pointed out in the prior art, when the myocardial region and a highly integrated region other than the myocardium are close to each other on the projection image, it is difficult to completely delete the highly integrated region. It has been difficult to completely remove the effect of the highly integrated site on the reconstructed image (Kamon Imai et al., Nuclear Medicine, 1995, 32, 3, p.307-310; Akihiro Takagi et al., Nuclear Medicine, 1999 36, 5, p.459-465). In the image processing method and the image processing program according to the present invention, a projection image in which a region of interest set so as to include a myocardial region and a highly integrated region other than the myocardial region overlap each other is deleted, and in other projection images, the interest Since the process of deleting regions other than the region is performed, it is possible to effectively remove the influence of highly integrated regions other than the myocardium in the reconstructed image.
Similarly, the image processing apparatus according to the present invention also deletes a projected image in which a region of interest set to include a myocardial region and a highly integrated region other than the myocardial region overlap each other, and in other projected images, the interest Since it has a function of deleting an area other than the area, it is possible to generate a reconstructed image in which the influence from a highly integrated part other than the myocardial part is completely removed.

In the image processing method and the image processing program according to the present invention, the non-interest region deletion step of deleting a portion other than the region of interest set in the myocardial region includes a method of deleting a portion other than the region of interest by mask processing, This can be done by a method of substituting a null code or an average count value in the background for pixels corresponding to a region other than the region of interest.
Similarly, the non-region-of-interest deletion unit in the image processing apparatus according to the present invention includes a method of deleting a portion other than the region of interest set in the myocardial region by mask processing, or a null code for pixels corresponding to the region other than the region of interest. Alternatively, a portion other than the region of interest may be deleted by a method of substituting the average value of the counts in the background.
In particular, if a method of substituting the average value of the background is used, it is possible to suppress the influence of discontinuity at the region of interest boundary.

The image processing method according to the present invention includes an image shift step in which the computer moves the center of the myocardial region to the center of rotation of the screen in each projection image included in the projection data set before the region of interest setting step, The region-of-interest setting step may be configured such that the computer sets a region of interest in a region including the myocardial region on each projection image in the projection data set after the image shift step processing.
With such a configuration, after the image shift step, the myocardial region is located at the center of rotation on all the projected images, so a region of interest is set on a representative projected image, Can be applied to projected images. Thereby, it is not necessary to set the region of interest separately for each projection image, and the processing burden can be greatly reduced.

Similarly, the image processing program according to the present invention causes the computer to move the center of the myocardial region to the rotation center of the screen in each projection image included in the projection data set before the region of interest setting step. In the region-of-interest setting step, the computer may be configured to execute processing for setting a region of interest in a region including a myocardial region on each projection image in the projection data set after the image shift step processing.
With this configuration, after execution of the image shift step, the myocardial region is located at the center of rotation on all projection images, as in the invention related to the image processing method. A region of interest can be set on the image and applied to all projected images. Thereby, it is not necessary to set the region of interest separately for each projection image, and the processing burden can be greatly reduced.

In addition, the image processing apparatus according to the present invention may include an image shift unit having a function of moving the center of the myocardial region to the rotation center of the screen for each projection image included in the projection data set.
By providing such a function, the myocardial region can be positioned at the center of rotation on all the projection images, and as a result, the region-of-interest setting unit sets the region of interest on the representative projection image, Processing such as applying it to all projection images can be performed. Thereby, it is not necessary to set the region of interest separately for each projection image, and the processing load in the region of interest setting unit can be greatly reduced.

The image processing method according to the present invention further includes a data complementing step in which the computer supplements the region deleted by the non-interest region deletion step with the value at the region of interest boundary on the image after the non-interest region deletion step, The image reconstruction step may perform image reconstruction using the projection data set after the data complementing step.
Similarly, the image processing program according to the present invention includes a data complementing step for complementing a region deleted by the non-interest region deletion step with a value at the region of interest boundary on the image after the non-interest region deletion step. Further, the image reconstruction step may be executed by causing the computer to perform image reconstruction using the projection data set after the data complementing step.
Similarly, the image processing apparatus according to the present invention further includes a data complementing unit having a function of complementing the deleted region with the value at the region of interest boundary on the image after the non-region of interest deletion processing. May be.
With such a configuration, it is possible to suppress the influence of the discontinuity of the region of interest boundary after the deletion process.
Complementation can be performed using, for example, a method of extending the value of the boundary in the region of interest in the vertical direction and the horizontal direction for the image after the non-region of interest deletion process. At this time, for pixels at coordinates whose values are extended from both the vertical and horizontal directions, either one of the values from the vertical or horizontal direction is filled, or the average value of the values from the vertical and horizontal directions is filled. Can be done. In addition, for pixels at coordinates whose values are not extended from both the vertical and horizontal directions, fill the values that have been extended to pixels that exist in the vertical and horizontal directions of the pixel, or calculate the average of those values. What is necessary is just to perform the filling process.
As another method of extending the boundary value, a method of extending the boundary value along a radial line drawn outward from the center of the image may be used. At this time, with respect to a pixel at a coordinate whose value has not been extended, a process of filling a value that has been extended to a pixel that exists in the vertical or horizontal direction of the pixel or filling an average value of those values may be performed. Note that the above-described configuration can be applied to the image processing apparatus according to the present invention as it is.

  According to the present invention, highly integrated sites other than the myocardial site can be substantially removed from the myocardial nuclear medicine image, and streak artifacts in the reconstructed image can be effectively reduced.

The figure which shows an example of the flow of the process in the preferable aspect of the image processing method which concerns on this invention. The figure which shows an example of the flow of a process in the image shift step in the image processing method which concerns on this invention. The figure which shows an example of the flow of the shift parameter calculation process in the image processing method which concerns on this invention The figure which shows an example of the structure in the preferable aspect of the image processing program which concerns on this invention The figure which shows an example of the functional block diagram in the preferable aspect of the image processing apparatus which concerns on this invention A figure showing an image (a) before shifting the myocardial region to the rotation center of the image and an image (b) after shifting to the rotation center (the rotation center is indicated by a circle) The example which performed the mask process. (A) An example of a projection image in which a highly integrated part exists outside the region of interest (before mask processing), (b) An image obtained by performing mask processing on the image of (a), (c) Highly integrated in the region of interest An example of a projected image in which a part exists, an image obtained by performing mask processing on the entire image for the images (d) and (c) Reconstructed image in a phantom that does not exhibit high accumulation in the liver (FBP processing) Reconstructed image of a phantom that is highly integrated in the liver (FBP processing) Reconstructed image of a phantom that is highly integrated in the liver (FBP processing after mask processing) Diagram showing the region of interest set in the myocardium

  The present invention will be described below with reference to the drawings. In addition, the example shown below demonstrates only a preferable form to the last, and the content of this invention is not limited at all by these description.

  First, the image processing method according to the present invention will be described. 1 to 3 are flowcharts showing the flow of processing in the image processing method according to the present invention. In the image processing method according to the present invention, first, a projection data set acquisition step is executed by a computer, and a projection data set to be processed is input to a computer system that implements the present invention (step S1). As the projection data set, a projection data set captured by a normal nuclear medicine image capturing apparatus such as a SPECT apparatus or a PET apparatus can be used. The projection data set can be obtained in a state stored in a computer-readable storage medium such as a hard disk, a CD-ROM, a DVD, or the like, stored in a computer-readable format such as DICOM format. it can. The acquired projection data set is read by a reader provided in the computer system and input to the system. Note that the projection data set may be directly input from a nuclear medicine imaging apparatus through a network.

  When the input of projection data is completed, the image shift step is executed by the computer, and the center of the myocardial region in each projection image is aligned with the rotation center of each projection image (step S2). This process can be performed by applying a separately calculated shift parameter to each projection image (steps S21 and S22). A preferred example of this shift parameter calculation method will be described with reference to FIG.

  In a preferred embodiment, the shift parameter is calculated using the reconstructed image of the representative slice. First, using the input projection data set, one cross-sectional image of a representative slice including a myocardial region is created (step S23). As the representative slice, a cross-sectional image including the vicinity of the center of the myocardium can be preferably used.

  When the representative slice is obtained, the central part of the myocardium is determined on the representative slice (step S24). The central part of the myocardium can be determined by a method in which the user visually determines it through a display or the like and designates it using a mouse pointer or the like. Alternatively, the center of gravity of the myocardial region may be calculated on the representative slice and used as the center of the myocardium.

  When the central part of the myocardium is determined on the representative slice, an image shift parameter is calculated (step S25). The shift parameter is a shift parameter for moving the myocardial region to the center of rotation in each projection image. The shift parameter can be easily calculated for the projection image at each projection angle based on the relationship between the coordinates of the central part of the myocardium determined above and the rotation center of the representative slice. The shift parameter calculated in this step is used as the shift parameter used in the image shift step.

Returning to FIG. 1, the image processing method according to the present invention will be described. When the image shift step is completed, the region of interest setting step is executed by the computer, and a region of interest including the myocardial region (shown as ROI in FIG. 1) is set (step S3). The region of interest can be set by a known method such as setting the region of interest including the myocardial region on the projection image. The shape of the region of interest does not need to be particularly limited, but it is preferable to use a region of interest having a predetermined shape such as a circle or an ellipse centered at the rotation center of the screen in order to improve reproducibility between users. In that case, a method of adjusting the size of a circular or elliptical region of interest centered at the center of rotation so as to completely include the myocardial region can be used. Further, the contour of the myocardial region is extracted in advance using a known method such as a threshold method, the maximum value of the distance from the rotation center of each point constituting the contour is calculated, and then a radius larger than the maximum value is calculated. It is also possible to obtain the area of interest as a circle.
Further, in the present embodiment, a process for shifting the myocardial region in each projection image to the rotation center of the image in advance by executing the image shift step is performed. Therefore, the myocardial region is displayed at substantially the same position in each projection image. Accordingly, if a region of interest is obtained by a method as described above for a representative projection image, the region of interest setting is completed by applying the same region of interest to all the projection images. That is, it is not necessary to set the region of interest separately for each projection image, and the processing burden is greatly reduced.
Needless to say, the size of the region of interest is preferably as small as possible including the entire myocardial region.

  When the region of interest is set, the computer is caused to execute an image deletion step, and a projection image in which a highly integrated region other than the myocardial portion overlaps the region of interest is deleted (step S4). The image deletion step can be performed by visually confirming each projection image and deleting the corresponding image. On the sinogram, a highly integrated region other than the myocardial region corresponds to the contour of the region of interest. Alternatively, a method may be used in which a projection angle existing inside a line (two existing across the center) is examined and a projection image corresponding to the projection angle is deleted.

  Next, the non-region of interest deletion step is executed by the computer for the projection data set after the image deletion step, and the portions other than the portion selected as the region of interest on each projection image are deleted (step S5). In this step, a known mask process is performed on each projection image to delete regions other than the region of interest, or a null code or an average count in the background is assigned to pixels corresponding to the region other than the region of interest. It can be done by the method of doing. Here, for the average value in the background, for example, a certain region is designated in a region (such as a lung field) that does not emit a signal on the projection image, and the average value of the count of each pixel in the region is calculated. Can be easily obtained.

When the non-interest region deletion step is completed, an image reconstruction step is performed to obtain a reconstructed image (step S6). Image reconstruction can be performed by a known method. Then, the obtained reconstructed image is output to an output device such as a display (step S7).
By executing the above processing, the image processing method according to the present invention is completed.

Next, an image processing program according to another aspect of the present invention will be described. FIG. 4 is a diagram showing the configuration of the most preferable mode of the image processing program 20 according to the present invention together with the storage medium 10. In a preferred embodiment, the image processing program 20 according to the present invention includes a main module 22 that supervises processing, a projection data set acquisition module 24, a representative slice creation module 26, a myocardial center determination module 28, and a shift parameter calculation module 30. An image shift module 32, a region of interest information input module 34, a region of interest setting module 36, an image deletion information input module 38, an image deletion module 40, a non-region of interest deletion module 42, and a reconstruction module 44. , And an output module 46.
In a preferred embodiment, the image processing program 20 according to the present invention is provided by being stored in a storage medium 10 such as a hard disk, a CD-ROM, or a DVD. When the storage medium 10 storing the image processing program 20 is inserted into a reading device provided in the computer, the computer can access the image processing program 20 and can operate as an image processing device 200 described later. It becomes. Note that the image processing program 20 may be provided directly via a network, not via a storage medium.

  The projection data set acquisition module 24 causes the computer to execute processing related to step S1. The representative slice creation module 26 causes the computer to execute processing related to step S23. The myocardial center information input module 28 causes the computer to execute processing related to step S24. The shift parameter calculation module 30 causes the computer to execute processing related to step S25. The image shift module 32 causes the computer to execute processing related to step S2.

  The region-of-interest information input module 34 receives information regarding information (shape, size, etc.) regarding the region of interest that includes the myocardial region. The shape and size of the region of interest can be determined by a method similar to the method described in the image processing method described above. That is, a method of inputting a region of interest including a myocardial region set by a user on a projection image via a user interface, a predetermined shape such as a circle or an ellipse having a center at the center of rotation, It is possible to use a method of adjusting the size so as to include the region of interest. The size may be adjusted by (1) a method in which the user visually confirms on a display or the like. (2) The contour of the myocardial region is extracted in advance using a known method, and the contour is extracted. May be obtained as a circle having a radius larger than the maximum value after calculating the maximum value of the distance from the rotation center of each point constituting the. Also, by combining the methods (1) and (2) above, the size of the region of interest is automatically determined by the same method as (2), and this is displayed on a display etc., and the user confirms it, which is necessary. It is also possible to use a method of performing fine adjustment according to the above.

  The region-of-interest setting module 36 causes the computer to execute processing related to step S3. The image deletion information input module 38 receives information regarding a projection image to be deleted by an image deletion module 40 described later, that is, an image in which a highly integrated region other than the myocardial portion is included in the region of interest set above. This process can be performed by a method in which the user visually confirms each projection image on a display or the like, designates the corresponding image, and the computer accepts the input. Further, on the sinogram, a projection angle where a highly integrated part other than the myocardial part is present inside a line (two existing across the center) corresponding to the contour of the region of interest is detected, and the projection angle is A method of inputting a corresponding projection image as an image to be deleted may be used. The image deletion module 40 causes the computer to execute processing related to step S4. The non-interest region deletion module 42 causes the computer to execute processing related to step S5. The reconfiguration module 44 causes the computer to execute processing related to step S6. The output module 46 causes the computer to execute processing related to step S7.

  Next, an image processing apparatus according to another aspect of the present invention will be described. FIG. 5 is a functional block diagram of a preferred embodiment of the image processing apparatus 200 according to the present invention. In a preferred embodiment, the image processing apparatus 200 according to the present invention is organically coupled to a nuclear medicine image capturing apparatus 100 such as SPECT or PET. In a preferred embodiment, the image processing apparatus 200 functionally includes a projection data set acquisition unit 220, a representative slice creation unit 320, a parameter calculation unit 340, an input unit 360, an image shift unit 240, and a region of interest setting unit. (In FIG. 5, the ROI setting unit is displayed) 260, a determination unit 380, an image deletion unit 280, a non-interest region deletion unit 300, a reconstruction unit 400, and an output unit 420.

  The image processing apparatus 200 according to the present invention is typically configured as a computer that reads the image processing program 20, but may be configured by a combination of components having a function of mechanically executing each function. Is also possible.

  The projection data set acquisition unit 220 executes the process related to step S1. The representative slice creation unit 320 executes the process related to step S23. The parameter calculation unit 340 uses the information regarding the myocardial center input via the input unit 360, and executes the process according to step S25. The image shift unit 240 executes processing related to step S2. The region-of-interest setting unit (displayed as the ROI setting unit in FIG. 5) 260 uses the information regarding the shape and size of the region of interest input from the input unit 360, and executes processing related to step S3. The determination unit 380 determines whether each projection image is a projection image to be deleted by the image deletion unit 280 using information input via the input unit 360. For example, on the sinogram, it is confirmed whether or not a highly collected site other than the myocardial site exists inside a line (two existing across the center) corresponding to the contour of the region of interest. The projection image at the projection angle is determined as the projection image to be deleted. The image deletion unit 280 executes processing related to step S4. The non-interest region deletion unit 300 executes the process according to step S5. The reconfiguration unit 400 executes the process related to step S6. The output unit 420 executes the process related to step S7.

  Hereinafter, the present invention will be described in more detail with reference to examples. However, the contents of the present invention are not limited to these contents.

(Verification using myocardial phantom)
Using a myocardial phantom (RH2 type, manufactured by Kyoto Kagaku Co., Ltd., with an artificial defect in the myocardial lower wall), hexakis (2-methoxyisobutylisonitrile) technetium ( ^99m Tc) was injected into the myocardium so as to be 15 MBq.
In addition, in order to reproduce a case with high accumulation in the liver, hexakis (2-methoxyisobutyl isonitrile) technetium ( ^99m Tc) was added to the above myocardial phantom and the liver to a radioactivity concentration 10 times that of the myocardial part. A phantom infused with was also prepared.
Each created phantom was imaged using a SPECT apparatus. Table 1 shows the SPECT apparatus and main imaging conditions.

The obtained projection data set was subjected to image reconstruction using a normal FBP process. In addition, the phantom provided with a highly integrated site in the liver was processed according to the following method.
(1) For each projection image, the myocardial region was shifted so as to be at the center of rotation of the image (FIG. 6).
(2) An elliptical region of interest was set in the myocardial region.
(3) For an image in which the liver portion (highly integrated portion) is included in this region of interest, the entire image is subjected to mask processing, and the entire image is deleted. On the other hand, for the other images, mask processing was performed outside the region of interest, and the region outside the myocardium including the liver portion (highly integrated portion) was deleted. (Fig. 7)
(4) Using the projection data set after the mask processing, image reconstruction using FBP processing by the method described in Table 1 was performed.

An example of the obtained reconstructed image is shown in FIGS. In a phantom that does not have high accumulation in the liver, a good image without streak artifacts was obtained in the reconstructed image obtained by normal FBP processing, and the population defect in the lower wall was also well depicted (FIG. 8).
In the phantom exhibiting high accumulation in the liver, streak artifacts occurred in the reconstructed image obtained by normal FBP processing, and no population deficit in the lower wall was also depicted due to the influence (FIG. 9).
On the other hand, in the phantom exhibiting high accumulation in the liver part, in the example in which the mask process was performed prior to the normal FBP process, the streak artifact disappeared and the population defect of the lower wall part was clearly depicted (FIG. 10). .

  Next, a donut-shaped region of interest is set in the myocardial region on a cross-sectional image in the same cross section of the image obtained with a phantom that is highly integrated in the liver (FIG. 11), and the% uptake is obtained to vary. A comparison was made. As a result, the reconstructed image subjected to the normal FBP process had a variance of 97.45, whereas the reconstructed image subjected to the mask process prior to the FBP process had a variance of 48.54. .

  The present invention can be used in the fields of nuclear medicine image processing programs and nuclear medicine image processing apparatuses.

20 Image processing program 10 according to the present invention Storage medium 22 Main module 24 Projection data set acquisition module 26 Representative slice creation module 28 Myocardial center determination module 30 Shift parameter calculation module 32 Image shift module 34 Region of interest information input module 36 Region of interest setting module 38 Image deletion information input module 40 Image deletion module 42 Non-interest region deletion module 44 Reconstruction module 46 Output module 100 Nuclear medicine imaging device 200 Image processing device 220 Projection data set acquisition unit 240 Image shift unit 260 Region of interest setting unit (Fig. 5 is displayed as ROI setting section)
280 Image deletion unit 300 Non-interest region deletion unit 320 Representative slice creation unit 340 Parameter calculation unit 360 Input unit 380 Judgment unit 400 Reconstruction unit 420 Output unit

Claims (13)

A projection data set acquisition step in which a computer acquires a projection data set of a myocardial nuclear medicine image;
A region of interest setting step in which a computer sets a region of interest in a region including a myocardial region on each projection image included in the projection data set;
An image deletion step in which a computer deletes a projection image in which a highly integrated portion other than the myocardial portion overlaps the region of interest;
In the region of interest setting step, for each projection image included in the projection data set after the computer deletes the projection image in which the highly integrated portion other than the myocardial region overlaps the region of interest in the image deletion step A non-interest region deletion step of deleting a portion other than the set region of interest;
An image processing method, wherein the computer includes an image reconstruction step of performing image reconstruction using the projection data set after the non-interest region deletion step. Prior to the region of interest setting step, the computer further includes an image shift step of moving the center of the myocardial region to the center of rotation of the screen in each projection image included in the projection data set,
The image processing method according to claim 1, wherein the region of interest setting step sets the region of interest in a region including a myocardial region on each projection image in the projection data set after the image shift step processing.   The image processing method according to claim 1 or 2, wherein the deletion of a portion other than the region of interest in the non-region of interest deletion step is performed by mask processing.   The deletion of a portion other than the region of interest in the non-region of interest deletion step is performed by complementing a portion outside the region of interest with an average value of the background. The image processing method as described. In each projection image after the non-interest region deletion step, further includes a data interpolation step of complementing the deleted region with a value at the region of interest boundary on the image after the non-interest region deletion process,
The image processing method according to claim 1, wherein the image reconstruction step performs image reconstruction using a projection data set after the data complementing step. Projection data acquisition step for acquiring a projection data set of myocardial nuclear medicine images in a computer;
A region of interest setting step for setting a region of interest in a region including a myocardial region on each projection image included in the projection data set;
An image deletion step of deleting a projection image in which a highly integrated region other than the myocardial portion overlaps the region of interest;
The interest set in the region-of-interest setting step for each projection image included in the projection data set after deleting the projection image in which the highly integrated portion other than the myocardial region overlaps the region of interest in the image deletion step A non-interest area deleting step for deleting a part other than the area;
An image processing program that executes an image reconstruction step of performing image reconstruction using the projection data set after the non-interest region deletion step. Prior to the region-of-interest setting step, the computer further causes an image shift step to move the center of the myocardial region to the center of rotation of the screen in each projection image included in the projection data set,
The region-of-interest setting step causes the computer to execute processing for setting a region of interest in a region including a myocardial region on each projection image in the projection data set after the image shift step processing. Image processing program.   The image processing program according to claim 6 or 7, wherein the deletion of a part other than the region of interest in the non-region of interest deletion step is performed by causing a computer to execute a mask process.   The deletion of a part other than the region of interest in the non-region of interest deletion step is performed by causing a computer to execute a process of complementing a part outside the region of interest with an average value of the background. The image processing program according to any one of the above. Causing the computer to further execute a data complementing step of complementing the deleted region with a value at the region of interest boundary on the image after the non-interesting region deletion process;
The image processing program according to any one of claims 6 to 8, wherein the image reconstruction step causes the computer to execute image reconstruction using the projection data set after the data complementing step. A projection data set acquisition unit for capturing a projection data set of myocardial nuclear medicine images;
A region-of-interest setting unit that sets a region of interest in a region including a myocardial region on each projection image included in the projection data set;
An image deletion unit that deletes a projection image in which a highly integrated region other than a myocardial region overlaps the region of interest among each projection image included in the projection data set;
A non-interest region deleting unit that deletes a portion other than the region of interest set by the function of the region of interest setting unit on each projection image;
An image processing apparatus comprising: an image reconstruction unit configured to perform image reconstruction using a projection data set.   The image processing apparatus according to claim 11, further comprising an image shift unit having a function of moving the center of the myocardial region to the rotation center of the screen for each projection image included in the projection data set.   The image processing apparatus according to claim 11, further comprising a data complementing unit having a function of complementing the deleted region with a value at the region of interest boundary on the image after the non-interesting region deletion processing. .

Images