JP2000338251A - Ri-imaging device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique for improving the quality of an image obtained by reconfiguration. SOLUTION: In the RI-imaging device, a measurement means for measuring the number of radiations that are emitted from a radioactive isotope being administered to a specimen on a two-dimensional plane that is defined by X and Y axes is rotated around a subject 108 for measuring the two-dimensional RI image of the subject 108, the two-dimensional RI image is converted into frequency space for performing filtering processing, and the sectional image of the subject 108 is reconfigured from the two-dimensional RI image after the filtering processing. The RI-imaging device is provided with an approximation means for calculating the approximation function of high-frequency components based on a count value distribution curve when the number of measured radiations is converted into the frequency space and a means for setting the cut-off frequency of a filter for performing the filtering processing based on the approximation function and the count value distribution curve.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an RI imaging apparatus, and more particularly to a single photon tomography (Single Pho) which is an emission CT of a single emission nuclide.
ton Emission Computed Tom
The present invention relates to a technique which is effective when applied to automatic correction of statistical noise included in each image data (projection data) at the time of an OGR (SPECT).

[0002]

2. Description of the Related Art A conventional RI imaging apparatus uses a radioisotope (R radioisotope: R) administered to a subject.
For example, by rotating a scintillation camera for detecting radiation emitted from I) around the entire circumference of a subject, or by arranging a large number of scintillators in a ring like an X-ray CT apparatus, a radioisotope of the subject can be obtained. After measuring projection data that is a distribution image and reconstructing a three-dimensional image that is a three-dimensional distribution of RI from the projection data, based on the three-dimensional image, a tomographic image or a three-dimensional reconstructed image of the measurement site Was displayed.

[0003] In a technique called RI imaging, in which RI is administered to a subject and the number of radiation emitted from the RI is measured from outside the body using a scintillation camera or the like arranged around the subject, a technique called RI imaging that has a strong penetrating power is used. Generally, γ-rays that are hardly absorbed by γ-rays are used. However, since the γ-ray emitted from RI is about 100 to 300 counts / pixel, which is not so large as to be measured as a continuous quantity, it has been identified and counted for each photon incident on the scintillation camera.

It is known that when the count value is small, when the count value is N, a statistical fluctuation represented by N1 / 2 occurs, and the number of γ rays is about 100 to 300 counts / pixel. In a conventional RI imaging apparatus, 5-1
Statistical fluctuation of about 0% was included in the projection data, and each projection data was generally of low image quality including statistical noise, which is noise accompanying the statistical fluctuation.

Therefore, in a conventional RI imaging apparatus, a power spectrum obtained by transforming projection data into a frequency space by Fourier transform at the time of reconstruction calculation of projection data is converted to a low-pass filter (Butterworth). By performing a filtering process using a filter or the like, a high-frequency component of a power spectrum that is a statistical noise component is removed, and the image quality of a tomographic image that is a reconstructed image is improved. FIGS. 10 and 11 show this state.
Therefore, the amount of statistical noise converted into the high-frequency component of the power spectrum shown in FIG. 10 greatly affects the image quality of a tomographic image or a three-dimensional image obtained by reconstruction.

[0006]

SUMMARY OF THE INVENTION As a result of studying the above prior art, the present inventor has found the following problems. In general, it is known that the number of γ-rays emitted from a subject varies depending on the amount of RI administered to the subject, the measurement site, and the like. In order to improve the image quality of tomographic images, statistical noise is required. The high-frequency component of the power spectrum corresponding to the component also needs to be varied depending on the amount of RI, the measurement site, and the like.

For this reason, in the conventional RI imaging apparatus, the cut-off frequency of the low-pass filter is set arbitrarily by the examiner, and the examiner sets the cut-off frequency based on each statistical data and the like. By
This prevents the image quality of the tomographic image from deteriorating due to the statistical noise. However, in the conventional RI imaging apparatus,
The cut-off frequency, which is a parameter for optimizing the reconstructed tomographic image, must be determined by the examiner's experience or trial and error, and the image quality of the tomographic image that can be obtained due to the experience of the examiner differs. There was a problem that would occur. Further, if the examiner does not have sufficient experience, a tomographic image with stable image quality cannot be obtained, so that a large amount of time is required for diagnosis, and there is a problem that the diagnosis efficiency is reduced. .

On the other hand, in order to obtain a tomographic image having a desired image quality,
It is also possible to sequentially change the cutoff frequency to reconstruct a tomographic image, but in this case, a reconstruction operation consisting of a Fourier transform, a filtering process, and a back projection process must be performed a plurality of times. In addition, there is a problem that the diagnosis efficiency is further reduced.

As a method of solving this problem, for example, by increasing the amount of RI administered to a subject,
Although it is possible to increase the number of γ-rays, there is a problem that the amount of exposure of the subject increases. As another method, a method of increasing the imaging (acquisition) time of one piece of projection data can be considered. In this case, the time required to obtain projection data of the subject from all circumferential directions is considered. Therefore, there is a problem that the subject must be restrained for a long time and a large load is imposed on the subject.

An object of the present invention is to provide a technique capable of improving the quality of an image obtained by reconstruction. Another object of the present invention is to provide a technique capable of accurately correcting statistical noise. Another object of the present invention is to provide an RI imaging apparatus capable of improving diagnostic efficiency. The above and other objects and novel features of the present invention will become apparent from the description of the present specification and the accompanying drawings.

[0011]

SUMMARY OF THE INVENTION Among the inventions disclosed in the present application, the outline of a representative one will be briefly described.
It is as follows.

(1) The measuring means for measuring the number of radiation emitted from the radioisotope administered to the subject on a two-dimensional plane defined by the X axis and the Y axis is rotated around the subject to rotate the measuring means. In a RI imaging apparatus for measuring the two-dimensional RI image of the above, converting the two-dimensional RI image into a frequency space, performing a filtering process, and reconstructing a tomographic image of a subject from the two-dimensional RI image after the filtering process, Approximation means for calculating an approximate function of a high-frequency component based on a count value distribution curve obtained by converting the obtained radiation number into a frequency space, and performing the filtering process based on the approximate function and the count value distribution curve. Means for setting a cutoff frequency of the filter.

(2) In the RI imaging apparatus according to the above (1), the frequency setting means cuts the filter based on the straight line calculated from the high-frequency component and the fluctuation width calculated from the high-frequency component. Set the off frequency.

(3) R of the above (1) or (2)
In the I imaging apparatus, the approximation unit sets a cutoff frequency of the filter based on 高周波 of a high frequency side of a frequency range when a measurement radiation number of each two-dimensional RI image is converted into a frequency space.

According to the above-mentioned means (1) to (3),
First, a two-dimensional RI image acquired by rotating the measuring means around the subject (imaging) is converted into a frequency space, and a low-pass filter for filtering a power spectrum, which is a distribution for each frequency component obtained by frequency components, is obtained. Prior to the integration, the frequency calculated from the high-frequency component of the power spectrum by the approximation means is used as the cut-off frequency of the low-pass filter, so that the cut-off frequency can be automatically set. It is possible to improve the image quality of the obtained tomographic image.

Accordingly, the examiner can always obtain a tomographic image having an image quality of a certain level or more regardless of his or her experience, so that the time required for the examination can be shortened,
The burden on the subject can be reduced. Further, since the time required for the examination can be reduced, the diagnosis efficiency can be improved.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail with reference to the drawings together with embodiments (examples) of the invention. In all the drawings for describing the embodiments of the present invention, components having the same functions are denoted by the same reference numerals, and their repeated description will be omitted.

FIG. 1 is a view for explaining a schematic configuration of an emission CT apparatus which is an RI imaging apparatus according to one embodiment of the present invention. In particular, FIG. 1A shows an emission CT apparatus according to this embodiment. FIG. 1B is a front view of the emission CT apparatus viewed from the body axis direction of the subject, and FIG. 1B is a side view of the emission CT apparatus of the present embodiment viewed from a direction perpendicular to the body axis direction of the subject.

In FIG. 1, 101 is a detector, 102 is a rotating plate, 103 is a frame, 104 is a detector support arm,
05 is a rotation axis, 106 is a collimator, 107 is a subject support base, 108 is a subject, 109 is control means, 110 is image processing means, 111 is a console, 112 is a keyboard, 1
Reference numeral 13 denotes a display device.

The detector 101 is, for example, a well-known scintillation camera. In the present embodiment, it is a two-dimensional scintillation camera. Therefore, in the present embodiment, a two-dimensional image (projection data) in which γ-rays emitted from RI distributed three-dimensionally in the body of the subject 108 are projected on the incident surface of the detector 101, that is, the surface on the collimator 106 side. Is detected as The detector 101 is supported by a detector support arm 104.

The rotating plate 102 is a well-known rotating plate for rotating the detector 101 supported by the detector supporting arm 104 around the body axis of the subject 108, and the rotating plate 102 is connected to the rotating plate 102. . Further, the rotating plate 102 rotates about the rotating shaft 105 as a center of rotation.

The frame 103 is a well-known frame that rotatably supports the rotary plate 102 about a rotary shaft 105 as a center of rotation. The rotary plate is mounted on a rotary shaft of a well-known rotary drive mechanism (not shown) provided on the frame 103. 102 is arranged. In this rotation drive mechanism, based on a rotation instruction from the console 111, the control unit 109 controls a well-known drive unit such as a motor of the rotation drive mechanism to control the rotation of the rotary plate 102.

The detector support arm 104 includes a rotating plate 102
One end on the side connected to the motor is disposed on a well-known support arm up-and-down mechanism including a well-known gear mechanism and a pulse motor (not shown). By controlling the rotation of the pulse motor by the control means 109, the detector is supported. The distance from the center of rotation of the detector 101 disposed at the other end of the arm 104 is adjustable.

The collimator 106 is a well-known collimator, and is attached to the front of the input surface of the detector 101.

The subject support table 107 is a well-known subject support table. For example, the support table can be moved back and forth and up and down by a well-known support table moving mechanism. However, in the present embodiment, the control means 109 controls the support base moving mechanism.

The control means 109 is a well-known control means for controlling the operation of the motor of each of the above-mentioned mechanisms in order to perform imaging from the position designated by the console 111, and is connected to the console 111, for example. It can be realized by a program operating on a known information processing apparatus and a drive control input / output of the information processing apparatus. Also, the control means 109
Is connected to each sensor (not shown).

The image processing means 110 counts the number of γ-rays detected by the detector 101 and stores it in a storage means (not shown), and an integrated value of the counted value based on the photographing mode input from the console 111. Alternatively, the program comprises a reconstructing means for reconstructing a tomographic image from the count value, for example, a program operating on a known information processing device connected to a console, and a magnetic disk device connected to the information processing device And a well-known external storage device such as an optical disk device. As a counting method in the present embodiment, a frame mode in which count values for each preset gate time are sequentially stored as projection data, and a γ-ray generation coordinate is set to a predetermined sampling time mark or a synchronization signal of a biological signal And a list mode for recording in chronological order. The details of the image processing means will be described later.

The console 111 is a well-known console, and is connected to the detector 101, control means 109, keyboard 112 and display device 113. The console 111 is composed of, for example, a well-known information processing device, and performs storage of data measured by the detector 101, data conversion, and the like.

The keyboard 112 is a well-known keyboard, and is connected to the console 111. The display device 113 is a known display device, for example, a television monitor using a known CRT. This display device 113 is
Connected to 1.

Next, the operation of capturing a tomographic image in the emission CT apparatus according to the present embodiment will be described with reference to FIG.

First, the subject 108 is placed on the subject support 107.
Set up in recumbent position. Next, as the imaging conditions for the emission CT imaging, for example, the imaging region, the number of divisions of the entire circumference of the subject, and the collection time (imaging time) of the projection data are set from the keyboard 112. Here, the subject 10
After the administration of a radioisotope (radio isotope: RI) to the keyboard 8, when the collection of projection data is instructed from the keyboard 112 of the console 111, the control unit 111 controls the rotating plate based on the conditions input from the keyboard 112. 10
The step rotation of 2 and the collection of projection data at the rotation angle, that is, the imaging of the emission image, are sequentially performed for the entire circumference of the subject 108.

Next, with respect to the emission image for the entire circumference,
First, the image processing unit 110 calculates a cutoff frequency of a Butterworth filter, which is a low-pass filter, from a high-frequency component of a power spectrum (Fourier spectrum) obtained by transforming projection data into a frequency space by Fourier transform. Next, the image processing unit 110 sets the cut-off frequency of the low-pass filter to the cut-off frequency determined in the above-described step, and performs filtering to reduce statistical noise that is a high-frequency component of the power spectrum. Next, the image processing unit 110 improves blur by performing filter processing using a Wiener filter or the like which is a high-frequency emphasizing filter. Next, after performing the inverse Fourier transform, the image processing unit 110 reconstructs a tomographic image by performing a so-called back-projection operation of sequentially adding the data, and outputs the tomographic image to the display device 113 to display the tomographic image on the display screen. To be displayed. As is clear from the above description, in the RI imaging apparatus according to the present embodiment, a well-known filter back projection method is used as an algorithm for reconstructing a tomographic image from two-dimensional projection data. Also, an appropriate value is assigned as an initial value to all pixels, and the difference between the actual measured value of the actual projection and the assigned initial value projection value is determined in each of the horizontal direction (X direction) and the vertical direction (Y direction). Needless to say, another reconstruction algorithm such as a successive approximation method for performing correction by a value divided by a number may be used.

FIG. 2 is a block diagram for explaining a schematic configuration of the image processing means of the present embodiment. 201 is a Fourier transform means, 202 is a filter processing means, 203 is an inverse Fourier transform means, and 204 is a back projection. Means, 205 is a linear regression means (approximation means), 206 is an estimation means, and 207 is an intersection calculation means. However, each unit illustrated in FIG. 2 can be realized by a program that operates on a known information processing apparatus that configures the emission CT apparatus according to the present embodiment. In addition, the main part according to the present invention,
As shown by the dotted line, the linear regression means 205 and the estimation means 20
6 and the intersection calculation means 207.

In FIG. 2, a Fourier transform means 201 is a well-known Fourier transform means for sequentially transforming input projection data into a frequency space, and in this embodiment,
The power spectrum obtained by the conversion is output to the filter processing means 202 and the linear regression means.

The filter processing means 202 includes a filter storage means for storing a Butterworth filter and an intersection calculation means 207 for calculating a cutoff frequency of the Butterworth filter.
Filter constant setting means for setting the frequency calculated by
The multiplication unit multiplies the power spectrum of the projection data by the Butterworth filter set by the filter constant setting unit, and outputs the filtered power spectrum as the multiplication result to the inverse Fourier transform unit 203.

The inverse Fourier transform means 203 is a well-known inverse Fourier transform means for performing a Fourier transform on the power spectrum after the filter processing, and outputs a value obtained by the inverse transform to the back projection means 204.

The back projection means 204 generates a tomographic image at an arbitrary position in the measurement area by sequentially adding the values obtained by the inverse Fourier transform means 203 over the entire circumference of the subject 108. Projection means for outputting the obtained tomographic image to a display device and displaying it on a display screen. At this time, it is needless to say that the tomographic image can be stored in an external storage means (not shown) according to an instruction from the console 111.

The linear regression means 205 is a means for linearly regressing a power spectrum in a frequency region larger than a preset frequency from the power spectrum of the projection data obtained by the Fourier transform by, for example, a least square method. In the present embodiment, one-fourth of the high-frequency side of the power spectrum obtained by the Fourier transform means 201 is linearly regressed. However, it is needless to say that the regression of the power spectrum in the linear regression means is not limited to a straight line, but may be a two-dimensional or more curved regression.

The estimating means 206 is a calculating means for calculating the standard deviation SD of the power spectrum in the same region as the linear regression means 205. In this embodiment, the calculated standard deviation SD is output to the intersection calculating means 207. I do.

The intersection calculating means 207 calculates a value obtained by doubling the standard deviation SD calculated by the estimating means, and calculates
Means for setting the variation range width of the regression line obtained in step 05, means for performing 5-point smoothing processing as interpolation processing for the power spectrum, and means for setting the regression line and standard deviation to 2 points.
Means for calculating an intersection with the power spectrum when the variation range width obtained by the multiplication is extended to the low frequency side, and a line obtained by adding twice the standard deviation to the regression line and the power spectrum. The intersection is output to the filter processing means 202 as the cutoff frequency of the Butterworth filter. However, in the present embodiment, the linear regression means 205,
The estimating means 206 and the intersection calculating means 207 constitute a cutoff frequency determining unit.

Next, FIG. 3 shows an operation flow for explaining the operation of the image processing means of the present embodiment.
The operation of the image processing means according to the present embodiment will be described based on FIG.

The start of this flow is the end of the imaging of the projection data from the entire circumference of the subject 108. First, the Fourier transform means 201 performs the projection from the storage means (not shown) of the image processing means 110 with the rotation angle of 0 degree. Data is sequentially read from the data, and the projection data is Fourier-transformed, and then the power spectrum of the projection data obtained by the Fourier transformation is output to the filtering means 202 and the linear regression means 205 (step 301). Next, the linear regression means 205 performs a linear regression by a least squares method on the data on the high frequency side ４ in the power spectrum (step 302). Next, the estimating means 206 calculates the standard deviation SD for the data on the high frequency side 1/4, that is, the fluctuation width of the statistical noise, and outputs the obtained standard deviation SD to the intersection calculating means 207 (step 303). In the intersection calculation means 107, first,
As a power spectrum interpolation process, a well-known five-point smoothing process is performed (step 304). Next, the intersection calculating means 207 sets a range in which the regression line calculated by the linear regression means 205 has a variation width twice as large as the standard deviation SD as a power spectrum portion of the statistical noise. Next, the intersection calculating means 207 calculates the intersection with the power spectrum by extending the straight line obtained by the regression line + 2SD, which is a range deviating from the fluctuation range, to the low frequency side, and uses the frequency of this intersection as the cutoff frequency. Is output to the filter processing means 202 (step 305). In the filter processing means 202,
After setting the cutoff frequency of the Butterworth filter to the cutoff frequency calculated by the intersection calculation means, the power spectrum is integrated with the Butterworth filter after setting to remove statistical noise in the power spectrum. Is output to the inverse Fourier transform means. In the inverse Fourier transform means, after transforming the power spectrum after removing the statistical noise into a real space,
The converted data is sequentially added by the back projection calculation means 204 to reconstruct a tomographic image.

FIG. 4 shows the manner in which the intersection of the power spectrum with the straight line of the regression line + 2SD is calculated in step 305 described above. As is clear from this figure, the low frequency component in the power spectrum is calculated. Is larger when the frequency space is converted, the intersection of the power spectrum and a line having a variation width twice as large as the standard deviation SD in the regression line is obtained by adding twice the standard deviation SD to the regression line. It can be calculated from the intersection of the power spectrum with the straight line of the regression line + 2SD, which is the calculated straight line.

FIG. 5 and FIG.
2 shows the power spectrum after filtering, and as is apparent from this figure, the high-frequency component of the power spectrum is always removed at the optimum cutoff frequency regardless of the measurement value that is the source of the power spectrum. It becomes possible.

FIG. 7 is a diagram showing a reconstructed image (tomographic image) in the RI imaging apparatus of the present embodiment, and FIG. 7A is calculated by the cut-off frequency determining section of the present embodiment. FIGS. 7B to 7D are tomographic images to which a cutoff frequency is applied, and FIGS. 7B to 7D are tomographic images to which frequencies before and after the calculated cutoff frequency are applied. As is clear from this figure, in the RI imaging apparatus of the present embodiment, it is possible to accurately correct statistical noise,
The image quality of the tomographic image obtained by the reconstruction can be improved.

FIG. 8 is a diagram showing the suitability of the cut-off frequency in the RI imaging apparatus of the present embodiment. The horizontal axis shows the frequency calculated by the cut-off frequency determining unit, and the vertical axis shows the frequency. The cut-off frequency set value by a sufficiently experienced examiner is shown. However, FIG.
Is the case where the number of samples N is 16.

As is apparent from this figure, the cut-off frequency (automatically calculated value) calculated by the cut-off frequency determining section of the present embodiment and the set value (experience calculated value) by an experienced examiner are different. There are largely three out of 16 cases. By using the RI imaging apparatus of the present embodiment, the cutoff frequency can be automatically set with a probability of 81.3%, and the performance is sufficient for practical use. I found it. The correlation coefficient r is 0.911, and a straight line 801 indicating the relationship between the automatically calculated value and the experience calculated value when the automatically calculated value is X and the experience calculated value is Y is Y = 1.21X. −0.04, which is an ideal straight line of 80
Since it almost coincides with 2, it can be judged that the performance is sufficient for practical use.

As described above, the RI of the present embodiment is
In the imaging apparatus, the range in which the intersection calculation means 207 gives a variation width twice the standard deviation SD to the regression line calculated by the least squares method is defined as the power spectrum portion of the statistical noise, and the range deviating from the variation width is defined as an image. Power spectrum part required for conversion. That is, by setting the frequency at which the power spectrum deviates from twice the fluctuation width of the standard deviation SD in the regression line as the cutoff frequency in the filter processing, it is closely related to the elimination of statistical noise which conventionally requires sufficient experience. Since the cutoff frequency of the Butterworth filter can be set accurately, the image quality of the tomographic image obtained by the reconstruction can be improved.

In addition, since it is possible to easily obtain a tomographic image having an image quality of a certain level or more at all times regardless of the experience of the examiner, it is possible to shorten the time required for the examination and reduce the burden on the subject. It becomes possible to reduce. Further, since the time required for the examination can be reduced, the diagnosis efficiency can be improved.

At this time, in the RI imaging apparatus according to the present embodiment, the width of variation with respect to the regression line is set to twice the standard deviation SD, and the intersection calculating means 207 calculates the smoothed power spectrum with the regression line. Since it is configured to calculate the intersection with the power spectrum,
It is possible to reduce the influence of a peak value or the like due to a measurement error appearing in the power spectrum, and it is possible to improve the calculation accuracy of the cutoff frequency.

In the present embodiment, the subject 1
The correction of the projection data by the absorption of γ-rays by 08 is not particularly described. However, prior to imaging of the emission image, the projection data is irradiated from the radiation source arranged opposite to the detector 108 via the object 108, and the object 108 is irradiated. Needless to say, a transmission image in which the transmitted radiation dose is collected is captured from the entire circumference of the subject 108, and the transmission image and the emission image can be corrected by a method of correcting the γ-ray absorption coefficient of the subject 108 from the transmission image and the emission image. Nor.

In the present embodiment, the subject 1
Although the tomographic image is reconstructed by performing the reconstruction operation after collecting the projection data from the entire circumference of 08, the invention is not limited to this, and the tomographic image is reconstructed together with the collection of the projection data. Needless to say, it is good.

Further, in the present embodiment, the case where the range of variation with respect to the regression line is set to twice the standard deviation SD has been described, but the range of variation is not limited to this and may be set to another value. It goes without saying that this may be done. For example, FIGS. 9A to 9D are views showing reconstructed images when the variation width is changed to 1 to 4 times the standard deviation SD for the same measurement image. 9 (b),
As is apparent from (c), the fluctuation range is set to the standard deviation SD of 3
It can be seen that even when the frequency is multiplied or quadrupled, the calculation accuracy of the cutoff frequency fc does not greatly vary. On the other hand, as is apparent from FIG. 9D, when the variation width is set to one time of the standard deviation SD, the cutoff frequency is set to 0.
27 greatly fluctuates. Therefore, if the variation width is set to be larger than one time of the standard deviation SD and about four times or less, it is possible to improve the image quality of the tomographic image obtained by the reconstruction. However, FIG.
The image quality of the tomographic image obtained most when the variation width is set to about twice the standard deviation SD from the tomographic image shown in (d) can be improved.

As described above, the invention made by the present inventor is:
Although specifically described based on the embodiments of the present invention, the present invention is not limited to the embodiments of the present invention, and it is needless to say that various modifications can be made without departing from the gist of the present invention. .

[0055]

The effects obtained by typical ones of the inventions disclosed in the present application will be briefly described as follows. (1) The image quality of the image obtained by the reconstruction can be improved. (2) Statistical noise can be accurately corrected. (3) The diagnostic efficiency can be improved.

[Brief description of the drawings]

FIG. 1 is a diagram for explaining a schematic configuration of an emission CT apparatus which is an RI imaging apparatus according to an embodiment of the present invention.

FIG. 2 is a block diagram illustrating a schematic configuration of an image processing unit according to the present embodiment.

FIG. 3 is an operation flow for explaining the operation of the image processing means of the present embodiment.

FIG. 4 is a diagram illustrating a state when an intersection point between a regression line + 2SD line and a power spectrum is calculated in the present embodiment.

FIG. 5 is a diagram for explaining a power spectrum after filtering by a filter processing unit of the present embodiment.

FIG. 6 is a diagram for explaining a power spectrum after filtering by the filter processing unit of the present embodiment.

FIG. 7 is a halftone image in which a reconstructed image (tomographic image) in the RI imaging apparatus of the present embodiment is displayed on a display device.

FIG. 8 is a diagram illustrating suitability of a cutoff frequency in the RI imaging apparatus according to the present embodiment.

FIG. 9 is a graph showing a standard deviation SD of a fluctuation width for the same measurement image;
5 is a photograph of a halftone image on a display device, the reconstructed image being displayed when the image is changed to 1 to 4 times.

FIG. 10 is a diagram for describing a power spectrum obtained by Fourier transform.

FIG. 11 is a diagram for explaining a degree of deterioration of a reconstructed image due to a difference in a statistical noise amount in collected data.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 101 Detector 102 Rotating plate 103 Frame 104 Detector support arm 105 Rotation axis 106 Collimator 107 Subject support base 108 Subject 109 Control means 110 Image processing means 111 Operator console 112 Keyboard 113 Display device 201 Fourier transform means 202 Filter processing means 203 inverse Fourier transform means 204 back projection means 205 linear regression means 206 estimation means 207 intersection calculation means

 ──────────────────────────────────────────────────続 き The continuation of the front page F term (reference) 2G088 EE02 FF04 GG20 JJ06 KK03 KK05 KK24 KK32 KK33 LL11 LL12

Claims (1)

[Claims] 1. A measuring means for measuring the number of radiation emitted from a radioisotope administered to a subject on a two-dimensional plane defined by an X axis and a Y axis is rotated around the subject, and An RI imaging apparatus that measures a two-dimensional RI image, converts the two-dimensional RI image into a frequency space, performs a filtering process, and reconstructs a tomographic image of a subject from the two-dimensional RI image after the filtering process. Means for calculating an approximate function of a high-frequency component based on a count value distribution curve obtained by converting the number of radiations into a frequency space, and a filter for performing the filtering process based on the approximate function and the count value distribution curve. Means for setting a cut-off frequency of the imaging device.