JP3101308B2 - SPECT device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial application field) The present invention relates to a radioisotope (hereinafter referred to as a radioisotope) administered to a subject.
SPECT (S) that reconstructs an image based on detection data obtained by detecting γ-rays emitted from RI) by a detector such as a gamma camera and obtains a RI concentration distribution tomographic image of the subject
The present invention relates to an ingle-photon emission computed tomography device, and more particularly to a fan beam SPECT device that obtains detection data using a fan beam collimator.

(Prior Art) This type of SPECT device includes a device that detects a γ-ray using a fan beam collimator as shown in FIG.
Some systems detect γ-rays using a parallel beam collimator as shown in the figure. When a fan beam collimator is used, there is an advantage that gamma rays emitted from the body of the subject can be efficiently detected. For example, when three fan beam collimators are provided, FIG.
As shown in FIG. 13, the γ-ray incident surface 100a of the fan beam collimator rotates around the subject P, and detects γ-rays emitted from the RI administered to the subject P from the 360 ° direction.
In FIG. 13, reference numeral 101 denotes a rotation center.

As a method of reconstructing an image from the fan beam data detected using the fan beam collimator in this manner, the fan beam data is converted into substantially the same data as the parallel data detected using the parallel beam collimator ( After that, the image is reconstructed by the parallel data reconstruction system using the parallel data, and the image is reconstructed by the fan beam data reconstruction system using the fan beam data as it is. There is a way to configure.

In the former case, as shown in FIG. 14, after collecting all the fan beam data obtained by detecting the γ-rays emitted from the subject P using the fan beam collimator 100, if necessary, uniform the data to these data. Preprocessing such as gender correction, center position correction, and spatial filter processing is performed to update the result. Then, the pre-processed fan beam data is taken out, and an image is reconstructed based on parallel data obtained by performing a fan-para transformation and post-processing, which will be described later. Get the data. On the other hand, in the latter case, as shown in FIG. 15, the collected fan beam data is directly taken out by the direct method reconstruction algorithm, preprocessed as necessary, reconstructed, and reconstructed. Obtain configuration data.

The above-mentioned fan-para conversion means finding data corresponding to parallel data from all collected fan beam data,
This is a process of interpolating and creating missing parallel data. For example, as shown in FIG.
When X _F (θ, ψ) and the parallel data are expressed by _XP (θ, t) (where 101 indicates the center of rotation and 102 indicates the focus of the fan beam), the following equation is used between the two data. Holds.

That is, if there is infinite fan beam data, necessary parallel data can be searched for and obtained from the above equation using the above equation. In practice, there is often no data that matches exactly, and parallel data to be obtained by interpolation is created.

(Problems to be Solved by the Invention) However, when the fan-para conversion algorithm is used in the above-described conventional technology, the fan-para conversion process cannot be performed unless all the fan beam data exists. Has a large number of times of reading collected data, it takes time to obtain parallel data, and SP that can collect data for multiple slices at the same time
ECT devices have the disadvantage of being unsuitable. Furthermore, when preprocessing is performed, data in the memory is written to update the original data (when the original data is needed, it is copied and used), so the data processing is complicated.
There is also a problem that data processing time tends to be long. On the other hand, when the direct method reconstruction algorithm is used, the calculation at the time of data processing is very complicated, and a conventional SPECT having a parallel data system (a system having a reconstruction program for reconstructing parallel beam data) is used. There was the disadvantage that the device could not be used.

The present invention has been made in order to solve the above-described problems of the related art, and its purpose is to simply collect fan beam data in a short time and perform data conversion.
An object of the present invention is to provide a SPECT apparatus capable of reconstructing an image using a parallel data system.

[Configuration of the invention]

(Means for Solving the Problems) In order to achieve the above object, according to the present invention, fan beam data corresponding to γ-rays emitted from a radioisotope administered to a subject is converted into a fan beam collimator. A two-dimensional detection means for detecting two-dimensionally in the X direction and the Y direction via X-ray; and X based on fan beam data in the X direction detected by the two-dimensional detection means. Direction parallel data detecting means, and fan beam data in the Y direction detected by the two-dimensional detecting means corresponding to the position of the parallel data detected by the X direction parallel data detecting means are detected as parallel data in the Y direction. Y direction parallel data detection means, and the X direction parallel data detection means and the Y direction parallel data detection means respectively detected Comprising an image reconstruction means reconstructs an image based on Parallel data.

(Operation) According to the present invention, the two-dimensional detecting means detects fan beam data corresponding to γ-rays emitted from the radioisotope administered to the subject via the fan beam collimator. At this time, the fan beam data is two-dimensionally converted, for example, in the X direction (channel direction) in which the detection elements of the fan beam collimator are juxtaposed, and in the Y direction in the body axis direction (slice direction) of the subject. To detect.

The X-direction parallel data detection means first detects parallel data of a predetermined slice plane based on the fan beam data in the X direction detected by the two-dimensional detection means.

Further, the Y-direction parallel data detecting means converts the Y-direction fan beam data detected by the two-dimensional detecting means corresponding to the position of the parallel data detected by the X-directional parallel data detecting means into parallel data in the Y direction. Detected as

Then, the image reconstructing means reconstructs an image based on the parallel data detected by the X direction parallel data detecting means and the Y direction parallel data detecting means, respectively.

As a result, the time required for collecting parallel data can be reduced, and the time required for image reconstruction can be reduced.

(Example) Hereinafter, an example of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram schematically showing a configuration of a SPECT apparatus according to one embodiment of the present invention.

Referring to FIG. 1, a SPECT apparatus 1 includes a gamma camera 3 for detecting, via a fan beam collimator 2, a gamma ray emitted from an RI administered to a subject P, and a gamma camera 3
Fan beam data collection / preprocessing means 4 for collecting fan beam data detected in step 1 and performing preprocessing on the data to improve the quality of a captured image; and collecting the fan beam data in parallel with the collection of fan beam data. The fan-parallel data is created from the fan-beam data of the image data, and the parallel data written in the image memory 5 is reproduced by using the fan-parallel conversion means 6 as data conversion means for arranging the parallel data in the image memory 5. It comprises an image reconstructing means 7 to be constituted, and a display 8 for displaying an RI density distribution tomographic image in the subject P in accordance with a video signal sent from the image reconstructing means 7.

When imaging the subject P, the gamma camera 3 detects γ-rays from a 360 ° direction while the fan beam collimator 2 and the gamma camera 3 rotate around the subject P, and the fan beam data is collected and preprocessed by the fan beam data. Sent to the means 4. Next, the operation of the fan beam data collection / preprocessing means 4 and the fan-para conversion means 6 will be described with reference to FIG.

First, select collected data to obtain a captured image, set fan-para conversion conditions (step 1), wait until unprocessed collected data is collected (step 2), and then collect collected fan beam data. Read one frame (step
3). The read data is subjected to preprocessing of uniformity correction (step 4), center position correction (step 5), and spatial filter correction (step 6). That is, in the uniformity correction, the data of the planar image of the uniform surface radiation source is previously collected using the fan beam collimator 2, and the sensitivity component of the fan beam collimator 2 is removed from the data to remove the change in uniformity. Only data is created, and uniformity correction data is created by taking the reciprocal of this data. Then, the non-uniform component of the gamma camera 3 is removed by multiplying the collected data by the uniformity correction data.

In the center position correction, the amount of deviation of the center of the field of view of the gamma camera 3 from the true rotation center is measured in advance for each data collection angle, the correction amount is obtained from this, and the image of the collected data is shifted by the correction amount. Correct the upper rotation center to the correct position. In the spatial filter processing, the acquired data is multiplied by a spatial filter for removing a noise component or the like. The filter type and filter matrix can be arbitrarily selected.

The fan-para conversion unit 6 performs fan-para conversion on the data on which the pre-processing has been performed (step 7). Here, this fan-para conversion will be described in detail. SPEC
In the T apparatus, since the data collection time per projection is long, in this embodiment, the collected data is sequentially subjected to the fan-parallel conversion process using the idle time of the CPU during data collection. In the fan-para conversion process of this embodiment, data that can be converted from the collected data of two frames into parallel data is obtained, parallel data is created based on this data, and the created parallel data is stored in an address (file) corresponding to the converted image. 3.) Writing into the image memory 5.

For example, as shown in FIG. 4 (a), the data of the pixel adjacent to the fan beam data F ₁ is F ₂ , the data of the next collection angle is F ₃ for the data F ₁ , Assuming that the data is F4, ₄ arranged as shown in FIG.
The parallel data P is obtained by interpolating from the point data F _{1 to} F ₄ to create parallel data as shown in FIG. That is, when the fan beam data is represented by the angle of the beam and the position of the pixel of the camera, it becomes as shown in FIG. The parallel data is created by interpolation from the four points of fan beam data (black points), and nothing is performed when there is no parallel data as in the area C.

This process is performed for all pixels of fan beam data of two consecutive frames. According to this method, two consecutive
If there is one fan beam data, it is possible to partially perform the fan-para conversion. If the same processing is performed on the fan beam data for the entire circumference, all parallel data is obtained and the fan-para conversion is completed.

In other words, when there is another one-frame data collected and preprocessed immediately before the one-frame data to be processed for the one-frame data to be processed, the fan-parallel conversion means 6 can first generate parallel data that can be created from the two-frame data. Examine the data and perform fan-para conversion. Here, a method of converting fan beam data into parallel data will be described.
As shown in FIG. 6, if the parallel data corresponding to the fan beam data Pf (θ, L) is P (ψ, r), the following equation holds.

tan μ = L / D μ = A tan (L / D) ψ = θ + μ (1) r = D ₀ × sin μ (2) where μ in the above equation is the opening angle of the fan beam,
D ₀ is the distance from the focus 11 of the fan beam to the center of rotation 10,
D indicates the distance from the focus 11 of the fan beam to the scintillator surface (detection surface) of the gamma camera 3. The above (1),
The parallel data corresponding to the fan beam data can be obtained by using the relational expression of θ, L, ψ, r expressed by the equation (2).

Then, as shown in FIG. 7, in the diagram in which the fan beam data is represented by the fan beam angle θ and the pixel r, four fan beam data P ₁ (θ
₁ , r ₁ ), P ₂ (θ ₂ , r ₁ ), P ₃ (θ ₃ , r ₃ ), P ₄ (θ ₄ , r ₃ ) Whether parallel data P (θ, r) exists Find out. First, it is checked whether or not r that satisfies the following equation exists.

r ₁ ≦ r <r ₃ (3) When r is present, the angles θ ₅ and θ ₆ in FIG. 7 are obtained using the following equations.

Next, it is checked whether or not θ that satisfies the following equation exists.

θ ₅ ≦ θ <θ ₆ (4) If r and θ satisfy the above equations (3) and (4), the parallel data P (θ, r) exists. However, a plurality of parallel data may exist depending on the condition. At this time, the parallel data is obtained in all cases. When the parallel data exists, the parallel data is created by area interpolation from the fan beam data of the four surrounding points.

At this time, a weight for multiplying the four data is calculated. For example, when weights a ', b', c ', and d' are set in proportion to the lengths a, b, c, and d in FIG. 7, these weights are first set to the normalization condition a '+ b' = 1. , c ′ + d ′ = 1. In this case, since c ′: d ′ = c: d = r−r ₁ : r ₃ −r, a ′, b ′, c ′, and d ′ obtained by the following equation can be used as weights.

Therefore, the parallel data P (θ, r) can be obtained using the following equation.

P (θ, r) = b ′ × d ′ × P ₁ + a ′ × d ′ × P ₂ + b ′ × c ′ × P ₃ + a ′ × c ′ × P ₄ The position of the parallel data in the x direction by the above method. However, since the data is parallel in the y direction (slice direction), when the parallel data position in the x direction is obtained, the parallel data can be obtained at the same position as the fan beam data in the y direction. Thus, for example, as shown in FIG. 8, two consecutive frames and i
Data of j pixels and j + 1 pixels in the x direction are extracted in the y direction by the image size (Y size) from the +1 frame fan beam data, and fan-parallel conversion data for a plurality of slices is created at a time. The obtained parallel data is written in the work area (image memory 5) as shown in FIG. 9 (step 8). Thereby, fan-para conversion of a plurality of slices can be easily performed.

The above operation is repeated until the collected data for one repeat is processed, and the collected data for one repeat is subjected to fan-parallel conversion (step 9), and post-processing is performed on the parallel data (s).
tep10). That is, after the fan-parallel conversion, the image size in the x direction and the image size in the y direction may be different depending on the processing conditions. This is repeated until all repeat data is processed (step1
1).

The image reconstructing means 7 reconstructs an image using the parallel data in the image memory 5 as shown in FIG.
First, after setting processing conditions, specifying a slice cutout position, and preprocessing (step 21), preprocessing for reconstructing an image is performed (step 22), and data for obtaining a slice image is created (step23). ). Perform convolution and back projection processing on these data (step2
4) Reconstruct the image. The processes in steps 23 and 24 are repeated until data for one repeat is processed (step 25), and the pre-process in step 22 and these processes are repeated until data for all repeats is processed (step 26). Then, a video signal containing the obtained reconstructed image is sent to the display 8, and a slice image showing the RI density distribution in the subject P is displayed on the screen of the display 8.

In this manner, in the present embodiment, collection, pre-processing, and fan-para conversion of fan beam data can be easily performed in a short time, and pre-processing for improving the image quality of an image is performed on data. Can be easily performed on the fan beam data without performing the writing process. Moreover, since the fan beam data only needs to be read once at the time of the fan-parallel conversion, it is easy to add various pre-processing and change the pre-processing conditions when reading the data.

Here, a fan beam SPECT device can be easily obtained simply by providing a fan beam data collection device for the parallel beam SPECT device and replacing the parallel beam collimator with a fan beam collimator. Strictly speaking, the information of the used fan beam collimator must be added to the fan beam data, but it is not necessary to largely change the configuration of the apparatus itself. That is, by providing the fan beam data collection / pre-processing means 4 and the fan-para conversion means 6 as described above, a conventional SPECT apparatus having a parallel data reconstructing system can be easily applied to the present invention apparatus using its functions as it is. Can be remodeled. In this case, the unique processing of the fan beam SPECT device is added as a function of the fan beam data collection / preprocessing means 4 or the fan para conversion means 6 so as not to affect other device configurations and systems. I do.

Further, in the apparatus of the present embodiment, the fan-para conversion software has a fan-para conversion algorithm suitable for SPECT data collection, so that the system configuration is simple. Further, if the conventional parallel data SPECT apparatus is modified and used as described above, only the operation of the fan-parallel conversion is added, so that the operation change is small and the burden on the operator can be reduced.

The fan beam data collection / pre-processing means 4 and the fan-para conversion means 6 also include a software package for performing the above-mentioned processing using computer software.

Although the embodiment of the present invention has been described above, the present invention is not limited to this, and it goes without saying that various modifications can be made.

〔The invention's effect〕

The SPECT apparatus of the present invention has the above configuration and operation, and can easily collect fan beam data and convert data in a short time to reconstruct an image using a parallel data system. Further, various pre-processes can be easily added to the fan beam data, and the image quality of a captured image can be improved.

[Brief description of the drawings]

FIG. 1 is a block diagram schematically showing a configuration of a SPECT apparatus according to one embodiment of the present invention, FIGS. 2 and 3 are flowcharts showing operations in the embodiment, and FIGS. 4 (a) to 4 (c). FIG. 5 is a diagram showing fan beam data in the embodiment and parallel data created from them; FIG. 5 is a diagram showing the fan beam data in the embodiment in terms of fan beam angles and camera pixel positions; FIG. FIGS. 7 to 9 show the relationship between fan beam data and parallel data.
FIG. 10 is an explanatory diagram for explaining fan-para conversion processing in the embodiment, FIG. 10 is a diagram illustrating a fan beam collimator, FIG. 11 is a diagram illustrating a parallel beam collimator, and FIG.
FIG. 13 is a diagram showing the state of the γ-ray incident surface of the fan beam collimator at the time of imaging the subject, and FIGS. 14 and 15 are diagrams showing the algorithm of image reconstruction in the conventional fan beam data collection SPECT device FIG. 16 is a diagram for explaining the conventional fan-para conversion. DESCRIPTION OF SYMBOLS 1 ... SPECT apparatus 2 ... Fan beam collimator 4 ... Fan beam data collection / preprocessing means 5 ... Image memory 6 ... Fan / para conversion means (data conversion means) 7 ... Image reconstruction means P ... Specimen

──────────────────────────────────────────────────続 き Continuation of front page (56) References JP-A-58-177636 (JP, A) JP-A-63-85480 (JP, A) JP-A-4-86585 (JP, A) (58) Field (Int.Cl. ⁷ , DB name) G01T 1/161

Claims (1)

(57) [Claims] 1. A two-dimensional detection means for two-dimensionally detecting fan beam data corresponding to γ-rays emitted from a radioisotope administered to a subject via a fan beam collimator in an X direction and a Y direction. X-direction parallel data detection means for detecting parallel data in the X direction based on fan beam data in the X direction detected by the two-dimensional detection means; and parallel data detected by the X-direction parallel data detection means. Y-direction parallel data detection means for detecting fan beam data in the Y direction detected by the two-dimensional detection means corresponding to the data position as parallel data in the Y direction; Image reconstruction means for reconstructing an image based on the parallel data respectively detected by the parallel data detection means; SPECT apparatus characterized in that it comprises.