JPH08184677A - Camera rotation type ect device - Google Patents

Abstract

PURPOSE: To correct the influence of absorption with sufficient accuracy, although the device is simple, in a practical clinical examination. CONSTITUTION: The output of pulse height analyzers 21, 22 having an energy window in the neighborhood of a photoelectric peak and a lower energy window than it respectively is counted by image memories 31, 32 every position signal X, Y, and image data C1i(x, y), C2i(x, y) are collected. In an image processing device 40 a ration of the whole counting value of these all picture elements is found, and the image data Caci(x, y) after correction are obtained by using a correction coefficient in response thereto. They are transmitted to an ECT image reconstruction processor 50.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a nuclear medicine diagnostic apparatus, and more particularly to a SPECT (Single Photon Emission C) type of rotating a two-dimensional radiation detector (radiation camera).
omputed Tomograph) device (camera rotation type ECT device)
Regarding

[0002]

2. Description of the Related Art A nuclear medicine diagnostic device such as a scintillation camera administers a radiopharmaceutical to a subject (patient), accumulates it at an inspection site (specific organ, etc.), and measures the accumulation state to make a medical diagnosis. It is useful for. That is, a radiation detector is placed outside the body of the subject, and radiation emitted from the body to the outside is detected to collect emission data. A radiation camera such as a scintillation camera obtains a concentration distribution image of a radiopharmaceutical viewed from a certain direction. E
The CT apparatus collects projection data in all directions within a plane (tomographic plane) of the subject and back-projects the data to obtain a concentration distribution image (tomographic image) of the radiopharmaceutical on that plane.

By the way, the radiation emitted from the radioactive substance in the subject is absorbed by the internal tissues before it goes out of the body. Since the nuclear medicine diagnostic apparatus detects the radiation emitted from the radioactive substance inside the body to the outside of the body as described above, it must be affected by the absorption as described above.

In the camera rotation type ECT device, a radiation camera (two-dimensional radiation detector) such as a scintillation camera is used.
By rotating the object around the subject, collecting image data at each angle, and extracting only the image data contained in the tomographic plane as projection data at each angle, and backprojecting this. To reconstruct the image. In almost all cases, the cross section of the subject (patient body) is not circular, and the radiopharmaceutical is not concentrated in the center of the cross section but is unevenly distributed. Therefore, the projection data in one direction is due to the radiation transmitted through the body tissue for a longer distance, while the projection data in the other direction is due to the radiation transmitted through the body tissue for a shorter distance. Then, the effect of absorption will vary depending on the direction.
Therefore, if the projection data in each direction are treated the same and the back projection processing is performed to reconstruct the image, an accurate image cannot be obtained.

The following methods have been conventionally known as correction methods for correcting this absorption. First, as an absorption correction method for a uniform absorber, Sorenso
Various methods such as the n method and the Chang method are known. For heterogeneous absorbers, see Transmission C
A method is known in which T data is collected to obtain absorption coefficient distribution data, and the emission data is corrected based on this. That is, using an X-ray CT apparatus or the like, transmission data (Tr
If the absorption coefficient distribution data for the same tomographic plane is obtained by collecting the projection data (anmission data) and backprojecting it, the ECT image (radioactive agent concentration distribution image) for that tomographic plane can be corrected. It is possible. Further, as in Japanese Patent Publication No. 6-40128, it is known to perform scattered ray correction and absorption correction based on energy spectrum information of each pixel of a scintillation camera.

[0006]

However, the conventional absorption correction method has the following problems. First, the correction methods such as the Sorenson method and the Chang method relating to the uniform absorber have a large error and are not suitable for the actual object which is the non-uniform absorber. In addition, although the method of collecting the transmission data and obtaining the absorption coefficient distribution image can certainly perform a fairly strict correction, the data is collected by irradiating the subject with radiation from the future using an external radiation source. It is necessary to reconstruct a CT image, and there is a problem that an X-ray CT apparatus must be used, and an extra acquisition time or reconstruction processing time is required. Furthermore, when absorption correction is performed from the energy spectrum information for each pixel of the scintillation camera, a correction coefficient is calculated for each pixel, and in applications with a small count value such as normal clinical examination, the effect of statistical error may occur. There is a problem that it is large and the amount of processing is enormous and it is not practical.

In view of the above, the present invention has been improved so as to be able to correct the influence of absorption with sufficient accuracy in an actual clinical examination while being simple enough to be practically adopted. An object is to provide an ECT device.

[0008]

In order to achieve the above object, the camera rotation type ECT apparatus according to the present invention outputs a two-dimensional position signal and energy signal of the incident position for each radiation incident event. -Dimensional radiation detection means, rotation means for rotating the two-dimensional radiation detection means around the subject, and detection that the crest value of the energy signal is contained in the first window provided near the photoelectric peak First crest analysis means for producing a signal, and the above first
Of the first and second wave height analyzing means for detecting a wave height value of the energy signal in a second window provided on the lower energy side of the window to generate a signal. First and second image data collecting means for collecting the first and second image data by counting the output for each of the above position signals, and a total count value of a predetermined area in the first and second image data. Correction operation means for applying a correction coefficient obtained from the ratio of at least the first image data, and projection data in each angular direction using the corrected image data, and an image is created from the projection data in each angular direction. It is characterized in that it is provided with a means for performing reconstruction calculation.

[0009]

[Function] The energy spectrum obtained by detecting radiation reflects the thickness and absorption coefficient of the absorber. Therefore, conversely, the thickness and absorption coefficient of the absorber should be estimated from the energy spectrum information. It is possible to correct the absorption of a non-uniform absorber. In this case, if the energy spectrum information is used with high energy resolution (comparing the count values for each of the subdivided energy ranges), more accurate absorption correction can be expected. Since count values must be collected for each energy range, the configuration is complicated and processing takes time, and if there are not many count values for each subdivided energy range, it will be rather inaccurate due to the influence of statistical error. . As described above, the first wave height analyzing means detects the wave height value of the energy signal in the first window provided near the photoelectric peak to generate a signal, and the second wave height analyzing means. Is generated as a signal by detecting that the crest value of the energy signal is contained in the second window provided on the lower energy side of the first window, and outputs from these first and second crest analysis means. If you count
Since it is a count for two energy windows, energy spectrum information can be used with a simpler configuration, and the count value for each energy window does not become small even in the case of ordinary clinical examinations, and statistics The influence of the error can be avoided. Further, if the count values for such two energy windows obtained for each pixel obtained by the two-dimensional radiation detecting means are obtained, absorption correction can be performed for the data for each pixel, and more accurate absorption correction can be performed. As you can expect, absorption correction processing must be performed for each pixel,
The configuration is complicated, it takes time to process, and if the count value for each pixel is not large, it becomes rather inaccurate due to the influence of statistical error. Here, as described above, the energy spectrum information is used in the form of a ratio between the total count values of the predetermined regions (entire region or region of interest) in the first and second image data. Therefore, even in the case of a normal clinical examination, the total count value of each is not too small and is not affected by the statistical error, and the configuration is simple and the processing does not take time. Therefore, even if the radiation source is localized in the subject and the thickness and absorption coefficient of the absorber are different in each angle direction, a Transmi
It is not necessary to collect ssion data, and it is possible to correct the influence of absorption with accuracy close to the limit that is possible in actual clinical tests while being simple enough to be practically adopted.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a preferred embodiment of the present invention will be described in detail with reference to the drawings. In FIG. 1, the gamma camera 10 is configured as a scintillation camera, and is rotated around the subject 60 by a rotation mechanism (not shown) (see arrow). This gamma camera 10 includes a collimator 11 and a scintillator 12
, Light guide 13, photomultiplier (PM
(Abbreviated as T) 14 and position and energy calculation circuit 15
And a housing 16 made of a radiation-shielding material for housing them.

A radiation-shielding material is removed from the front surface of the housing 16, and a large number of radiation transmitting holes, for example, parallel to a flat radiation shielding plate, are provided on the front surface to regulate the direction of radiation. The configured collimator 11 is attached. The scintillator 12 is formed in a flat plate shape and emits light when radiation enters. The light is guided by the light guide 13 to the light-receiving surfaces of many PMTs 14. When the position of the PMT 14 is closer to the light emitting position, the amount of incident light is larger and a large output can be obtained. Therefore, the output of each PMT 14 can be converted into the surface of the flat scintillator 12 (X
Position signals X and Y are obtained by performing weighted addition calculation in the X and Y directions on the (-Y plane). Further, the sum of the outputs of all PMTs 14 corresponds to the total amount of light emission and corresponds to the energy of the incident radiation, so this sum is output as the energy signal E. It is determined that the energy signal E is within a predetermined energy window, and an unblank signal UNB indicating that there is an incident event of radiation from a predetermined nuclide is obtained. These calculations are performed by the position and energy calculation circuit 15, and the position signals X and Y, the energy signal E and the unblank signal UN are calculated.
B is output.

The energy signal E is an unblank signal U
Wave height analyzers 21 and 2 whose operation timing is controlled by the NB
It is determined whether or not the wave height is input to the windows W1 and W2 respectively, and when it is in the window W1, the unblank signal UNB1 from the wave height analyzer 21 is in the window W2. An unblank signal UNB2 is output from the wave height analyzer 22.

Here, in order to simplify the explanation, it is assumed that a single-energy γ-ray emitting nuclide (for example, a nuclide such as Tc-99m in which the emission γ-ray has a photoelectric peak near 140 keV) is measured. . Normally, a window of about 20% is provided around the photoelectric peak and it is detected whether or not the wave height of the energy signal E is included in it, but this is divided into two as shown in FIG. Two windows W1 and W2 (for example, W1 is the lower 10% of the photoelectric peak and W2 is the upper 10%).

The γ rays from the subject 60 are gamma cameras 10.
When incident on, the position signals X and Y are output for each incident event and sent to the image memories 31 and 32. At this time, if the wave heights of the energy signal E output for each incident event are included in the windows W1 and W2, respectively, the wave height analyzers 21 and 22 perform the unblank signals UNB1 and UNB.
2 is output, and these are sent to the image memories 31 and 32, respectively. In the image memories 31 and 32, each time the unblank signals UNB1 and UNB2 are input,
"1" is added to the count value of the address designated by the position signals X and Y. As a result, the count value of incident γ rays is accumulated for each two-dimensional pixel, and two-dimensional image data is collected.

Thus, the subject 60 of the gamma camera 10
When the angle with respect to is the i-th angle, the image data C1i (x, y), C2i is stored in the image memories 31 and 32.
(X, y) is collected. C1i (x, y) and C2i (x, y) are the window W1 and the window W2.
Represents the count value of the pixel (x, y) collected for each. Such image data collection is performed for each angle of the gamma camera 10.

Collected C1i (x, y), C2i
(X, y) is read from the image memories 31 and 32 and sent to the image arithmetic processing unit 40, and the image arithmetic processing shown by the following mathematical formulas 1, 2 and 3 is performed.

[Equation 1]

[Equation 2]

(Equation 3) In these mathematical expressions, Cnci (x, y) is the count value of the pixel (x, y) in the image of absorption correction OFF at the i-th angle, and Caci (x, y) is the absorption correction ON at the i-th angle. Represents the count value of the pixel (x, y) for each image. Also,
fi (Pi) represents an absorption correction coefficient at the i-th angle. Since Pi is the ratio of the sum of the count values in the entire field of view regarding the windows W1 and W2 at the i-th angle as shown in Expression 3, the absorption correction coefficient fi is determined as a function of this ratio Pi.

The Cnci (x, y) or Caci (x, y) thus obtained is the ECT image reconstruction processor 50.
Sent to These Cnci (x, y) or Caci
Of (x, y), the pixel (x,
By extracting only the data relating to y) to form projection data and performing backprojection calculation processing on this, an ECT image with absorption correction OFF or absorption correction ON is reconstructed.

Next, the image data Caci (x, y)
Let us explain in more detail that is the data that has been properly absorption-corrected. First, as shown in FIGS. 2A and 2B, the target organ 61 of the subject 60 in which the radiopharmaceutical is accumulated.
However, it is assumed that the desired tomographic plane is not located at the center but localized. At this time, when the same gamma camera 10 is rotated around the subject 60 and image data is collected, the subject 60, the nuclide administered therein, the organ 61 in which the nuclide is accumulated, the collimator 11, etc. are all the same. The condition 60 is satisfied, and only the subject 60 of the gamma camera 10
Only the angle with respect to is different as shown in FIGS. 2 (a) and 2 (b).

In FIG. 2A, the distance La until the γ-rays emitted from the nuclide in the target organ 61 and reaching the gamma camera 10 reach the body surface is short, and in FIG. 2B, the distance Lb. Is long. That is, the thickness of the absorber interposed between the nuclide and the gamma camera 10 is small in FIG. 2 (a), but large in FIG. 2 (b). Therefore, the energy spectrum of the sum of the count values of all fields of view (all pixels) is shown in FIG.
In (a) of FIG. 3, it becomes like (a) of FIG. 3, and in (b) of FIG.
Then, it becomes as shown in FIG.

This is because the distance of γ-rays reaching the gamma camera 10 to the body surface is longer in FIG. 2B than in FIG. 2A (La). <L
2B and FIG. 2B, since the absorber is thick, the attenuation due to absorption is

[Equation 4] As described above, the count value of the window W2 is significantly reduced and the γ scattered by the Compton at a low angle in the subject 60.
This is because the probability of incidence of a line increases and the count value of the window W1 increases.

Therefore, the ratio P of the total count value of the window W1 to the total count value of the window W2 (Equation 3)
2) is larger in FIG. 2B than in FIG. 2A. This ratio P corresponds to the thickness of the absorber interposed between the nuclide and the gamma camera 10. The thicker this is, the larger P becomes. Therefore, in order to correct the influence of this thick absorber. It is understood that the larger the P, the larger the absorption correction coefficient f.

The function of the absorption correction coefficient f relating to the ratio P can be obtained by measuring using a phantom 70 as shown in FIG. Here, in the phantom 70, an object (liquid) having an absorption coefficient similar to that of a human body is put in an appropriate container 71 as an absorber 72, and a radiation source 73 of the same nuclide as the nuclide to be measured is placed therein. Made by sinking. The absorber thickness L is adjusted by adjusting the amount of the absorber 72.

The absorber thickness L is variously changed, and for each L, with the same gamma camera 10, the window W1 with respect to the total count value of the window W2 for the entire field of view.
The ratio P of the total count value of is measured. Then, the data as shown in FIG. 5 is obtained. In the field of view, the radiation source 7
For a partial area of the portion corresponding to 3, the sum Cncr of the count values of the window W1 and the window W2 is measured. L
When Cncr when = 0 is defined as Cncro, Cncr / Cncro for each L is obtained as shown in FIG.

This Cncr / Cncro indicates the damping ratio due to absorption. In other words, it is understood that the reciprocal of this damping ratio can be used as the absorption correction coefficient. Therefore, from FIGS. 5 and 6, by using the absorber thickness L as a parameter, the function of the absorption correction coefficient f with respect to the ratio P can be obtained as shown in FIG. 7. If the function of the absorption correction coefficient f for this ratio P is prepared as a conversion table of the ratio P and the coefficient f, the actual object 60
For the i-th angle, the absorption correction coefficient fi can be immediately obtained by referring to the conversion table, and the image data Cnci (x,
y) (see Equation 1 above) is absorption-corrected, and the image data Caci (x, x,
This is desirable because y) can be obtained (see the above equation 2).

Let us now consider the case where the absorption coefficient changes. Even if the thickness of the absorber is the same, if the absorption coefficient is different, the energy spectrum having the same tendency as when the thickness of the absorber is changed can be obtained. That is, if the absorbers have the same thickness (La = Lb) and the absorption coefficients μa and μb have a relationship of μa <μb,

(Equation 5) As shown by, the attenuation due to absorption is larger in the case of the absorption coefficient μb than in the case of μa. Therefore, the energy spectrum is as shown in FIG. 3A for the absorption coefficient μa and as shown in FIG. 3B for the absorption coefficient μb.
In addition, the window W relative to the total count value of the window W2
The ratio P of the total count value of 1 is larger in the case of the absorption coefficient μb than in the case of μa. Therefore, it is understood that such a difference in the absorption coefficient is equivalent to a change in the absorber thickness, and the absorption correction can be performed by using the above formulas 1, 2, and 3.

In the above description, the ratio Pi is calculated by using the total count value of each window for the entire field of view (all pixels) as shown in the mathematical expression 3, but it is possible to calculate the ratio Pi in a part. It is also possible to obtain using the total count value of each window for the region (which can be distinguished from others because the count value is relatively large in this region). Here, a part of the area may be an area within the fault plane,
Alternatively, it may be an area in a direction perpendicular to the tomographic plane.

Further, in the formula 3 for obtaining the ratio Pi, as the denominator thereof, instead of the total count value of the window W2, the total count of the image data Cnci of the absorption correction OFF collected near the photoelectric peak as shown in the formula 1 It is also possible to use the value and use the total count value of another window set on the lower energy side instead of the total count value of the window W1 as the numerator.

Further, the number of windows may be three or more instead of two as described above, and the absorption correction OFF image data Cnci and the ratio Pi may be obtained.

By providing another window on the higher energy side than the vicinity of the photoelectric peak (windows W1 and W2), the total count value of the entire screen in that window is subtracted from the total count value of windows W1 and W2, respectively. , The background noise included in the total count value of these windows W1 and W2 is removed, and then the ratio P
You may make it obtain i.

Image data Cnci with absorption correction OFF
(X, y) is not obtained as simple addition image data of the two windows W1 and W2 near the photoelectric peak as shown in the above Equation 1, but the image data after scattering correction by weighted addition is performed. It can also be used. If the image data after such scatter correction is used, it is particularly effective at the time of multinuclide measurement.

In addition, it goes without saying that the specific configuration can be variously modified without departing from the spirit of the present invention. For example, as a gamma camera as a two-dimensional radiation detector, one using a semiconductor radiation detector in addition to a scintillation camera can be used.

[0032]

As described in the above embodiments, according to the camera rotation type ECT apparatus of the present invention, when the radiation source is localized in the subject, the absorption correction is performed with sufficient accuracy for practical use. be able to. This is simple because it is not necessary to collect the Transmission data using an X-ray CT apparatus or the like. Therefore, the data collection time may be the same as when absorption is not corrected, the burden on the patient can be reduced, and exposure to extra radiation can be avoided. Furthermore, absorption correction can be performed with sufficient accuracy even in the case of data collection with a small count value such as a normal clinical test.

[Brief description of drawings]

FIG. 1 is a block diagram of an embodiment of the present invention.

FIG. 2 is a schematic diagram showing a state of γ rays incident on a gamma camera at each angle.

FIG. 3 is a graph showing an energy spectrum at each angle.

FIG. 4 is a schematic diagram showing measurement using a phantom.

FIG. 5 is a graph showing the relationship of the ratio to the absorber thickness obtained by phantom measurement.

FIG. 6 is a graph showing the relationship between the damping ratio and the absorber thickness obtained by phantom measurement.

FIG. 7 is a graph showing the relationship between the absorption correction coefficient and the ratio obtained by phantom measurement.

[Explanation of symbols]

 Reference Signs List 10 gamma camera 11 collimator 12 scintillator 13 light guide 14 photomultiplier 15 position and energy calculation circuit 16 housing 21, 22 wave height analyzer 31, 32 image memory 40 image calculation processing device 50 ECT image reconstruction processing device 60 subject 61 target Organ 70 Phantom 71 Container 72 Absorber 73 Radiation source

Claims (1)

[Claims] 1. The incident position of each radiation incident event is 2
Two-dimensional radiation detecting means for outputting a two-dimensional position signal and energy signal, rotating means for rotating the two-dimensional radiation detecting means around the subject, and the energy signal for the first window provided near the photoelectric peak. First crest analysis means for generating a signal by detecting that the crest value of
The second provided on the lower energy side than the above first window
Second wave height analyzing means for generating a signal by detecting that the wave height value of the energy signal is contained in the window, and outputs of the first and second wave height analyzing means are counted for each of the position signals. First and second image data collecting means for collecting first and second image data, and
A correction calculation unit that applies a correction coefficient obtained from the ratio of the total count value of the predetermined area in the second image data to at least the first image data, and projection data in each angular direction by using the corrected image data. A camera rotation type ECT device, which is provided with means for performing image reconstruction calculation from the projection data in each angular direction.