JPH068856B2 - Method for correcting coincidence coincidence count in PCT apparatus - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (a) Field of Industrial Application This invention relates to a method for correcting random coincidence in a PCT apparatus, that is, an ECT apparatus (emission computer tomography apparatus) using a positron-emitting nuclide as a radionuclide. Regarding

(B) Prior art As is well known, the PCT device utilizes the fact that two γ-rays are emitted in the 180 ° direction when the positron disappears, and two detectors are used to direct two γ rays in the 180 ° direction. γ
It is based on knowing that the positron-emitting radionuclide exists on the line connecting the two detectors by detecting the line. In an actual device, a large number of radiation detectors are arranged in a ring in one plane, and the radiation emitted in the 180 ° direction when the positron disappears inside the ring arrangement is one of the detectors. The simultaneous incidence on two is used to determine on which straight line the positron has disappeared, and data is obtained on the coincidence count for each of the straight lines in the plane. Then, those pieces of the count data having the same straight line direction are collected as one profile data, and the profile data in each direction is processed by the image reconstruction processing algorithm (for example, after the profile data in each direction is subjected to a correction filter). , Etc.) to reconstruct the concentration distribution image of the positron-emitting nuclide in the plane.

When obtaining this coincidence count, there is a problem that separate gamma rays, which are not due to one positron-emitting nuclide, are incident on two detectors at the same time and are coincident with each other. This is called random coincidence counting, and is an error in profile data.

Therefore, conventionally, various methods for correcting the coincidence coincidence count have been proposed. As one of them, a method of counting Delay Coincidence is known as a highly accurate correction method. This is to count the number of radiation incident events (On Time Coincidence) that actually occurred at the same time, including the coincidence coincidence, while the other is intentionally appropriate the signal from one detector. Good time (eg 5
0 nsec), counting only the simultaneity (Off Time Coincidence) between this delayed signal and the undelayed signal from the other detector,
Then, since the off-time coincidence is generated only by accident, the subtraction of the off-time coincidence count value from the on-time coincidence count value excludes the coincidence coincidence count included in the on-time coincidence count value. It can be corrected.

Specifically, the on-time coincidence address is incremented by 1 and the off-time coincidence is decremented by -1, or the on-time coincidence count value and the off-time coincidence count value are collected in separate memories. However, in any case, the statistical fluctuation included in the off-time coincidence count value is subtracted,
It will be added to the final data. As a result, the final image had statistical fluctuations as if there were twice as many coincidence coincidences, and the S / N of the image was deteriorated.

In particular, in the case of transmission data used for absorption correction, it is extremely problematic that the S / N thereof deteriorates due to the statistical fluctuation of the off-time coincidence count value.

(C) Purpose The present invention is a method of counting off-time coincidence by delay coincidence and subtracting this count from the count of on-time coincidence to correct the error due to coincidence coincidence, and the count of off-time coincidence. It is an object of the present invention to provide a correction method that can suppress the deterioration of S / N to the minimum by removing the influence of the statistical fluctuation included in.

(D) Configuration In the method for correcting the coincidence coincidence count in the PCT apparatus according to the present invention, the on-time coincidence data and the off-time coincidence data are collected in different memories, and the off-time coincidence data is smoothed (smoothed). It is characterized in that it is subtracted from the data of on-time coincidence after being processed.

(E) Example In the present invention, the coincidence coincidence is generally distributed over the entire field of view, and even in the case of a point source, the spatial coincidence is moderately distributed. Therefore, smoothing (smoothing processing) is performed on the coincidence coincidence data. It is noted that the distribution does not change even if it is applied, and only the effect of correcting and removing the statistical fluctuation due to smoothing is obtained.

In the figure, a large number of radiation detectors 1, 1, ... Are arranged in a ring in one plane, and the radiation emitted in the 180 ° direction when the positron disappears inside the ring arrangement is as described above. Two detectors 1 can be simultaneously incident. The subject 2 is placed in this ring-shaped array, and a drug containing a positron-emitting nuclide has been administered to the body of the subject 2 in advance. Two detectors 1, 1 which emit a line and which lie, for example, on one straight line 21.
Incident at the same time.

Here, the simultaneous detection circuit 3 and the delayed simultaneous detection circuit 4 are provided for each set in which two detectors 1, 2, ... The simultaneous detection circuit 3 is for counting the on-time coincidence and is composed of an AND circuit 31 and a flip-flop 32, and produces an output when the outputs of the two detectors 1 are simultaneously input. The delay coincidence detection circuit 4 is for coincident counting of off-time coincidences, and the AND circuit 41 and the flip-flop 4 are provided.
2 and the output of one detector 1 for a certain period of time, for example 50 nse
and a delay circuit 43 that delays by c. When the delayed output of one detector 1 and the undelayed output of the other detector 1 are simultaneously input to the AND circuit 41, the flip-flop 42 Will produce output.
The outputs of the simultaneous detection circuit 3 and the delayed simultaneous detection circuit 4 are sent to the memories 5 and 6, and the contents of the address corresponding to the set of the detector 1 are incremented by one.

When the γ-rays are simultaneously incident on the two detectors 1 and 1 on the straight line 21 as described above, an output is generated from the simultaneous detection circuit 3 to which the outputs of these detectors 1 and 1 are sent,
The contents of the address corresponding to the straight line 21 are incremented by one. The count value of the on-time coincidence thus obtained is not only the count value of two γ rays generated from one positron-emitting nuclide as described above, but also the incident value of γ rays incident at the same time. The value of the coincidence count is also included.

On the other hand, 5 for each of the two detectors 1, 1 on the straight line 21
When γ-rays are incident with a time difference of 0 nsec, an output is generated from the delayed simultaneous detection circuit 4, this output is sent to the memory 6,
The contents of the address corresponding to the straight line 21 are incremented by 1, and the off-time coincidence is counted. The fact that an output is produced from the delayed simultaneous detection circuit 4 in this way means that the γ-rays are incident on the two detectors 1 and 1 with a time difference of 50 nsec. That is, this is a case that never occurs in the case of two γ rays from one positron-emitting nuclide, and it can be said that this occurs only by accidental ones. Therefore, the coincidence counting data is collected in the memory 6. The data of the count value of the off-time coincidence is smoothed by the smoothing circuit 7 and then sent to the subtraction circuit 8 to be collected in the memory 6 from the simultaneous count data collected in the memory 5. Subtraction is made to the contingent coincidence data. Here, smoothing means smoothing data between neighboring data. Data is collected for all straight lines including the straight line 21 as described above, and this straight line can be represented by an angle and a position, and data having a close angle and position can be said to be neighboring data. Since the data are arranged two-dimensionally by angle and position, smoothing may be performed between the data arranged in the vicinity in such a two-dimensional array. This smoothing may be performed, for example, by simply averaging the neighboring 3 × 3 9-point data in the two-dimensional data array, or by filtering processing or more complicated arithmetic processing.

Here, it can be said that accidental simultaneous incidence occurs with equal probability at any time. That is, in the case of an accidental γ-ray, the probability of being incident on the two detectors 1 and 1 at the same time and the probability of being incident with a time difference of 50 nsec are the same. Therefore, by subtracting the off-time coincidence data collected in the memory 6 from the on-time coincidence data collected in the memory 5, the contingent same clock value included in the count value of the memory 5 can be removed. It is possible. Since the off-time coincidence count value data is smoothed, statistical fluctuations included in the off-time coincidence count value data are excluded. Note that, in general, coincidence coincidence is distributed over the entire field of view, and even in the case of a point source, it is distributed spatially gently, so even if smoothing is applied to the off-time coincidence count data, its distribution hardly changes. Therefore, there is no problem even if smoothing is applied.

Therefore, the statistical fluctuation of the count value of off-time coincidence does not remain in the data after this subtraction, and by reconstructing the image using this data, the statistical fluctuation due to the correction of the coincidence coincidence count is increased. Minimize, S /
An excellent final image of N can be obtained. This effect is almost the same as when the temporal detection width at the time of simultaneous detection is halved.

The present invention can be applied not only to the emission data for reconstructing the image as described above, but also to the correction of the coincidence coincidence count in the transmission data, the blank data, or the normalization data, By suppressing the increase in statistical fluctuations associated with the correction of the coincidence coincidence count of these data, the S /
It is effective in improving N.

(F) Effect According to the present invention, it is possible to reduce the statistical fluctuation in the count data of the off-time coincidence, and it is possible to reduce the statistical fluctuation in the data after correction of the coincidence coincidence counting performed using the count data of the off-time coincidence. S / N of concentration distribution image of reconstructed positron-emitting nuclide
Can be improved. In particular, when correcting the coincidence coincidence count in transmission data, the S / N of the data can be improved, which is extremely effective.

[Brief description of drawings]

 The figure is a block diagram of an embodiment of the present invention. 1 ... Radiation detector 2 ... Subject 3 ... Simultaneous detection circuit 4 ... Delay simultaneous detection circuit 5, 6 ... Memory 7 ... Smoothing circuit 8 ... Subtraction circuit

Claims (1)

[Claims] 1. A plurality of radiation detectors are arranged in a ring in one plane, and radiation emitted in a 180 ° direction when the positron disappears inside the ring arrangement is one of the detectors. It is possible to determine on which straight line the positron disappears by utilizing the simultaneous incidence on two pieces, and to reconstruct an image from the data on the coincidence counting for each of the straight lines in the plane. In a PCT apparatus for reconstructing a concentration distribution image of a positron-emitting nuclide in the plane by the method, it is detected that signals which cannot be delayed from any two of the detectors are simultaneously input. And simultaneously detect, count each combination of the two detectors detected at the same time, and collect in the first memory. On the other hand, from any two detectors, , A combination of two detectors that have been intentionally delayed by a predetermined time and that the other signal that has not been delayed is simultaneously input to detect the delay simultaneously, and the delay simultaneous detection is performed. Data is collected and collected in the second memory in advance, the data collected in the second memory is subjected to smoothing processing, and then subtracted from the data in the first memory to correct the coincidence coincidence count. Method.