JPS63193086A - Pet apparatus - Google Patents

Abstract

PURPOSE:To cover inferior resolution in the direction of the thickness of a slice by a method wherein the data collected in regard to a cross slice are multiplied by a weight coefficient being smaller as a distance from the central axis of the cross slice. CONSTITUTION:Radiation detectors 1 disposed in the shape of a ring are superposed in a plurality of layers. A coincidence counting circuit 3 conducts coincidence counting among the detectors 1 in the same layer (e.g. a second layer), and the data on the direct slice thereof are collected in a data memory 5. A coincidence counting circuit 4 conducts coincidence counting among the detectors 1 in each of layers (e.g. first and third layers), and the data on a cross slice are collected in a data memory 6. Then, a uniform coefficient is given to the data on the direct slice from a weight coefficient memory 9, while a weight coefficient being smaller as a distance from the central axis of the cross slice becomes larger is given to the data on the cross slice from said memory 9. After the data are multiplied by the coefficients by multipliers 7 and 8 respectively, they are added up by an adder 10 and sent to an image recomposition computing device 11 to be subjected to an arithmetic processing.

Description

[Detailed description of the invention]

[Industrial Application Field 1] The present invention relates to a PET device, that is, an ECT device (emission type computer tomography device) using a positron-emitting nuclide R1 (radioisotope). [Prior Art] In a PET apparatus, a subject to whom a positron-emitting nuclide RI is administered is placed in a ring-shaped arrangement of radiation detectors, and radiation from the RI in the subject's body is detected. . When the positron disappears, two gamma rays are emitted in a 180° direction, and these rays simultaneously enter the ring-shaped detector. Therefore, by performing this coincidence counting, it can be determined that RI exists on the line connecting the two incident positions. By collecting RI position information in this way and processing it using an image reconstruction algorithm using a computer, it is possible to reconstruct an RI distribution image (tomographic image) within one slice. As shown in Fig. 4, the detectors 1 arranged in a ring shape are overlapped in the rotational layer in the direction of the central axis, and each layer is separated by a septa (collimator) 2 to simultaneously image a large number of slices. In addition to simultaneous counting between detectors 1 in the same layer (for example, the first layer), simultaneous counting is performed between the detectors 1 across the pond layers (for example, the detectors in the first and second layers). (Question 1), it is common practice to count simultaneously. In the former case, data on slices perpendicular to the central axis (this is called a direct slice) as shown by the dotted line can be obtained; in the latter case, data on
A slice tilted to the central axis as shown by the dotted line (
This is because it is possible to obtain data regarding the cross slice (this is called a cross slice), to simultaneously image a direct slice and a cross slice located between them, and to increase the number of slices that can be simultaneously imaged. Furthermore, in order to increase the number of slices that can be captured simultaneously,
As shown in the figure, the slice thickness of detector 1 and the width in the central axis direction)
D is also made thinner and arranged in multiple layers. In this case, since the sensitivity per slice is proportional to 1/D2, a problem arises in that the sensitivity decreases significantly as the width of the detector 1 becomes narrower. Therefore, as shown in Fig. 6, if you want to obtain an image of the slice indicated by arrow A, the collected data of the direct slice (dotted line) for the second layer is combined with the cross slice (double-dashed line) for the first and third layers. It is being considered to add collected data to increase sensitivity and reconstruct images. In addition, when obtaining an image of the slice shown by the arrow in FIG. 6, the sensitivity is increased by adding the cross-slice collected data regarding the 5th and 6th layers and the cross-slice collected data regarding the 4th and 7th layers. To increase.

[Problems to be solved by the invention]

However, when using data collected regarding cross-slices in this way, there is a problem in that the response (resolution) in the slice thickness direction (center axis direction) deteriorates particularly in the peripheral part of the visual field. That is, to explain the slice shown by arrow A in FIG. 6 as an example, the cross slice regarding the first layer and the third layer coincides in the slice thickness direction near the central axis, but when the central axis is As you move away, you move away. Therefore, when cross-slice data regarding the first and third layers are added to direct slice data regarding the second layer, the responses in the slice thickness direction are as shown in A to C, and the central axis 1
It is good at 4 near, but it gets worse as you move away from the central axis and towards the periphery of the visual field. An object of the present invention is to provide a PET apparatus that is improved so as to compensate for poor resolution in the slice thickness direction of data collected regarding cross slices.

[Means to solve the problem]

The PUT device according to the present invention includes radiation detection means arranged in a ring shape and arranged in a plurality of layers, and means for performing coincidence counting between the detection means in each layer. , a means for collecting data regarding the direct slice from now on, a means for performing coincidence counting between the detection means spanning each of the above layers, a means for collecting data regarding the cross slice from now on, and a means for collecting data regarding the cross slice from now on. It consists of means for applying a weighting coefficient that decreases as the distance from the central axis increases, and means for processing the collected data to reconstruct an image.

[For production]

The resolution in the slice thickness direction of data collected regarding cross-slices deteriorates toward the periphery of the field of view, but this data is multiplied by a weighting coefficient that decreases as the distance from the central axis increases. the result,
Data with good resolution in the slice thickness direction near the central axis is effectively used for image reconstruction, and conversely, data in the peripheral part of the visual field with poor resolution in the slice thickness direction has a low contribution to the reconstructed image. Therefore, an image reconstructed using the data processed in this way becomes an image with less deterioration in resolution in the slice thickness direction in the peripheral part of the visual field.

【Example】

In FIG. 1, detectors 1 arranged in a ring shape are stacked in seven layers in the central axis direction. A subject (not shown) is inserted into this ring. RI has been administered to this subject, and radiation emitted from the RI inside the body is collimated by the septa 2 and enters the detector 1 in each layer. In this example, it is assumed that an image is obtained for each slice indicated by arrow A. Coincidence between the detectors 1 in the second layer is detected by the coincidence circuit 3, and data of the direct slice of the second layer is collected in the data memory 5. Further, the coincidence circuit 4 detects simultaneous incidence of radiation between the first and third layer detectors 1, and the data memory 6 collects cross-slice data regarding the first and third layers. In this way, direct slice and cross slice data is stored in the memory in the form of a sinobram (projection data arranged by projection angle) as shown in FIGS. 2A and 3A, for example.
6 will be stored. These data are stored in memory 5
．． 6 and are respectively multiplied by weighting coefficients in multipliers 7.8. The weighting coefficients are stored in advance in the weighting coefficient memory 9' depending on whether the slice is a direct slice or a cross slice, and what type of loss slice it is. In the case of die reflex I slices, the resolution in the slice thickness direction does not decrease as you move toward the periphery, so the weighting coefficient is uniformly set to "1" in the diametrical direction as shown in Figure 2B. . On the other hand, in the case of cross slicing, the resolution in the slice thickness direction is good at the center, but worsens toward the periphery. '', the function gradually becomes rQJ as it moves away from the center. The sharpness of this weighting coefficient curve increases as the slope with respect to the central axis increases, for example, in a cross slice between the first layer and the fourth layer. The data multiplied by the weighting coefficient in this way is sent to the adder 1.
0 is added and sent to the image reconstruction arithmetic unit 11, where arithmetic processing for image reconstruction is performed. Therefore, in reconstructing an image, data with good response in the slice thickness direction is effectively used, and conversely, data with poor response is used less frequently. Therefore, the resulting image has increased sensitivity near the center of the field of view by adding cross-slice data to direct slice data, thereby minimizing the reduction in resolution in the slice thickness direction at the periphery of the field of view. It becomes what it is. In this embodiment, the present invention is applied when adding cross-slice data of two layers adjacent to a certain layer to direct slice data of a certain layer, and in addition, cross-slice data of two layers adjacent to that layer are added. The present invention can be applied in the same way when adding up a plurality of cross-slice data, and can also be applied when a plurality of cross-slice data are added together, as in the case of photographing slices indicated by arrows in FIG. Furthermore, the curve of the weighting coefficient may be changed for each case of imaging. Furthermore, such operations on the collected data can be done by hardware or software.

【Effect of the invention】

According to the PET apparatus of the present invention, the resolution in the slice thickness direction of data collected regarding cross-slices is improved to compensate for the decrease in the periphery of the visual field.
Even when sensitivity is increased by adding loss slice data to direct slice data, poor response in the slice thickness direction can be minimized.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of an embodiment of the present invention, FIG. 2A,
B and Figures 3 A and B are for explaining the relationship between the collected data and the weighting coefficients that act on it.
Figure 3A represents the collected data, Figure 2B, Figure 3
FIG. B is a graph showing weighting coefficients, FIGS. 4 and 5 are schematic diagrams of the conventional example, and FIG. 6 is a schematic diagram for pointing out problems. DESCRIPTION OF SYMBOLS 1... Detector, 2... Sebuta, 3.4... Coincidence circuit, 5.6... Data memory, 7.8... Multiplier, 9... Weighting coefficient memory, 10 ...Adder, 11
...Image reconstruction calculation device.

Claims (1)

[Claims] (1) Radiation detection means arranged in a ring shape and arranged in such a way that the ring-shaped arrangement is stacked in multiple layers, a means for performing coincidence counting between the detection means in each layer, and direct slicing from this. a means for collecting data on the cross slice, a means for performing coincidence counting between the detection means spanning each of the layers, a means for collecting data on the cross slice from now on, and a means for collecting data on the cross slice from the central axis thereof. A PET apparatus comprising means for applying a weighting coefficient that decreases as the distance increases, and means for processing collected data and reconstructing an image.