JP3315513B2 - Positron emission CT system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a positron emission CT apparatus for acquiring image data using a radiopharmaceutical.

[0002]

2. Description of the Related Art Positron emission CT (PET)
Is to administer a radiopharmaceutical labeled with an isotope and to image its distribution in a living body.In positron emission CT, it is generated by pair annihilation of positrons using a γ-ray detector arranged in a ring. γ-rays are detected by coincidence counting. The detector ring has multiple layers (4-
16 layers), an image (intra-layer image) of the slice is obtained from coincidence between detectors belonging to each ring, and an image between layers is also obtained from coincidence between adjacent rings. In addition, by providing a slice shield (collimator) between each detector ring, unnecessary gamma rays and scattered radiation in the living body are shielded, and unnecessary counting of detectors is reduced, so that counting or accidental counting is performed. The coincidence count is reduced. The resolution is mainly determined by the detection element (BG
O scintillator), the cross-sectional resolution of recent devices has reached 3-5 mm (half width).

FIG. 5 schematically shows the entire positron emission CT. A gamma ray is detected by an image data acquisition device 310 including a detector ring, and is emitted in the opposite direction by pair annihilation by a signal processing circuit 315. Is detected. Data for constructing an image by the coincidence counting of the signal processing circuit 315 is stored in the data memory 340 and integrated for a certain period of time. Computer 350 constructs an image from the data stored in data memory 340,
The image data is stored in the data storage device 370 and the entire system including the image data collection device 310 is controlled.

In the positron emission CT, generally, data accumulated for a certain period of time is stored in the order of measurement by a computer 350 or the like. The data represents the true degree of change of the object to be measured, and always includes attenuation information of the radiopharmaceutical with respect to time. PE
In general, T data is spatial information such as slice plane information (X, Y) and axial direction information (Z) and time information (t).
This is four-dimensional information having the above four combinations. In the time axis direction, this data is a value obtained by integrating the radiation count number (At) at time (t) from a certain time to an arbitrary time, and is called frame data. As described above, since the resolution in the axial direction is restricted by the physical size of the detection element, Z-motion scan is one method for increasing the resolution in the Z-axis direction and enriching information. .

In the Z-axis direction, PET data is collected as two-dimensional information called slices (so-called sliced cross-sectional views), and as described above, each slice has a surface connecting an opposing detector. (Direct plane)
And the plane (cross plane) connecting the adjacent detectors
There are types. FIG. 6 shows the sensitivity distribution in the Z-axis direction of each slice. A sensitivity peak appears on the plane connecting the centers of the opposing detectors or the centers of the adjacent detectors, and a valley of sensitivity is generated at the center of the peaks. As a result, a so-called “unviewed portion” occurs as shown in FIG. Therefore, information between slices can be obtained by moving the detector by a half pitch of the slice width and collecting data as shown in FIG. 7B. This is a Z-motion scan.

FIG. 8 shows a PE having a Z-motion mechanism.
1 shows an example of a T device together with its process flow. Γ-rays are captured by detectors 312 arranged in a ring of a detector system (image data collection device) 310,
The coincidence circuit 320 determines whether the γ-ray detection signal from the detector 312 is coincidence counting. The coincidence count data is converted by the address conversion circuit 330 from the address of the detector pair to the address of sinogram data (also called T-θMap). Data collection memory 340a / 34 for temporarily storing sinogram data
A plurality of 0b are prepared and function like a FIFO buffer. After reading and storing the address value from the address conversion circuit 330, "1" is added.

The Z-motion motor 390 has the detector system 3
10 is moved to axial positions A and B, and its drive is controlled by a motor drive circuit 380.
The main computer 350 sends a command for measurement to the detector system 310 and generates a frame switching signal 250. The signals 340a and 340b storing the sinogram data are switched by this signal, and the Z-
The signal is sent to the motion motor drive circuit 380 and used as a start signal for scanning the positions A and B by the Z-motion motor 390.

[0008] Thus, the main computer 350 stores the memory 340 in accordance with the axial positions A and B of the detector system 310.
a and 340b are switched to store the sinogram data measured at the positions A and B in the memories 340a and 340b, respectively, while the memories 340a and 340 on the side where the measurement has been completed.
The frame data is transferred from 40b to the image reconstruction device 360. That is, while one of the memories 340a and 340b stores sinogram data, the other reads the stored sinogram data. The image reconstructing device 360 performs computer processing of the frame data to reconstruct the image, and the reconstructed data obtained by this process is stored in the data storage device 370. If the time interval between frames is short, the image reconstruction device 360
Is stored in the form of sinogram data without image reconstruction during data acquisition of the detector system,
After data collection, image reconstruction is performed.

The actual configuration of the detector system (image data acquisition device) 310 is of two types. A linear motion device such as a ball screw attached to the detector ring as shown in FIG. Some implement a Z-motion by moving the detector between positions A and B that are open by minutes. Further, as shown in FIG. 9 (B), the subject is moved by positions A and B corresponding to a half pitch of the slice width by a linear motion device such as a ball screw attached to the subject fixing table, thereby achieving Z-motion. There is something that realizes. As described above, there are two types of PET. As an example in which the former is adopted, SHR-2400 manufactured by Hamamatsu Photonics is used.
And is in operation at its PET center. An example of adopting the latter is SHR-2000, which is sold as PET for animals and is in operation at its PET center. With such a mechanism, the amount of information in the body axis direction has been successfully increased.

Another example of PET using Z-motion is provided with a correction circuit for correcting sinogram data, as described in JP-A-63-115079 (Document 1). is there. This prior art aims at correcting attenuation information with an electric circuit for frame data without using a computer CPU. Since the processing is not performed via the CPU, the processing is very fast. Since the correction of the attenuation information has been completed before the data processing, the processing time can be more efficiently used. Further, as described in Japanese Patent Application Laid-Open No. 2-40586 (Document 2), when it is desired to observe a wider range than the image data collection device, the image data collection device is moved in the axial direction to the range, and There is something to observe in a wide range. Furthermore, as described in Japanese Patent Application Laid-Open No. 2-134589 (Document 3), sinogram data is continuously collected when the image data collecting device is moved in the axial direction, and the information amount in the body axis direction is increased. Some try to.

[0011]

An advantage of the Z-motion scanning method as described above is that the apparent number of image slices is increased, thereby facilitating image analysis in the body axis direction. On the other hand, the disadvantage of this method is that information of the same slice is not temporally continuous. That is, the frame data at the position A is measured at a different time from the frame data at the position B, and the information of the slices at the positions A and B respectively has the count number (At) at different times. )
Has been integrated.

FIG. 10 schematically shows image data obtained by performing a Z-motion scan, using a slice number on the vertical axis and a frame number on the horizontal axis. The slice numbers when the detector is at the origin (position A) are 1, 2, 3,..., And the slice numbers when the detector is moved by Z-motion (position B) are 1 ', 2', 3 '.
…. As can be seen from the right figure, when the predetermined position in the axial direction is considered as a continuous frame as an odd frame and the moved position is considered as a continuous frame as an even frame, the frame data is temporally related to a specific slice. The data is not continuous, and the data indicated by oblique lines and dots is missing.

FIG. 11 schematically shows a change in a time axis direction with respect to a specific slice (1.2.3,... Or 1 ′, 2 ′, 3 ′) of the PET data. End of data collection of frame 1 from measurement start time t ₀ (frame 2
Data collection start) until time t ₁ , and frame 3 data collection start (frame 2 data collection end) time t
Data collection at the position A is performed from ₂ to the data collection end time t _{3 of the} frame 3, and data collection at the position B is performed from the data collection start time t ₁ of the frame 2 to the data collection end time t _{2 of the} frame 2. assuming that is done, the position from time t ₁ with respect to data in the a to time t ₂ (approximate them if time t ₁ and t ₂ of the intermediate time t
_{At 12} ) the data will be missing.

It is an object of the present invention to improve the lack of time information while increasing the amount of information in the body axis direction due to the missing data.

[0015]

According to the positron emission CT apparatus of the present invention, a large number of detectors arranged in a ring shape around an object to be measured are axially moved from a first position to a second position.
A positron emission CT apparatus that reciprocates to a position and detects gamma rays from inside the object to be measured by a detector, and constructs an image of the inside of the object to be measured in a slice shape. a coincidence circuit for detecting whether the γ-rays are simultaneously emitted, an address conversion circuit for generating sinogram data from the position of the detector pair that has detected the pair of γ-rays emitted simultaneously, and a first position. A first data acquisition memory for storing first frame data obtained by acquiring sinogram data for a predetermined time;
A second data acquisition memory for storing second frame data obtained by collecting sinogram data at a position for a certain period of time, and a first frame data corresponding to a slice at a first position. The first interpolation data corresponding to the time when the first frame data is obtained is created, and the image inside the DUT at the time when the second frame data is obtained is processed corresponding to the slice at the second position. An image inside the measured object is constructed from the first frame data, and an image inside the measured object corresponding to the slice at the first position is stored in the first frame data.
And arithmetic means for constructing from the interpolation data.

The first interpolation data S ₂ corresponds to an intermediate time between the _first frame data S ₁ and S _{3 which} are temporally adjacent to each other,
For the frame data S ₁ and S ₃ , “S ₂ = k ₁ S ₁
+ K ₃ S ₃ (however, the coefficients k ₁ and k ₃ are coefficients determined by the radionuclide and time inside the object to be measured) ”.

The calculating means calculates the second frame data from
A second interpolation data corresponding to the time point when the first frame data is obtained is further created corresponding to the slice at the second position, and the inside of the device under test at the time when the first frame data is obtained is obtained. For the image, an image inside the device under test corresponding to the slice at the first position is constructed from the first frame data, and an image inside the device under test corresponding to the slice at the second position is obtained from the second frame data. May be constructed from the interpolation data of

A third data collection memory for storing the first frame data and a fourth data collection memory for storing the second frame data are further provided, and the first to fourth data collection memories are provided. First and second frame data obtained by collecting sinogram data at the first and second positions are sequentially stored in the memory, and the calculating means stores the second or fourth data in the second or fourth data collection memory. When storing the frame data, the first or third frame data stored in the first or third data collection memory is used to determine the second stored frame data.
Alternatively, first interpolation data corresponding to the time point when the second frame data in the fourth data acquisition memory is obtained is created, and the second or fourth data acquisition that has been stored is completed from the first interpolation data. Constructs an image of the inside of the device under test at the time when the second frame data in the memory is obtained,
When the first frame data is stored in the first or third data collection memory, the calculation means starts storing the second frame data in the second or fourth data collection memory. The second interpolation data corresponding to the time when the first frame data in the completed first or third data acquisition memory is obtained is created, and the stored first or third data is stored from the second interpolation data. The method may be characterized in that an image of the inside of the device under test corresponding to the time when the first frame data in the data acquisition memory of No. 3 is obtained is constructed.

The image processing apparatus may further include an interpolation data memory for storing the first and second interpolation data.

[0020]

In the positron emission CT apparatus of the present invention, a pair of γ's is generated by annihilation of a pair of electrons inside the measured object.
When the pair of gamma rays are detected simultaneously by the detector pair, the coincidence circuit detects whether the pair of gamma rays are simultaneously emitted. If the γ-rays are emitted simultaneously, it is determined that the γ-rays have been emitted at the same time due to the annihilation of the pair of electrons.
The sinogram data is collected according to the first and second positions and stored as first or second frame data in the first and second data collection memories, respectively. With respect to the image inside the DUT at the time when the second frame data is obtained by the calculation means, not only the image inside the DUT corresponding to the slice at the second position can be obtained from the second frame data, , An image inside the DUT corresponding to the slice at the first position is obtained from the first interpolation data. Therefore, the obtained image of the inside of the measured object has a higher resolution.

Here, the calculating means further creates, from the second frame data, second interpolation data corresponding to the slice at the second position and corresponding to the time when the first frame data is obtained. Using the second interpolation data, for the image of the DUT at the time when the first frame data is obtained, the image of the DUT corresponding to the slice at the second position is obtained from the second interpolation data. By constructing, the constructed image of the measured object can further increase the information amount in the body axis direction.

The apparatus further comprises third and fourth data acquisition memories, wherein the frame data is sequentially stored in the first to fourth data acquisition memories. Since the first or second interpolated data can be created from the completed first and second frame data to construct an image of the inside of the device under test, processing can be performed at higher speed.

[0023]

An embodiment of the present invention will be described with reference to the drawings.
The description of the same or equivalent components as those of the above-described conventional example will be simplified or omitted.

The present invention is an apparatus for collecting image data using a radiopharmaceutical. The apparatus searches for the measurement start / end time of each stored frame data, and searches for data loss in the time axis direction. And generate interpolation data. In this way, data loss is interpolated to obtain more continuous data.

FIG. 1 schematically shows the hardware of the positron emission CT apparatus of the present invention.
As in the above-described conventional example, the image processing apparatus includes an image data collection device 310, a signal processing circuit 315, a computer 350, a data memory 340, and a data storage device 370. The computer 350 and the data memory 340 generate interpolation data. This is different from the conventional example described above.

In brief, first, the computer 350 sends a data collection command to the data collection device 310 via a path 355 (control line). Data collection device 3
Numeral 10 reciprocates the gantry according to the frame switching signal to start data collection corresponding to each of the positions A and B. Here, when the gantry is moved, the position B is exactly at the center of the slice, and the distance between the positions A and B is 1/4 of the interval between the detector rings (half the distance between slices = the distance between rings). 1/4).

After the obtained data is stored in the data memory 340 through the signal processing circuit 315, it is stored in the data storage device 370 through the path 356 (data bus) at regular intervals. In response to the stored data, a separately prepared sub CPU 352 (this is a case of a multi-CPU; in the case of a multi-task, becomes another process) reads data from a path 359 (data bus) and outputs each frame. Retrieve data time information (frame start / end time). At this time, if the measurement end time of one frame data does not match the measurement start time of the next frame data, interpolation data is created on the assumption that data loss has occurred. If real-time processing is not always necessary and interpolation is not a problem even after the end of data collection, instead of using multiple CPUs using sub CPUs,
The processing of the sub CPU can be substituted by the processing of the computer 350.

FIG. 2 is a diagram showing an example of the configuration of a positron emission CT apparatus of the present invention together with its processing flow. Compared to the above-described conventional example, the data collection memory 340
c · 340d, a data interpolation computer 352, and an interpolation data memory 411 are added. Here, the case where the detector is at the origin is denoted as position A, and the case where the detector is moved is denoted as position B. The frame data name at position A is a and the frame data name at position B is b. The frame switching signal 25 from the main computer 350 is displayed.
0, the data collection memory 340a
340c stores the frame data at the position A, and the data collection memories 340b and 340d store the frame data at the position B. Frame data is stored in the memories 340a, b,
The data is stored in the order of c and d, and the data collection memory 340a
40c stores the frame data at the position A alternately with time, and the data collection memories 340b and 340d store the frame data at the position B alternately with time. The interpolation data memory 411 is used for performing processing in the data interpolation computer 352.

When the measurement is started, the coincidence circuit 320 determines whether or not the γ-ray detection signal captured by the detectors 312 arranged in a ring of the detector system 310 is coincidence counting. The coincidence count data is converted by the address conversion circuit 330 from the address of the detector pair to the address of sinogram data (also called T-θMap). Each frame data is stored in the data collection memories 340a, 340b,
The data is temporarily stored in the order of 340c and 340d, and the main computer 350 reads these data as before.

On the other hand, the data interpolation computer 352
Is the frame switching signal 2 obtained from the main computer 350
50, use state of the data collection memory is detected,
Data collection memory 340a / 340b / 340c / 34
Data for creating interpolation data is determined from among 0d that are not selected by the frame switching signal 250. After that, using the data stored in the non-selected one, the data interpolation computer 352 creates interpolation data by interpolation calculation described later.

Table 1 shows, for each frame from frame 1, the detector positions A and B at that time and the memories 340a, 340b, 340c and 340c in which the frame data is stored.
340d is shown, and the stored data is shown by adding a lowercase letter of the position A or B of the detector to the frame number. From frame 1 to frame 3, the frame data 1a, 2b, 3a are stored in the memory 340a.
After being stored in 340b and 340c, the image is composed by the image reconstruction device 360 and stored in the data storage device 370, respectively.

[0032]

[Table 1]

When the measurement of frame 4 is performed, the frame data 1 of frame 1 and frame 3
Interpolation data 2a of frame 2 is created using a and 3a, and stored in the interpolation data memory 411. The interpolation data 2a is read out by the main computer 350, an image of the frame 2 is formed by the image reconstruction device 360 using the interpolation data 2a and the frame data 2b of the frame 2, and is stored in the data storage device 370. Interpolation data 2
a is slice 1 of frame 2 without “′” in FIG.
Are used for the configuration of the images of 2, 3,... That is, using the frame data of frame 1 and frame 3 where the detector is at position A, interpolation is performed on an image (slice without “′”) of the slice at position A of the detector at the time of frame 2. Become.

Similarly, after frame 5, the two frames of data when the detector is at position A are used, and the frame when the intermediate detector is at position B is used for each slice at position A. Interpolate the image. Interpolation data for the frame at the immediately preceding time point is created using the frame data of the immediately preceding and the immediately preceding frame, and an image corresponding to the frame of the immediately preceding time point is configured.
0 is stored.

Using the interpolation data, the image of the missing part in FIG. 10 is constituted by the image reconstructing device 360 and stored in the data storage device 370, and the entire image of FIG. 10 is constituted. In this manner, the interpolation of the image in the previous two frames is performed. When each slice is viewed in the time direction (the frame direction in FIG. 10), the frame data of each slice is interpolated, and the time resolution is improved. Will be. When each frame is viewed in the Z-axis direction (slice direction in FIG. 10), the slice is interpolated for each frame, and the spatial resolution (in the body axis direction) is improved.

FIG. 3 shows PET data in the time axis direction. The number of radiation counts obtained per unit time is represented by the theoretical attenuation f (t) with respect to the elapsed time of measurement as shown in (a) and the physiological change g with respect to the elapsed time of measurement as shown in (b). (T) is superimposed on the value f
(T) * g (t) ((c) in the figure). The obtained image data is an integrated value at the frame time as shown by the shaded portion in FIG. Therefore, when a certain frame data is missing, f (t) * g (t) for the frame time is obtained.
f (t) is estimated from the theoretical attenuation curve from before and after the frame in which g (t) is missing, and interpolation is performed as follows.

Here, as shown in FIG. 4, from the measurement start time t ₀ to the end of frame 1 data collection (start of frame 2 data collection) time t ₁ , and from the start of frame 3 data collection (data of frame 2) Data collection at the position A is performed from the time t ₂ to the data collection end time t _{3 of the} frame 3, and the position is collected from the data collection start time t ₁ of the frame 2 to the data collection end time t _{2 of the} frame 2. Assuming that data is collected at B,
Without interpolation, the data at the time t ₁ to time t ₂ is missing for data at position A. That is, in the conventional example, and the data of the frame 2 is missing at position A at time t ₁ and the time t ₁₂ approximating the middle of t _2, so that the image at this position is missing.

[0038] If the biological variation is linear, the time t ₁₂ can be set at an arbitrary time between time t ₁ and time t _2. However, because it is actually non-linear (a function of time),
Time t ₁₂ is it is reasonable to approximate set to the midpoint of the time t ₁ and time t _2. Therefore, in this embodiment, as shown in FIG. 4, the detector is the boundary of the time t ₁₂ in the position B, and interpolating and reproducing sinogram data of the frame 2 from the frame 3 of the frame 1 and rear side front.

Here, the measured images of frame 1 and frame 3 are converted to frame data S in order to make it easy to write equations.
Assuming that ₁ · frame data S ₃ , these indicate the number of radiation counts obtained per unit time as frames 1, 2.
The frame 3 and the frame 3 are integrated within the frame time. Therefore, the frame data S ₂ to be interpolated can be approximated by expressing the frame data S ₁ and the frame data S ₃ by multiplying them by a certain coefficient. In this embodiment, the interpolation is performed as shown in Expression (1). (The same applies to other frames).

[0040]

(Equation 1)

The coefficients k ₁ and k ₂ are determined as follows.

First, the frame data S ₁ is represented by the following equation (2) (where λ = ln (2) / Half time (half-life)).

[0043]

(Equation 2)

Accordingly, the activity (Activity) A _{01 at} time t ₀ obtained from S ₁ is expressed by the following equation (3).

[0045]

(Equation 3)

Next, the portion S _2a of the frame 2 corresponding to the time t ₁ to the time t ₁₂ is expressed in the upper part of the equation (4), and when the equation (3) is substituted into this, the data S _2a becomes Can be represented using ₁ .

[0047]

(Equation 4)

Thus, the coefficient k ₁ is expressed by equation (5).

[0049]

(Equation 5)

Similarly, the frame data S ₃ is expressed by the following equation (6), and the time t ₀ obtained from S _{3 based} on the expression (6).
Activity (Activity) A ₀₃ is expressed by Equation (7) in the.

[0051]

(Equation 6)

[0052]

(Equation 7)

The portion S _2b of the frame 2 corresponding to the period from the time t ₁₂ to the time t ₂ is represented by the upper part of the equation (8), and when the equation (7) is substituted into this, the data S _2b uses the data S ₃ Can be expressed as From this, the coefficient k ₃ is expressed by Equation (9).

[0054]

(Equation 8)

[0055]

(Equation 9)

The frame data S ₂ to be interpolated is the data S
_{Since this} is the algebraic sum of _2a and data S _2b , data S ₂ is represented by equation (10).

[0057]

(Equation 10)

As described above, the coefficients k ₁ and k ₃ to be multiplied at the time of interpolation take different values depending on the half-life and time, and are obtained by using the equation (10) as the interpolation source data shown in Table 1. Interpolate the data shown in the bottom column of the table by multiplying by a coefficient. This coefficient is calculated in advance before the measurement and referred to during interpolation, so that high-speed processing can be performed. Table 2 shows that the nuclide ¹⁵ O was used, and the frame data collection intervals were 10 seconds and 100 seconds.
It shows an example of the coefficients k ₁ and k ₃ .

[0059]

[Table 2]

In the present invention, even-numbered data at position A and odd-numbered data at position B have been lost in the past. In the present invention, however, the portion is interpolated to obtain an image with higher axial resolution. Since the main measurement system including the main computer 350 is not burdened during the interpolation processing, high-speed interpolation can be realized.

In the invention described in Japanese Patent Laid-Open Publication No. Hei 2-134589, as a means for realizing this, the ring type detector is continuously moved within a range requiring interpolation, and data is continuously collected during the movement. ing. As a means for the interpolation, mechanical scanning is used, and complicated calculations are not required. However, since data is collected while moving, the spatial resolution in the Z-axis direction is reduced by an amount corresponding to the moving distance. . In this publication, the gazette states that "data was collected substantially at the same time", but data is not collected at the same time. Also, the Z-motion mechanism itself is mechanically complex.

On the other hand, in the present invention, interpolation data is created by using data before and after frame data requiring interpolation as an implementation means, and data is not collected while moving. , The spatial resolution in the Z-axis direction does not decrease. Further, since the interpolation process operates independently of the main measurement system, no load is imposed on the main measurement system. Further, the Z-motion mechanism may be any mechanism that can perform a simple step operation.

In the above-described embodiment, when it is desired to interpolate the start (or end) frame data (for example, frame 1), the "previous or subsequent data" (for example, frame 0)
Since it does not exist, it is necessary to use it together with another interpolation method. In such a case, the method can be covered by automatically performing a preliminary scan that does not require storage, or adding a function of automatically performing interpolation using a plurality of frame data after measurement is completed. In this case, it is possible to cope only by adding a program on the interpolation computer.

In this embodiment, even if interpolation is performed, loss of quantitativeness and deterioration of the S / N ratio are small and the frame data is stored as it is, so that loss interpolation can be performed even after the measurement is completed. Further, even if data loss occurs for some reason, the missing portion can be immediately interpolated. Furthermore, Z
Even if it is known that the same part of the image data cannot be obtained successively as in the case of motion, the continuous data can be pseudo-interpolated. The parameters required for the interpolation are only the half-life of the nuclide used and the frame start and end times of the data on which the nuclide is based, so that the calculation is easy.

The present invention is not limited to the above-described embodiment, but can be variously modified.

Here, the calculation method using the frame data before and after the lost frame is taken as an example, but there are some other interpolation methods themselves. Examples include a least square method using a plurality of frame data, and a method of directly calculating a biometric change g (t) using a higher-order approximation formula. In the above-described embodiment, “fast” and “easy” A simple interpolation method with linear approximation is described in the sense of calculating. Also,
This simple interpolation method has a drawback in that interpolation cannot be performed unless data exists on both sides of the lost frame (for example, data in a hatched portion in FIG. 1). In such a case, another interpolation method ( For example, interpolation of a lost frame can be performed using slice 1 and slice 2 to interpolate slice 1 ′. As the interpolation method, an interpolation method for approximating the g (t) component using the previous and next frames, an interpolation method using a least square method using a plurality of frame data, and a biological change g ( An interpolation method by directly calculating t), a combination with a simple interpolation method, a combination of various interpolation methods with all data, and the like.

While the principle of operation of the present invention is illustrated in FIG. 2, several further developments are possible.

First, four data collection memories are shown. However, by using a plurality of data acquisition memories, not only interpolation using two frame data but also interpolation calculation using a plurality of frame data becomes possible.

By providing a plurality of interpolation data memories, it is not necessary to access the main computer during the measurement time, so that it is possible to cope with a case where the frame time is short.

Further, by increasing the number of data acquisition memories and interpolation data memories to a plurality, the frame time and the interpolation method are widened (although the system becomes complicated and expensive).

[0071]

As described above, according to the present invention, with respect to the image inside the DUT at the time when the first and second frame data are obtained, the inside of the DUT corresponding to the slice at the first and second positions is obtained. Is obtained not only from the first and second frame data, but also the image inside the DUT corresponding to the slice at the first position is obtained from the second and first interpolation data. The image inside the object has a higher resolution.

[Brief description of the drawings]

FIG. 1 is a diagram schematically illustrating a hardware configuration of an embodiment of the present invention.

FIG. 2 is a system block diagram of an embodiment of the present invention.

FIG. 3 is a diagram showing PET data in a time axis direction.

FIG. 4 is a diagram for explaining an interpolation method.

FIG. 5 is a diagram schematically showing the entirety of a conventional positron emission CT.

FIG. 6 is a diagram illustrating a sensitivity distribution in a Z-axis direction of each slice.

FIG. 7 is a diagram illustrating a sensitivity distribution in a Z-axis direction of each slice when a Z-motion scan is performed and not performed.

FIG. 8 is a diagram showing an example of a PET apparatus having a Z-motion mechanism together with its process flow.

FIG. 9 is a diagram illustrating an actual configuration of an image data collection device.

FIG. 10 is a diagram schematically illustrating image data obtained by performing a Z-motion scan, using a slice number on a vertical axis and a frame number on a horizontal axis.

FIG. 11 shows a specific slice (1.2.3) of PET data.
.. Or 1 ′, 2 ′, 3 ′,...) In the time axis direction.

[Explanation of symbols]

310: detector system, 320: coincidence circuit, 33
0: Address conversion circuit, 340a to 340d: Data acquisition memory, 350: Main computer, 352: Data interpolation computer, 369: Image reconstruction device, 411
Memory for data interpolation.

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-2-134589 (JP, A) JP-A-2-40586 (JP, A) JP-A-63-238487 (JP, A) JP-A-5-205 11054 (JP, A) JP-A-61-159178 (JP, A) JP-A-63-115079 (JP, A) Japanese Utility Model Application No. Sho 61-181385 (JP, U) (58) Fields investigated (Int. ⁷ , DB name) G01T 1/161

Claims (5)

(57) [Claims] 1. A large number of detectors arranged in a ring shape around an object to be measured are reciprocated in an axial direction from a first position to a second position to detect γ-rays from inside the object. What is claimed is: 1. A positron emission CT apparatus configured to detect an image of the inside of an object to be measured in a slice shape by detecting with the detector, wherein coincidence for detecting whether a pair of γ-rays detected by the detector are simultaneously emitted. A circuit; an address conversion circuit for generating sinogram data from a position of the detector pair that has detected the pair of gamma rays emitted at the same time; and acquiring the sinogram data at the first position for a certain period of time. A first data acquisition memory for storing first frame data; and a second data acquisition memory for storing second frame data obtained by acquiring sinogram data at the second position for a predetermined time. Data acquisition memory, and, from the first frame data, create first interpolation data corresponding to a slice at the first position and corresponding to a time point when the second frame data is obtained; For an image inside the device under test at the time when the second frame data is obtained, an image inside the device under test corresponding to the slice at the second position is constructed from the second frame data, and Calculating means for constructing an image of the inside of the device under test corresponding to the slice at the first position from the first interpolation data. Wherein said first interpolation data S ₂ corresponds to the intermediate time of the first frame data S _1, S ₃ that chronologically successive, relative to the frame data S _1, S _3, S _2. The positron according to claim 1, wherein: ₂ = k ₁ S ₁ + k ₃ S ₃ (where coefficients k ₁ and k ₃ are coefficients determined by radionuclide and time inside the object). Emission CT device. 3. The arithmetic unit further includes, from the second frame data, second interpolation data corresponding to a slice at the second position and corresponding to a time point when the first frame data is obtained. While creating, for the image inside the DUT at the time of obtaining the first frame data,
An image inside the device under test corresponding to the slice at the first position is constructed from the first frame data, and an image inside the device under test corresponding to the slice at the second position is obtained. The positron emission C according to claim 1, wherein the positron emission C is constructed from the second interpolation data.
T device. 4. The apparatus further comprises: a third data collection memory for storing the first frame data; and a fourth data collection memory for storing the second frame data. The fourth data acquisition memory sequentially stores the first and second frame data obtained by acquiring sinogram data at the first and second locations, When the second frame data is stored in the data collection memory of No. 4,
Alternatively, from the first frame data stored in the third data collection memory, the second frame data corresponding to the time when the second frame data stored in the second or fourth data collection memory is obtained. 1 is created, and the first
And constructing an image inside the device under test corresponding to a point in time when the second frame data stored in the second or fourth data acquisition memory that has been stored is obtained from the interpolation data of When the first frame data is stored in the first or third data collection memory, the second
Alternatively, from the second frame data stored in the fourth data collection memory, the first frame data corresponding to the time point when the stored first frame data in the first or third data collection memory is obtained. 2 interpolation data, and the second
The image of the inside of the device under test corresponding to the time when the first frame data stored in the stored first or third data acquisition memory is obtained from the interpolated data of (1) or (2). 4. The positron emission CT apparatus according to 3. 5. The positron emission CT apparatus according to claim 3, further comprising an interpolation data memory for storing said first and second interpolation data.