CN108763758A - A kind of GATE emulation modes of non-complete ring-type PET rotation sweep patterns - Google Patents

Abstract

The invention discloses a GATE simulation method of an incomplete annular PET rotation scanning mode. The steps of the method are as follows: pre-simulation model and parameter preparation, including establishment of an incomplete annular PET geometric model under the GATE platform, radioactive source model setting, crystal Sensor parameter setting, calculation and setting of detector single rotation angle and number of rotations, physical interaction setting between particles, analog-to-digital conversion circuit parameter setting, output file format setting, initial scanning time, time segment, overall simulation time setting; simulation In the first stage, the initial position of the detector is determined. After completing the static scanning at this position, the incomplete detector ring is rotated and the above scanning is repeated. When the overall scanning time reaches the preset value, the simulation ends. Through the simulation method of the present invention, simulation verification is provided for a small amount of crystal sensors to complete the PET equipment construction scheme, which is helpful to the optimization of the rotary PET collection mode and equipment development, and reduces the equipment manufacturing cost.

Description

A GATE Simulation Method of Non-Complete Annular PET Rotary Scanning Mode

technical field

The invention belongs to the field of nuclear physics simulation experiments, and in particular relates to a GATE simulation method aiming at an incomplete annular PET rotation scanning mode.

Background technique

In recent years, radiation detection technology has been widely used in industry, agriculture, medical and health, and scientific research. Due to the continuous research and development of various powerful and complex radiation detection devices and spectrometers, early detector development and research methods and inability to Adapt to needs. At present, various scientific research institutions and equipment development companies generally use computer simulation to assist the design of detectors. As a random number method, the Monte Carlo method is similar to the physical process of energetic particles in the detector medium, so this method has become It is an important means for scientific researchers and engineers to assist in the design of detectors.

Positron Emission Tomography (PET) is a cutting-edge imaging method in today's medical field. It is based on the principle of positron annihilation and can noninvasively, quantitatively and dynamically It can accurately assess the metabolic level, biochemical reaction and functional activity of various organs. With the continuous development of PET technology, its application fields are no longer limited to the medical field. For example, PET imagers for plant research and PET imaging equipment for particle tracking in the industrial field all reflect the application value of PET imaging in various fields. . The γ-photon pairs produced by the annihilation of positrons have strong penetrating ability and are less affected by materials and environmental factors. The positron nuclide is marked on a suitable carrier such as liquid, gas or solid particles, injected into the device, and then Detection and imaging based on peripheral PET equipment can realize the positioning of marked particles inside industrial equipment. However, once the current medical PET detection ring is built, the layout is difficult to change, and the number of detectors in the whole ring is large, and the hardware is complex, which is not suitable for industrial equipment with complex and changeable structures; at the same time, due to the high price of PET sensors, many of them are under research In the stage of detector design, especially the feasibility verification test, it is impossible to directly carry out the test through PET hardware equipment.

Contents of the invention

Purpose of the invention: In view of the above existing problems in the prior art, the present invention aims to provide a GATE simulation method of incomplete annular PET rotation scanning mode, and proposes to use crystal sensors to add rotation scanning to complete the construction of incomplete PET equipment. Unlimited, making it more flexible in the field of industrial applications; using the GATE simulation method to simulate the rotation scanning mode of real PET equipment, providing simulation verification and optimization reference for the construction plan of rotary PET equipment, and providing assistance for the development of rotary PET equipment .

Technical scheme: in order to realize the purpose of the present invention, the technical scheme adopted in the present invention is: a kind of GATE emulation method of incomplete circular PET rotation scanning mode, and the steps of this emulation method are as follows:

(1) The simulation environment is the GATE platform under the Linux Ubuntu system, and a PET detector model is established on the platform. There are multiple choices for the number of sensor pairs in the detector, and they are arranged in a ring at equal intervals;

(2) Merge and convert the two-dimensional grayscale slice images of the preset number of radioactive source models into a radioactive source model format readable by GATE, adjust the size of the model, and set the material distribution and kernel of the model according to the pixel grayscale and the GATE material database Configuration documentation for prime activity distribution;

(3) Set the parameters of the crystal sensor and determine the numbering rules of the crystal strips inside the sensor;

(4) Solve the rotation parameters of the PET detection ring according to the size of the crystal sensor, select an integer multiple of the center angle corresponding to the single crystal strip as the rotation step angle according to the imaging resolution requirements, and calculate the detection ring through the logarithm of the sensor and the set rotation step angle The number of revolutions to complete a full circular scan;

(5) Set the analog-to-digital conversion circuit parameters, particle interaction, and output file format in the PET detector;

(6) Set the initial scan time of the detector, the time segment and the overall scan time;

(7) In the simulation stage, the initial position of the detector is determined, and the scan is performed according to the time segment. After the position scan is completed, the incomplete detector ring is rotated for the next scan. When the overall scan time reaches the preset value, the simulation is stopped;

(8) After the whole scanning process is completed, angle correction is performed on the output coincidence event data, and the correction is used for data reorganization, and the reorganized sinogram is used for image reconstruction.

Among them, in step (1), a three-dimensional coordinate system is established on the GATE simulation platform, and a PET detector model is constructed. The detector model includes a detection ring T, n crystal sensors M0～Mn, and the sensors are distributed at equal intervals with an inner diameter of D mm On the detection ring T, and the plane where the ring is located is parallel to the xOy plane, where M0 is the initial sensor number, and the sensor system increases clockwise from M0 to Mn when viewed from the negative direction of the z-axis.

Among them, in step (2), the two-dimensional slices of the radioactive source model are merged and converted into ".img" and ".hdr" formats, and the size of the radioactive source model is set by adjusting the pixel size, and set in different pixel grayscale areas. Different materials and different activities of radioactive sources in the model, the final GATE platform can directly read the radioactive source models stored in ".img" and ".hdr" formats.

Wherein, in step (3), the sensor parameters are set, including the number of crystal strips inside the sensor, array mode and size, the internal array of the crystal sensors M0-Mn is L*H, wherein the number in the y direction is L, the number in the z direction is H, and the number of crystals in the z direction is H. The width of the bottom of the bar is w mm, the x direction is the length of the crystal bar, and the crystal bars are numbered in the order of increasing in the y direction and then increasing in the z direction, wherein the positive direction of the y axis is a column, and the positive direction of the z axis is a row.

Among them, in step (4), the rotation scanning step angle θ _{n is} set according to the integer multiple β of the central angle α corresponding to the single crystal strip, and the detection ring is calculated to complete the entire The number of rotations M of the circular scan is calculated as follows: the diameter of the detection ring is D millimeters, and the width of a single crystal strip is w millimeters. By approximation, the central angle corresponding to a single crystal strip can be obtained Single rotation angle θ _n = βα°, number of rotations

Among them, in steps (6) and (7), set the simulation start scan time, time segment, overall scan time and initial scan position of the detector, the simulation start time is set as T _start , and the time segment is determined by the detector at each position The scan time T _slice is determined, the overall scan time T _total is determined by the number of detector rotations M, satisfying T _total = M*T _slice , determine the initial position R0 of the detector in the simulation stage, scan according to T _slice , after completing the scan at this position Rotate the detector and complete the next scan, repeat the scan until the overall scan time reaches the set value to stop the simulation.

Wherein, in step (8), after the scanning is completed, the document that stores the event data is obtained, by establishing a virtual complete ring, the inner diameter is D mm, the crystal strips on the ring are connected and the size is also the same as that of the incomplete detector The size of the crystal strips on the ring T is the same, and the number of the crystal strips at its starting position is 0, and the virtual numbering is performed clockwise from the negative direction of the z-axis.

Among them, in step (8), the output coincident event data includes the energy, time, and crystal number information of the received gamma photons, and the response line is the connection between two crystal bars that receive gamma photons almost simultaneously. The rotated total angle is combined with the number information of the crystal strip, and mapped with the number of the virtual ring, and the number of the coincidence of the position of the crystal strip with the crystal strip on the virtual ring is corrected to the number on the virtual ring, and the coincidence event collected by rotating the detector M times All the data are corrected by the number of crystal strips, and the corrected number can reflect the position of the response line on the virtual ring, and then the single-slice data reorganization algorithm is used for the coincident event data and stored in the sinogram.

Beneficial effects: Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

(1) Through the present invention, to a certain extent, simulation verification is provided for various numbers of crystal sensors to complete the PET equipment construction scheme, and the proposed scanning mode with an integer multiple of a single crystal bar as the rotation step value reduces the lack of response lines to a certain extent horn.

(2) The performance indicators of the detectors obtained from the simulation will provide assistance for the development of rotary PET equipment, help to optimize the rotation acquisition mode, and increase the flexibility of PET detectors.

Description of drawings

Fig. 1 is the implementation flow chart of the GATE simulation method of incomplete circular PET rotation scanning mode;

Figure 2 is a schematic diagram of the distribution model of the non-complete annular PET sensor under the GATE simulation platform;

Figure 3 is a schematic diagram of the radioactive source model and two-dimensional slices in the GATE simulation platform;

Fig. 4 is a schematic diagram of the internal crystal array of the crystal sensor in Fig. 2 and its array encoding rules;

Fig. 5 is a schematic diagram of the simulation process of the PET model rotation acquisition mode;

Fig. 6 is a schematic diagram of the correction principle based on crystal coding;

Fig. 7 is the sinogram diagram obtained by the angle-corrected data reorganization algorithm of the complete ring and incomplete PET detectors.

Detailed ways

The technical solutions of the present invention will be further described below in conjunction with the accompanying drawings and embodiments.

Example 1

The invention discloses a GATE simulation method of a non-complete ring-shaped PET rotation scanning mode. The method flow is shown in Figure 1. The ring-shaped PET model is set under the GATE platform before the simulation, the radiation source model is set, the crystal sensor parameter is set, and the detector is set. Calculation and setting of single rotation angle and number of rotations, setting of physical interaction between particles, setting of analog-to-digital conversion circuit parameters, setting of output file format, setting of simulation start scanning time, static scanning time of each angle during rotation, that is, time slice, And determine the overall scan time based on the product of the number of rotations and the time slice. In the simulation stage, the initial position of the detector is determined. After the static scanning at this position is completed, the incomplete detector ring is rotated and scanned repeatedly. When the overall scanning time reaches the preset value, the simulation ends, and the corresponding event data is obtained, and the angle is corrected to realize Data reorganization. The specific steps of the simulation method are as follows:

(1) The PET detector model is established on the simulation platform. There are multiple choices for the number of sensor pairs, and they are distributed at equal intervals. As shown in Fig. 2(a), incomplete detection ring T1, crystal sensor modules M0~M11; establish a three-dimensional coordinate system on the GATE simulation platform, this coordinate system is helpful for the subsequent spatial position adjustment of detectors and radioactive sources, generating The detector model is also shown in this coordinate system. Figure 2(a) is a PET detector model composed of 6 pairs of crystal sensors M. The crystal sensors M0~M11 are equally spaced on a ring with an inner diameter of 100 mm, and the plane where the ring is located is parallel to the xOy plane, where M0 is the starting point The sensor number, viewed from the negative direction of the z-axis, the sensor system number increases clockwise from M0 to M11.

(2) By superimposing a certain number of two-dimensional grayscale slice images of radioactive sources, a three-dimensional voxel model of radioactive sources is established, and the material distribution and nuclide activity distribution are set by pixel grayscale. The radioactive source model is a model established to simulate the nuclide distribution of the research object. As shown in Figure 3, S is the established three-dimensional model of the radioactive source, which is characterized by a cube containing four through holes of different shapes, which are respectively elliptical cylinders , large cylinder, small cylinder, and cube. The GATE platform cannot directly read 3D format files. L is a slice image made for the 3D model of the radioactive source. The slice image is a grayscale image with a pixel size of 128*128. The default pixel size is pixel /mm, the size of the model can be adjusted by setting the scale through the medical software. After superimposing a certain number of slices, the medical software converts and generates the model file format ".hdr" and ".img" that can be read by the GATE platform.

In the example, the model material and nuclide distribution are divided into two parts: the aluminum shell without nuclide, the black part, the gray value is 0, and the water containing nuclide, that is, the white part, the gray value is 255. The radioactive source model is in the first quadrant of the 3D coordinate system by default when imported, so if you want to place the model in another position in the coordinates, you need to adjust it through coordinate transformation. Finally, the radioactive source material configuration file and the nuclide activity distribution configuration file are respectively set up to complete the model establishment, and the materials are directly transferred to the material database in the GATE platform.

Among them, "Alphantom.dat" is the material configuration file, and "ActivityRange.dat" is the activity configuration file. An example of the content in the configuration document is as follows:

"Alphantom.dat"

2 #Material type

0～5Aluminium #The material whose gray value is in [0,5] has been defined in the material database

200～255Water #The material whose gray value is in [200,255] has been defined in the material database

"ActivityRange.dat"

2 # Number of nuclide values

0～5 0 #The activity of the grayscale value in [0,5] pixel is 0

200～255 2000.Bq #The activity of the gray value in [200,225] pixel is 2000Bq

(3) Set the parameters of the sensor, including the size, array mode and number of crystal strips inside the sensor. As shown in Figure 4(a), the internal array of the crystal sensor M0 in this example is 6*16, of which the number in the y direction is 6, and the number in the z direction is 16. mm, the x direction is the length of the crystal strip, which may be 15 mm to 20 mm, perpendicular to the positive direction of the x axis and intersecting at point p in the center of the crystal array. In FIG. 4 , the crystal sensor module M, the single crystal strip c, and the system numbers of the crystal strips are c0-c95.

For the numbering rules of the internal crystal array of the sensor M0, as shown in Figure 4(b), c0~c95 are crystal strip system numbers 0~95, and the crystal strip numbers are along the positive direction of the y-axis (this direction is specified as a column) and the z-axis ( It is stipulated that this direction is the row) increasing in the positive direction, and increasing first in the y direction and then in the z direction. Similar to M0, the internal crystal bar numbers of M1~M11 are seen from the negative direction of the z axis, first increase clockwise in the column direction, and then go to the z axis Positive increase.

(4) Solve the rotation parameters of the detector according to the size and logarithm of the crystal sensor, and obtain a single rotation angle, that is, the step angle. In the case of a small number of detectors, in order to reduce the incompleteness of the response line angle, the minimum step value is the central angle α corresponding to a single crystal strip inside the sensor. According to different imaging resolution requirements, you can choose 1 to 4 times α is a step value, and when α is larger, the scanning time can be shortened under the premise of satisfying the image quality. The rotation angle is calculated according to the size of the detection ring and the specifications of the crystal sensor. If the diameter of the detection ring is D millimeters and the width of a single crystal strip is w millimeters, the central angle corresponding to a single crystal strip can be obtained by approximate calculation:

According to the limitation of imaging accuracy and overall scanning time, the number of crystals for a single rotation is required to be β, and the single rotation angle is

_θn = βα° (2)

In order to prevent too many missing response lines, 1≤β≤4.

Among them, the number of rotations is calculated according to the logarithm of the crystal sensor and the rotation step angle. If there are N pairs of sensors distributed on the ring at equal intervals, and the rotation angle is the above-mentioned θ _n , then the number of rotations

Special case When N pairs of sensors form a complete detection ring, then M=0.

From the parameters in step (3), it can be seen that in this embodiment, w=2.15mm, D=100mm, N=6, and β=3, then a single rotation angle θ can be obtained according to formula (1) and formula (2) _n = 7.5°, the number of rotations M = 4 is obtained according to formula (3).

(5) The parameters of the analog-to-digital conversion circuit, particle interaction, and output file format need to be simulated according to the actual hardware performance or the performance index of the proposed detector, whether to study scattering and other issues, and set the data processing means. The energy resolution of a crystal sensor selected in this example is about 15%, the time resolution of the hardware circuit is 600ps, and the time window and energy window are respectively 10ns, 350Kev～650Kev, and the relevant simulation parameters can be set accordingly; the hardware requires rotation scanning The total time, whether to study scattering problems, data storage tools, etc. all determine the particle interaction and output format settings in the simulation process. In this example, positron annihilation, photoelectric effect, Compton scattering effect, Rayleigh scattering, etc. are mainly set. Only output matches events and are in ROOT format.

(6) Set the detector start scan time, time segment and overall scan time. The start time of the simulation is set to T _start , the time segment is determined by the scanning time T _slice of the detector at each position, and the overall scanning time T _total is determined by the number of rotations M of the detector, which satisfies formula (4):

T _total = M*T _slice (4)

In the actual situation, the scan start time is generally set to 0. In this embodiment, the time segment T _slice is set to 10s, and the number of rotations is known to be 4. Then according to formula (4), it can be determined that T _total is 40s, and the simulation time is determined accordingly. Set related parameters.

(7) Determine the initial position of the detector in the simulation stage, scan according to the T _slice , rotate the detector after completing the position scan and complete the next scan, and stop the simulation until the overall scan time reaches the set value. As shown in Figure 5, the detection rings R0~R3 under four rotation positions, (a) is the initial state R0 of the simulation model. Rotate R0 to obtain model R1, scan according to the time slice T _slice , and then rotate in the same steps to obtain R2 and R3, and stop the simulation until the overall scanning time reaches the preset value of 40s.

(8) After the scan is completed, the result obtained from the simulation is a large amount of coincident event data, which includes the energy, time, and crystal number information of the received gamma photons, and the response line is two crystals that received gamma photons almost at the same time strip connection. Extract the number data of the crystal bar in the coincidence event, correct the number according to the rotation angle position, and finally reorganize to obtain the sinogram. Considering the original single-slice reorganization algorithm, determine the position of the coincident event on the single-layer ring according to the crystal sensor number M and the crystal bar number c, as shown in Figure 5, after the rotation acquisition, the same crystal number of the original M0 will be Appears in multiple positions on the ring, and at this time it cannot be uniquely positioned based on the number. In FIG. 6( a ), after the crystal sensor M0 is rotated, the numbers c0 to c5 of the crystal bars appear repeatedly at multiple positions on the circumference.

The present invention establishes a complete ring with an inner diameter of D mm, the crystal bars on the ring are closely connected and the size is consistent with the size of the crystal bars on the incomplete detector ring, and the number of the crystal bars at the starting position is 0. From the negative direction of the z-axis, virtual numbering is performed clockwise in increments. Map the sensor crystal bar numbers in R1_M0, R2_M0, and R3_M0 in 6(a) to the sensor crystal bar numbers in VR1_M0, VR2_M0, and VR3_M0 in 6(b), and use an approximate method to align the crystal bar numbers and positions on the ring Match them one by one. In Fig. 6, R0_M0, R1_M0, R2_M0, and R3_M0 are the sensor module M0 in the four rotation positions respectively, and VR0_M0, VR1_M0, R2_M0, R3_M0 are the sensor module M0 in the four rotation positions after correcting the crystal numbers.

(9) Similarly, the crystal sensors M1-M11 and the crystal bars c6-c95 can also be mapped one-to-one by using the same mapping relationship. After the mapping relationship after each rotation is obtained, the coincident event data of multiple rotation scans are corrected, superimposed and reorganized to obtain the sinogram. Figure 7(a) is the sinogram after a single scan recombination under a complete circle, Figure 7(b) is the sinogram after a single scan recombination in Example 1, and Figure 7(c) is the recombination using numbering correction in Example 1 The sine diagram obtained by the algorithm, it can be seen that the effect of (c) and (a) is closer, and in (b) because of incomplete collection, many response lines are lost.

Example 2

In the specific implementation process, the number of crystal sensor pairs in the PET detection ring can also be other numbers, and the detector aperture and crystal strip size can also be determined according to requirements. Compared with Embodiment 1, this embodiment mainly embodies the calculation of the rotation parameters of the simulation process under the situation of different sensor logarithms, detector ring apertures, and crystal strip sizes. The specific simulation method steps are as follows:

(1) The PET detector model is established on the simulation platform. There are multiple choices for the number of sensor pairs, and they are distributed at equal intervals. As shown in Fig. 2(b), incomplete detection ring T2, crystal sensor modules M0-M11; establish a three-dimensional coordinate system on the GATE simulation platform, this coordinate system is helpful for the subsequent spatial position adjustment of detectors and radioactive sources, generating The detector model is also shown in this coordinate system. Figure 2(b) is a PET detector model composed of 4 pairs of crystal sensors M. The crystal sensors M0~M11 are equally spaced on a ring with an inner diameter of 100mm, and the plane where the ring is located is parallel to the xOy plane, where M0 is the starting point The sensor number, viewed from the negative direction of the z-axis, the sensor system number increases clockwise from M0 to M11.

The radioactive source model established in step (2) is the same as that in Example 1, and the steps are consistent.

(3) Set the parameters of the sensor, including the size, array mode and number of crystal strips inside the sensor. In this example, the internal array of the crystal sensor M0 is still as shown in Figure 4(a), which is 6*16, where the number in the y direction is 6, and the number in the z direction is 16. The difference from Figure 4(b) is that the width of the bottom surface of the gap crystal strip is considered Both are 3.15mm, and the x direction is the length of the crystal strip, preferably 15mm to 20mm, perpendicular to the positive direction of the x axis and intersecting at point p in the center of the crystal array. In Fig. 4, crystal sensor module M, single crystal strip c, crystal strip system numbers c0-c95;

For the numbering rules of the internal crystal array of the sensor M0, as shown in Figure 4(b), c0~c95 are crystal strip system numbers 0~95, and the crystal strip numbers are along the positive direction of the y-axis (this direction is specified as a column) and the z-axis ( It is stipulated that this direction is the row) increasing in the positive direction, and first the y direction and then the z direction, similar to M0, the internal crystal bar number of M1~M7 is viewed from the negative direction of the z-axis, first increases clockwise in the column direction, and then goes to the positive direction of the z-axis Increase.

(4) Solve the rotation parameters of the detector according to the size of the crystal sensor to obtain a single rotation angle, that is, the step angle. The PET detector model is composed of 4 pairs of crystal sensors M, the detection aperture D is 200mm, and the crystal strip width w is 3.15mm. If the incompleteness of the response line is expected to be smaller, the smaller the β should be, in this case β is 2, According to the calculation formula of step (4) in embodiment 1, can obtain Then the single rotation angle θ _n =βα°=3.6°.

Calculate the total rotation acquisition times of the detection ring according to the logarithm of the crystal sensor and the step angle. According to the calculation formula (3) of step (4) in embodiment 1, obtain the number of rotations It is not difficult to draw such a relationship: the number of sensors and the single rotation angle are inversely proportional to the number of rotation acquisitions.

Step (5) is the same as in Example 1.

(6) In this embodiment, the time segment T _slice is still set to 10s, and the number of rotations is known to be 12. According to the calculation formula (4) in Embodiment 1, T _total can be obtained as 120s, and the parameters related to the simulation time are set accordingly.

(7) Determine the initial position of the detector in the simulation stage, scan according to the T _slice , rotate the detector after completing the position scan and complete the next scan, and stop the simulation until the overall scan time reaches the set value. As shown in Figure 2(b), the initial state of the simulation model is that the x-axis passes through the center of the sensor M0. After the initial position scanning is completed with T _slice , the model is rotated at a rotation angle of 3.6° at the beginning of the next time slice. Scan according to the time slice T _slice , and then rotate in the same steps until the overall scanning time reaches the preset value of 120s to stop the simulation.

Steps (8), (9) will use the same virtual ring number in Example 1 to correct the angle of the collected coincident data, the difference is only the increase in the number of rotations, and the number of overlapping crystal strips is changed from 3 to 2. Still use the similar mapping relationship in Figure 6 for correction, where c2, c3 of R0_M0 will coincide with c0, c1 of R1_M0 after a single rotation, and the crystal numbers of VR1_M0 need to be corrected to c2, c3, c4, c5, c6, c7 , and the rest of the crystal strips can be similarly corrected.

Claims (8)

1. a kind of GATE emulation method of non-complete circular PET rotation scanning pattern, it is characterized in that, this emulation method step is as follows: (1) The simulation environment is the GATE platform under the Linux Ubuntu system. A PET detector model is established on this platform. There are multiple choices for the number of sensor pairs in the detector, and they are arranged in a ring at equal intervals; (2) Merge and convert the two-dimensional grayscale slice images of the preset number of radioactive source models into a radioactive source model format readable by GATE, adjust the size of the model, and set the material distribution of the radioactive source model according to the pixel grayscale and the GATE material database Configuration files with nuclide activity distribution; (3) Set the parameters of the crystal sensor and determine the numbering rules of the crystal strips inside the sensor; (4) Solve the rotation parameters of the PET detection ring according to the size of the crystal sensor, select an integer multiple of the central angle corresponding to the single crystal strip as the rotation step angle according to the imaging resolution requirements, and calculate the detection ring through the logarithm of the sensor and the set rotation step angle The number of revolutions to complete a full circular scan; (5) Set the analog-to-digital conversion circuit parameters, particle interaction, and output file format in the PET detector; (6) Set the initial scan time of the detector, the time segment and the overall scan time; (7) In the simulation stage, the initial position of the detector is determined, and the scan is performed according to the time segment. After the position scan is completed, the incomplete detector ring is rotated for the next scan. When the overall scan time reaches the preset value, the simulation is stopped; (8) After the entire scanning process is completed, angle correction is performed on the output coincidence event data, and the correction is used for data reorganization, and the reorganized sinogram is used for image reconstruction. 2. the GATE emulation method of a kind of non-complete annular PET rotation scan pattern according to claim 1, it is characterized in that, in step (1), in GATE emulation platform, three-dimensional coordinate system is set up, builds PET detector model, the The detector model includes a detection ring T, n crystal sensors M0~Mn, and the sensors are equally spaced on the detection ring T with an inner diameter of D mm, and the surface of the ring is parallel to the xOy plane, where M0 is the starting sensor number, Viewed from the negative direction of the z-axis, the sensor numbers increase clockwise from M0 to Mn. 3. The GATE simulation method of a kind of incomplete circular PET rotation scan mode according to claim 1, characterized in that, in step (2), the two-dimensional slice diagram of the radioactive source model is merged and converted into ".img ", ".hdr" format, set the size of the radioactive source model by adjusting the pixel size, set different materials in the model in different pixel gray areas, and different radioactive source activities, and finally the GATE platform can directly read ".img ", ".hdr" format stored radioactive source model. 4. the GATE emulation method of a kind of non-complete circular PET rotation scan pattern according to claim 1, it is characterized in that, in step (3), sensor parameter is set, comprises sensor inner crystal bar number, array mode and size, The internal array of crystal sensors M0~Mn is L*H, where the number in the y direction is L, the number in the z direction is H, the width of the bottom surface of the crystal strip is w mm, and the x direction is the length of the crystal strip. The crystal strips increase first in the y direction and then The numbers are numbered in ascending order in the z direction, where the positive y-axis is the column, and the positive z-axis is the row. 5. The GATE simulation method of a non-complete annular PET rotation scanning mode according to claim 1, wherein in step (4), the rotation scanning step is set according to the integral multiple β of the central angle α corresponding to the single crystal strip Angle θ _n , through the set step angle θ _n and the logarithm n of the crystal sensor, calculate the number of rotations M for the detection ring to complete the entire circular scan. The calculation method is as follows: the diameter of the detection ring is D mm, and the width of a single crystal strip is w mm, by approximation, the central angle corresponding to a single crystal strip can be obtained Single rotation angle θ _n = βα°, number of rotations 6. the GATE emulation method of a kind of incomplete annular PET rotation scanning pattern according to claim 1, is characterized in that, in step (6), (7), setting emulation initial scanning time, time segment, overall scanning Time and the initial scanning position of the detector, the simulation start time is set to T _start , the time segment is determined by the scanning time T _slice of the detector at each position, and the overall scanning time T _total is determined by the number of times M of the detector rotation, satisfying T _total =M*T _slice , determine the initial position R0 of the detector in the simulation stage, scan according to T _slice , rotate the detector after completing the position scan and complete the next scan, repeat the scan until the overall scan time reaches the set value to stop the simulation. 7. the GATE emulation method of a kind of non-complete circular PET rotation scan pattern according to claim 1, it is characterized in that, in step (8), after finishing scanning, obtain the file that stores conforming event data, by setting up a virtual The complete ring of , the inner diameter is D mm, the crystal strips on the ring are connected and the size is also consistent with the size of the crystal strips on the incomplete detector ring T, the number of the crystal strip at its starting position is 0, viewed from the negative direction of the z-axis Clockwise in increments for virtual numbering. 8. The GATE simulation method of a kind of non-complete annular PET rotation scanning mode according to claim 7, is characterized in that, in step (8), in the coincident event data of output, included the energy, time of receiving gamma photon , crystal strip number information, the response line is the connection of two crystal strips that receive γ photons almost simultaneously, and the total angle rotated by the detection ring is combined with the crystal strip number information to map with the virtual ring number, and the crystal strip position and The overlapping number of the crystal strips on the virtual ring is corrected as the number on the virtual ring, and all the coincident event data collected by the detector rotation M times are corrected for the crystal strip number, and the corrected number can reflect the position of the response line on the virtual ring. A single-slice data reorganization algorithm is then used for the coincident event data and stored in a sinogram.