CN110368009B - Method for correcting detection efficiency of PET detector - Google Patents
Method for correcting detection efficiency of PET detector Download PDFInfo
- Publication number
- CN110368009B CN110368009B CN201910735219.9A CN201910735219A CN110368009B CN 110368009 B CN110368009 B CN 110368009B CN 201910735219 A CN201910735219 A CN 201910735219A CN 110368009 B CN110368009 B CN 110368009B
- Authority
- CN
- China
- Prior art keywords
- source
- detection efficiency
- barrel
- crystals
- pet
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus for radiation diagnosis, e.g. combined with radiation therapy equipment
- A61B6/02—Devices for diagnosis sequentially in different planes; Stereoscopic radiation diagnosis
- A61B6/03—Computerised tomographs
- A61B6/037—Emission tomography
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus for radiation diagnosis, e.g. combined with radiation therapy equipment
- A61B6/58—Testing, adjusting or calibrating apparatus or devices for radiation diagnosis
- A61B6/582—Calibration
- A61B6/585—Calibration of detector units
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01T—MEASUREMENT OF NUCLEAR OR X-RADIATION
- G01T1/00—Measuring X-radiation, gamma radiation, corpuscular radiation, or cosmic radiation
- G01T1/29—Measurement performed on radiation beams, e.g. position or section of the beam; Measurement of spatial distribution of radiation
- G01T1/2914—Measurement of spatial distribution of radiation
- G01T1/2985—In depth localisation, e.g. using positron emitters; Tomographic imaging (longitudinal and transverse section imaging; apparatus for radiation diagnosis sequentially in different planes, steroscopic radiation diagnosis)
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01T—MEASUREMENT OF NUCLEAR OR X-RADIATION
- G01T7/00—Details of radiation-measuring instruments
- G01T7/005—Details of radiation-measuring instruments calibration techniques
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- High Energy & Nuclear Physics (AREA)
- Medical Informatics (AREA)
- Radiology & Medical Imaging (AREA)
- Surgery (AREA)
- Pathology (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Biophysics (AREA)
- Optics & Photonics (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- General Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Nuclear Medicine (AREA)
Abstract
The invention discloses a method for correcting the detection efficiency of a PET detector, which comprises the following steps: a1, acquiring scanning data of a PET system after a barrel source is placed; a2, determining the actual position information of the barrel source in the PET system according to the scanning data; a3, acquiring barrel source related correction information and intermediate information for correcting detection efficiency according to the actual position information; and A4, obtaining correction values of all crystal detection efficiencies according to the correction information, the intermediate information, the scanning data, the actual position information of the barrel source and the intensity of the radionuclide in the barrel source. The invention greatly reduces the complexity of the operation of the detection efficiency measurement experiment, can effectively reduce the radiation of workers in the experiment and effectively reduce the experiment cost.
Description
Technical Field
The invention relates to the field of medical imaging, in particular to a method for correcting the detection efficiency of a PET detector.
Background
Positron Emission Tomography (PET) is the only new imaging technology that can show the metabolism of living biological molecules, the activity of receptors and neuro-mediators at present, and is now widely used in the aspects of differential diagnosis, treatment effect evaluation, organ function research, new drug development and the like of various diseases. The working principle is to metabolize the necessary substances in the life of the organism, such as: glucose, proteins, nucleic acids, fatty acids, etc., labeled with short-lived radionuclides (e.g. 18 F， 11 C, etc.), and the characteristics reflecting the metabolic activity of the life are displayed through images by utilizing the different metabolic states of different tissues of the human body on the marked substances, such as vigorous glucose metabolism, more accumulation and the like in high-metabolic malignant tumor tissues, so that the aim of early diagnosing diseases such as tumors and the like is fulfilled.
During a PET scan, the radionuclide decays releasing a positron which annihilates with a free negative electron, producing two photons nearly back-to-back. The detector detects the photon pair so as to deduce the real occurrence position of the event; at the same time, how many photon pairs are detected reflects the intensity of the occurrence of the event. Only when the detection efficiency of all the detectors for photons is consistent, an image which truly reflects the activity distribution of the radionuclide can be obtained. However, the efficiency of detecting photons incident on the surface of different detection units of a PET system is not uniform due to factors such as the structure of the detector and the non-uniformity of the detector crystal. In order to obtain an unbiased estimation of the radionuclide activity distribution image, the inconsistency of the photon detection efficiency by different detection units is corrected back.
The correction experiment of the detection efficiency of present PET detector needs the people to place the bucket source at the center of detector, and then carries out the correction of the detection efficiency of PET detector. However, in practice, to ensure that the bucket source is centrally located, it is often necessary for the operator to touch the radioactive bucket source multiple times to precisely adjust the position of the bucket source. This not only makes the experiment complicated and difficult to operate, but also requires the experimenter to be subjected to large doses of radiation.
Disclosure of Invention
To solve the above problems, an object of the present invention is to provide a method for correcting the detection efficiency of a PET detector.
In order to achieve the purpose, the invention adopts the main technical scheme that:
in a first aspect, the present invention provides a method of correcting for the detection efficiency of a PET detector, comprising:
a1, acquiring scanning data of a PET system after a barrel source is placed;
a2, determining the actual position information of the barrel source in the PET system according to the scanning data;
a3, acquiring barrel source related correction information and intermediate information for correcting detection efficiency according to the actual position information;
and A4, obtaining correction values of all crystal detection efficiencies according to the correction information, the intermediate information, the scanning data, the actual position information of the barrel source and the intensity of the radionuclide in the barrel source.
Optionally, the step A2 includes:
a21, based on the scanning data, adopting a selected reconstruction algorithm to obtain a reconstructed image of a bucket source;
and A22, acquiring the gravity center according to the reconstructed image, and obtaining the actual position information of the bucket source by using a gravity center method.
Optionally, the step a21 includes:
a21-1, carrying out non-attenuation correction reconstruction on the scanning data to obtain a PET activity distribution image I;
a21-2, calculating an average value of all pixels in the first PET activity distribution image, selecting a preset threshold value matched with the average value, and selecting pixels which are larger than the preset threshold value in the first PET activity distribution image;
giving attenuation coefficients corresponding to the barrel source substances to the selected pixels according to the priori knowledge to obtain attenuation coefficient distribution I;
a21-3, reconstructing the scanning data with attenuation correction by using the attenuation coefficient distribution to obtain a PET activity distribution image II;
repeating the process of the step A21-2 and the step A21-3 for P times to obtain a PET activity distribution output image subjected to accurate attenuation correction;
where the iteration P is determined based on the performance of the PET detector, the activity size of the radionuclide in the barrel source, and/or the acquisition time.
Optionally, the step a21 includes:
a211, acquiring the acquired data of the barrel source according to other imaging systems, and acquiring attenuation coefficient distribution corresponding to a reconstructed PET activity distribution output image;
and A212, reconstructing the scanning data with attenuation correction by using the attenuation coefficient distribution to obtain a PET activity distribution output image.
That is, when the experiment is assisted by other imaging systems and a high-precision structural image can be provided for the reconstruction of the PET image, the distribution of the attenuation coefficient can be accurately defined, and the process of attenuation correction reconstruction can be directly carried out to obtain an output image subjected to accurate attenuation correction.
Optionally, the revision information associated with the bucket source includes one or more of:
correction factor SG representing the uneven detection efficiency caused by the influence of detector geometry on different LORs uivj ；
Correction factor FG representing non-uniformity of detection efficiency due to different LORs affected by phantom geometry uivj ；
The intermediate information for correcting the detection efficiency includes: number of crystals N ui 。
Optionally, the obtaining of the intermediate information for correcting the detection efficiency in step A3 includes:
a3-1, selecting LORs which pass through a barrel source part from LORs formed by any crystal and crystals opposite to the crystal and are symmetrical to the diameter of a detector based on actual position information of the barrel source;
a3-2, determining effective LORs according to the selected LORs and the actual position information of the barrel source to obtain the number N of crystals ui ；
Alternatively, the first and second liquid crystal display panels may be,
a31, based on the actual position information of the bucket source, selecting all LORs which pass through the bucket source and are symmetrical about the diameter of the bucket source;
a32, determining effective LOR according to the selected LOR and the actual position information of the barrel source to obtain the number N of crystals ui 。
Optionally, the step A4 includes:
a41, actual detecting the number of cases E according to each LOR in the scanning data uivj The intensity of the radionuclide in the barrel source and the correction information are obtained based on the following formula I uivj ；
Wherein E is uivj The actual detection case number u, v represents the axial number of any two crystals, and i, j represents the annular number of the crystals;
a represents the radionuclide intensity, SG uivj Image representationA correction factor for the detector geometry that is responsive to the detection efficiency;
FG uivj a correction factor representing a corresponding bucket source geometry that affects detection efficiency;
and A42, acquiring the corrected detection efficiency according to a formula II and a formula III.
Wherein N is ui Indicating the number of crystals.
In a second aspect, the invention provides a PET system comprising PET detectors which perform the method of any one of the first aspects.
The beneficial effects of the invention are:
centered placement of the bucket source is a requirement that is not easily met when the detection efficiency of the prior art is corrected. The method of the invention enables the detection efficiency measurement experiment to be free from strict centered placement of the barrel source, greatly reduces the operation difficulty of the experiment, and simultaneously reduces the radiation risk of workers.
In addition, the method of the invention fully utilizes the bucket source scanning data to obtain the real position of the bucket source through a computer program. And then modeling the deviation of theoretical detection efficiency caused by non-uniform irradiation by using the real position of the barrel source, and correcting the difference of the radioactivity integral value of the barrel source by the coincidence line formed by different detector pairs due to the eccentricity of the position of the barrel source so as to enable the integral value to be equivalent to that of all crystals to be uniformly irradiated. The detection efficiency distribution reflecting the intrinsic difference of the crystal efficiency can be obtained by correcting the eccentric barrel source scanning data.
Further, in the detection efficiency calculation process, valid data are utilized adaptively, and calculation errors caused by invalid data are eliminated. Valid data refers to the fact that the coincidence line formed by the detector pair passes through the barrel source; and invalid data, meaning that the conforming line did not pass through the bucket source.
Drawings
FIG. 1 is a schematic view of a sector formed by a crystal ui and an opposing set of crystals aligned in accordance with the present invention;
FIG. 2 is a schematic illustration of determining the sector area in the present invention;
FIG. 3 is a schematic diagram of the present invention illustrating the calculation of the minimum number of conforming crystals based on the actual position of the eccentric barrel source;
FIG. 4 is a schematic diagram of the present invention showing the calculation of the number of corresponding crystals for each crystal based on the actual position of the eccentric barrel source;
fig. 5 is a schematic diagram of a method for correcting the detection efficiency of a PET detector according to an embodiment of the present invention.
Detailed Description
For the purpose of better explaining the present invention and to facilitate understanding, the present invention will be described in detail by way of specific embodiments with reference to the accompanying drawings.
In order to better understand the scheme of the embodiment of the invention, the following outlines the scheme of the embodiment of the invention.
To better understand the improvement of the present invention, the following correction procedure for the detection efficiency of the existing PET detector is explained as follows.
In order to obtain an unbiased estimation of a radionuclide activity distribution image, the inconsistency of photon detection efficiency among different detection units needs to be corrected back by human, and the process is called normalization correction.
The common normalization correction methods include a direct measurement method and a detection efficiency modularization measurement method. The former needs a large number of counts to reduce statistical noise, and the acquisition time is long, so that the ideal statistical quantity is difficult to achieve. Currently, in clinical application, a detection efficiency modular measurement method is generally adopted. In the case of ensuring uniform illumination of all detector units, the count of the detector unit reflects the relative detection efficiency. In order to satisfy the precondition that all detector units are uniformly illuminatedEfficiency is typically measured using a bucket source centered in the Field-of-View (FOV). Wherein the barrel source is a barrel with a certain radius, an axial length covering the detector axial FOV, and a certain activity of radioactive nuclide (such as radionuclide containing 18 A liquid source of F or containing 68 A solid state source of G, or containing other radionuclides).
Experiments with detection efficiency correction factor measurements require that all detector crystals are illuminated uniformly, i.e. that the bucket source is placed centrally. In practice, to ensure this, it is often necessary for the operator to touch the radioactive source of the bucket multiple times to accurately adjust the position of the bucket source. This not only makes the experiment complicated and difficult to operate, but also requires the experimenter to be subjected to large doses of radiation.
During the detection efficiency measurement experiment, the centered placement of the bucket source is a requirement which is not easy to meet. In practice, the barrel source is generally placed approximately at the center of the cross-sectional FOV, and the position of the barrel source is finely adjusted by calculating the center of gravity of the barrel source through preliminary imaging. This process needs to be repeated several times to ensure that the bucket source is strictly centered in the FOV.
Example one
The invention provides a method for correcting the detection efficiency of a PET detector, which comprises the following steps as shown in figure 1.
A1, obtaining scanning data of the PET system after the PET system is placed on the barrel source.
It is understood that the coincidence Data acquired by the PET detector system is generally divided into List Mode Data (List Mode Data) and Sinogram Data (Sinogram Data), wherein the List Mode Data can be converted into Sinogram Data by Data Sorting (Sorting).
A2, determining the actual position information of the barrel source in the PET system according to the scanning data;
for example, the step A2 may include the following sub-steps not shown in the figure:
a21, based on the scanning data, adopting a selected reconstruction algorithm to obtain a reconstructed image of a bucket source;
and A22, acquiring the gravity center according to the reconstructed image, and obtaining the actual position information of the bucket source by using a gravity center method.
A3, acquiring barrel source related correction information and intermediate information for correcting detection efficiency according to the actual position information;
for example, the revision information associated with the bucket source includes one or more of:
correction factor SG representing the uneven detection efficiency caused by the influence of detector geometry on different LORs uivj ；
Correction factor FG representing the non-uniformity of detection efficiency due to the influence of the geometry of the phantom (i.e., the object to be detected) on different LORs uivj ；
The intermediate information for correcting the detection efficiency includes: number of crystals N ui 。
And A4, obtaining correction values of all crystal detection efficiencies according to the correction information, the intermediate information, the scanning data, the actual position information of the barrel source and the intensity of the radioactive nuclide in the barrel source.
That is, the correction of the detection efficiency of all crystals in the PET detector.
Because the detection efficiency of the prior art is correct, the centered placement of the bucket source is an unmet need. The method of the invention enables the detection efficiency measurement experiment to be free from strict and centered placement of the barrel source, greatly reduces the operation difficulty of the experiment, and simultaneously reduces the radiation risk of workers.
Example two
The most common structure of a PET detector is cylindrical, and a structure composed of crystals is integrally arranged on the surface of the cylinder. The detected object is placed on the bed body and goes deep into the cylinder to carry out data acquisition. Different crystals judge whether the detected photons come from the same case or not through electronic signals and record the photons. If two crystals detect two photons from the same instance, the instance must occur on the line connecting the two crystals. In the analysis, this Line is called a Line of Response (LOR). The coincidence data acquired by the PET detector system is typically classified as list mode data (L)ist Mode Data) and Sinogram Data (Sinogram Data), wherein the List Mode Data may be converted into Sinogram Data by Data Sorting (Sorting). The data format that the scanning data adopted in this application is Sinogram data. Each element size in the Sinogrm data represents the actual number of probe instances E for the corresponding LOR uivj 。
Number of actual detection cases E of LOR uivj Can be expressed as:
E uivj ＝A×SG uivj ×FG uivj ×ε ui ×ε vj ………………………………………………(2)
the detection efficiency of any one LOR can be expressed as the product of the two-terminal crystal detection efficiencies:
η uivj ＝ε ui ×ε vj ……………………………………………………(1)
wherein u, v represents the axial number of any two crystals, and i, j represents the circumferential number of the crystals.
In the above formula, a represents the radionuclide intensity; SG (steam generator) uivj Correction factors representing the uneven detection efficiency caused by the influence of the geometric structures of the detectors (scanners) on different LORs are derived from factors such as different incident angles of the different LORs to the crystal, and are generally obtained by measuring through a rotating line source experiment; FG (FG) uivj The non-uniform correction factor, which represents the detection efficiency due to the Phantom (Phantom) geometry affecting the different LORs, results from the non-uniform length of the different LORs through the Phantom (detected object), which is strictly dependent on the shape and position of the Phantom. Through SG uivj And FG uivj The corrected LOR distribution is equivalent to all crystals being uniformly illuminated, and it is the intrinsic difference in detector crystal efficiency that is reflected. By analyzing the distribution, the intrinsic detection efficiency of the detector unit can be obtained.
As shown in FIG. 1, a detector crystal-ui is arbitrarily selected, a group of crystals A is selected opposite to the circumferential direction of the detector crystal-ui, and ui coincides with any crystal vj in A. The detection efficiency of any crystal ui can be expressed as:
combining the formula (1) to obtain
The above formula holds if the sum of the face crystal efficiencies (the crystals included in Group A in FIG. 1) is approximately equal to the number of crystals, i.e., the
Wherein N is the number of crystals corresponding to the crystal ui directly opposite thereto (GroupA in FIG. 1). In the present invention, N is defined as the number of coincidence crystals in the detection efficiency calculation process. The more crystals the Group A contains, the more intrinsic differences of the detection efficiency of different crystals will be compensated, and the closer the average value is to 1, i.e. the sum is close to the total number of crystals, the closer the above formula is to the same. Theoretically, as many LORs as possible should be utilized to ensure that equation (5) holds, and statistics can be increased to reduce noise.
However, in the actual normalization experiment operation, the effective length of the source passing through the barrel is short, the count is small, and the statistical fluctuation is large through the LOR of the edge of the source. To avoid the effect of the LOR passing through the bucket source edge on the result, the LOR passing through the bucket source edge is generally removed in the calculation of the detection efficiency during the experiment. Specifically, as shown in FIG. 2, only LORs that pass near the center of the bucket source are selected.
Given strict requirements for centered placement of bucket sources, FG is typically given in advance uivj The number of coincident crystals in the course of the distribution and detection efficiency calculation. However, when the bucket source is not centered as intended, the geometry factor FG associated with the bucket source is first made uivj Correcting errors, and secondly, causing some crystal efficiencies to be calculated, invalid LORs are introduced. Both of the above two points will cause the error of the detection efficiency calculation.
The method of the invention overcomes the limitation of the existing algorithm for the centered placement of the bucket source. Firstly, obtaining a reconstructed image of a bucket source by using scanning data of a detector, and further calculating the actual position of the bucket source by a gravity center method; secondly, obtaining a geometric factor correction factor related to the bucket source according to the actual position of the bucket source, and correcting the scanning data; and finally, according to the actual position of the barrel source, obtaining the number of the conforming crystals in the detection efficiency calculation process under the condition of ensuring that the researched LOR effectively passes through the barrel source.
For the multi-modal scanning mode, the position of the bucket source can also be obtained through the image distribution given by other scanning modes.
1)Obtaining the true center of the barrel source by using the gravity center method
Step one, obtaining a reconstructed image of the bucket source based on a selected reconstruction algorithm according to the scanning data.
For example, in practice, the scan data may be processed using a PET system integrated Ordered-Subset-Expectation-Maximization (OSEM) algorithm to obtain a reconstructed image of the bucket source.
In other embodiments, the method for reconstructing the image by using the OSEM may not be limited, and may be selected according to actual needs, and any method for reconstructing the image may be used.
It can be understood that the reconstructed image in the first step can be, for example, a reconstructed image obtained without the assistance of other imaging systems in one implementation. The other is a reconstructed image obtained in the presence of the assistance of other imaging systems.
And step two, calculating the gravity center of the bucket source based on the reconstructed image. The acquisition paths of the image center of gravity are slightly different for different scan modes.
And step three, determining the position information of the barrel source in the PET system according to the gravity center of the barrel source.
1.1 ) acquire reconstructed images without assistance from other imaging systems
When the experiment is without the assistance of other imaging systems, the attenuation coefficient distribution of the bucket source is unknown. For this purpose, the attenuation coefficient distribution needs to be obtained by using the acquired data, i.e., the scan data in the present embodiment. The method comprises the following specific steps:
step 1: and reconstructing the acquired data, namely the scanning data, without attenuation correction to obtain a PET activity distribution image I.
And 2, step: calculating the average value of pixels in the PET activity distribution image I, setting a proper threshold value (such as 0.5 time of the average value) according to the value, and endowing the pixels larger than the threshold value with an attenuation coefficient corresponding to the barrel source material according to the prior knowledge, such as the attenuation coefficient of liquid water is 0.096. And the pixels smaller than the threshold are set to 0, and the first attenuation coefficient distribution is obtained.
And 3, step 3: and (3) reconstructing the acquired data by using the attenuation coefficient distribution pair obtained in the step (2) and having attenuation correction to obtain a PET activity distribution image II. And (5) repeating the step (2) to obtain the second attenuation coefficient distribution.
And successively iterating, obtaining an output image of the PET activity distribution with accurate attenuation correction (namely, the reconstructed image of the PET reflects the activity distribution map of the detected object, and the activity distribution map of the detected object is consistent with the activity distribution map of the PET). The number of iterations depends on the performance of the PET detector, the size of the activity of the radionuclide, the acquisition time, etc.
In other embodiments, the reconstructed image of the PET system may also be assisted by other imaging systems, such as MR modality imaging.
When the experiment is assisted by other imaging systems and a high-precision structural image can be provided for PET image reconstruction, the attenuation coefficient distribution can be accurately defined, and the step 3 can be directly carried out to obtain an output image subjected to accurate attenuation correction. The method for reconstructing the image assisted by the imaging of other modalities is not described in detail in the present application. The following exemplary embodiments are described without the aid of other imaging systems.
1.2 For step two above, calculate the center of gravity of the bucket source, as described in detail below.
By analyzing the output image, the true position coordinates of the bucket source are obtained in two steps. The method comprises the following specific steps:
step 01: the center of gravity of the whole image is calculated as a research object. The specific calculation formula is as follows:
wherein, pixel i Is the value of an arbitrary pixel i, x i And y i Are the x and y coordinates of the corresponding pixel. All pixels are taken as research objects, and the gravity center x 'of the barrel source can be obtained' c And y' c 。
Step 02: due to calculation of the center of gravity x' c And y' c All pixels are utilized and there may be a surrounding background effect on the result. And in addition to step 01, with (x' c ，y’ c ) A drum range with a radius slightly larger than the radius of the drum source is selected as the center of a circle. The selected drum range can be adjusted according to actual conditions. The true center of gravity x of the bucket source can be found by substituting the values of all pixels in the selected region into equation (6) c And y c . The eccentric distance of the barrel source can pass through x c And y c Obtaining the eccentric distance of the barrel source:
2) Calculating a bucket source dependent geometric correction factor from a true position of the bucket source
As shown in fig. 2, different LORs travel different lengths in the bucket source, i.e., different total numbers of photons occurring on the LOR path. Through modeling of a PET system, integral values of the radioactivity activity distribution of the barrel source on different LORs can be obtained, namely, a geometric correction factor FG related to the barrel source in acquired data uivj . SG obtained by joint preamble experiment uivj Using equation (2), the corrected distribution of the detection efficiency equivalent to all LORs under the condition that all crystals are uniformly irradiated can be obtained. By analyzing the distribution, a relative detection of the detector units can be obtainedEfficiency.
So-called SG uivj It refers to the relative difference in the efficiency of detection of photons within different LORs due to the different angles of incidence on the respective crystals, the different positions of the crystals in the detector, and so on. Typically by other experiments such as a rotating line source. It is customarily referred to as the geometric correction factor of the detector, i.e. SG uivj 。
So-called FG uivj Obtained by system modeling. FG (FG) uivj Which refers to the probability of detection by each LOR at different locations within the FOV, is determined by the size of the two detector field angles corresponding to the LORs at different points within the FOV. The calculation process is complicated, and the calculation is performed in advance, stored in a certain format and directly called clinically. Conventionally, this is referred to as system modeling. When the detected objects are different, the probability of being detected by each LOR at different positions in the FOV is multiplied by the radioactivity at the position, and the integral value of the detected object on each LOR can be obtained.
3) Calculating the number of conforming crystals in detection efficiency calculation from the true position of the bucket source
Generally, to obtain the detection efficiency, the LOR of any crystal and its opposite crystal is selected and studied. However, when the detected object is placed eccentrically, some of the LORs selected previously may be caused to become invalid LORs, resulting in erroneous calculation results. To ensure that the LOR of all detector units and their corresponding crystals passes through the bucket source, the number of corresponding crystals should be determined by the actual location of the physical object being detected.
In the embodiment of the invention, two methods for finding the number of the crystals in accordance with the detected physical actual position in a self-adaptive mode are provided.
Firstly, selecting a barrel source part in an LOR formed by any crystal and the crystal opposite to the crystal as a research object, namely the LOR to be researched is symmetrical about the diameter of a detector;
second, all the LORs that pass through the bucket source are selected as study objects, i.e., the LORs under study are symmetric about the bucket source diameter.
For each situation, the effective LOR can be determined through system modeling on the basis of the actual position of the barrel source to obtain the number of the conforming crystals, and the number of the conforming crystals can also be directly calculated through the geometric relationship between the actual position of the barrel source and the detector.
3.1 By selecting the number of crystals corresponding to the diameter of the detector as the axis of symmetry
System modeling
To adaptively obtain the number of coincident crystals used in the calculation of detection efficiency, any crystal ui is selected, the actual position of the barrel source is used, modeling is performed by the system described above, and the number of crystals coincident with the crystal located directly opposite the crystal is estimated, denoted as N, while ensuring that the LOR passes through the barrel source ui 。
To ensure that the selected set of corresponding crystals is located directly opposite crystal ui, crystal vj is first selected by ensuring vj- (N) ui -1)/2 to vj + (N) ui The LOR composed of the crystals and the crystals ui between-1)/2 passes through the barrel source to obtain N ui . And traversing all crystals to obtain the number of corresponding crystals corresponding to all the crystals.
Considering the problem of relatively small counts of LOR's passing through the bucket source edge and large statistical fluctuations, the crystals associated with LOR's passing through the bucket source edge are removed. This value can be optimized according to specific experimental data. The final selected value is used as the number N of corresponding crystals corresponding to each crystal in the detection efficiency calculation process ui 。
Geometric relationship
As shown in fig. 3, for any crystal ui, at point a. The spatial coordinates of which are known. The distance between the center of gravity of the detected object and the center of gravity of the detected object, which is obtained through the previous calculation, at the point O ', is calculated, namely the length of | O' A | in FIG. 3 is denoted as d. The FOV has a center O. In the triangle OO' a, three sides are known, and the solving process of the angle is a conventional formula and is not stated herein. AB is the tangent line of the barrel source connecting the crystal under study and near the center O of the FOV, and the tangent point is E. B' is the symmetry point of B with respect to C. Wherein
Wherein R is cyl Is the radius of the barrel source, transaxialcrytal is the number of annular crystals of the detector, N ui I.e. the number of crystals included in Group a as shown in fig. 3. And traversing all crystals to obtain the corresponding number of the corresponding crystals of all the crystals.
Considering the problem of relatively small counts of LOR's passing through the bucket source edge and large statistical fluctuations, the crystals associated with LOR's passing through the bucket source edge are removed. This value can be optimized according to specific experimental data. The final selected value is used as the number N of corresponding crystals corresponding to each crystal in the detection efficiency calculation process ui 。
3.2 By selecting the number of crystals with the barrel source diameter as the axis of symmetry
In the detection efficiency calculation, the algorithm of selecting the number of coincident crystals by taking the diameter of the detector as a symmetry axis is a conventional algorithm, but when the detected physical eccentricity is too much, if only the crystals are considered to be coincident with the crystal set opposite to the crystals, some crystals cannot find the effective number of coincident crystals. To account for this, we choose the number of conforming crystals with the barrel source diameter as the axis of symmetry.
System modeling
In order to adaptively obtain the number of corresponding crystals in the detection efficiency calculation process, selecting any crystal ui, utilizing the actual position of a barrel source, modeling through the system, and evaluating the number of crystals corresponding to the crystal on the premise of ensuring that the LOR passes through the barrel source, wherein the number of the corresponding crystals is recorded as N ui . And traversing all crystals to obtain the number of corresponding crystals corresponding to all the crystals.
Considering the problem of relatively small counts of LOR's passing through the bucket source edge and large statistical fluctuations, the crystals associated with LOR's passing through the bucket source edge are removed. This value can be optimized according to specific experimental data. The final selected value is used as the number N of corresponding crystals corresponding to each crystal in the detection efficiency calculation process ui 。
Geometric relationship
In fig. 4, for any crystal ui, the point a is located. The spatial coordinates of which are known. The distance between the center of gravity of the detected object and the center of gravity of the detected object, which is obtained through the previous calculation, at the point O ', is calculated, namely the length of | O' A | in FIG. 4 is denoted as d. The center of the FOV is O. All LORs formed coincident with the crystal ui pass through the portion of the object being detected, as shown between AB and AB' in the figure. The number of crystals contained in Group A is proportional to the angle ≈ BOB'. Let < BOB' = θ. By using the geometric relation in the graph, the < BOB '=2 < BAB' can be obtained.
And then to
In accordance with the previous discussion, R cyl The radius of the barrel source, transaxialcrytal is the number of annular crystals of the detector, N ui I.e. the number of crystals included in Group a as shown in fig. 4. And traversing all crystals to obtain the number of corresponding crystals corresponding to all the crystals.
Considering the problem of relatively small counts of LOR's passing through the source edge of the bucket, with large statistical fluctuations, the crystals associated with LOR's passing through the source edge of the bucket are removed. This value can be optimized according to specific experimental data. The final selected value is used as the number N of corresponding crystals corresponding to each crystal in the detection efficiency calculation process ui 。
4) Calculating detection efficiency using coincidence crystal number
Under the condition that the eccentric amplitude of the barrel source is not large, different crystals generally select the same number of conforming crystal numbers in order to ensure that the statistical fluctuation in the detection efficiency calculation of different crystals is consistent. I.e. the number of coincident crystals N per crystal obtained before ui The minimum value is selected as the number N of the coincidence crystals in the calculation of the detection efficiency. With the formula (4), the detection efficiency can be obtained.
However, when the detected object is eccentric too much, the difference between the numbers of coincidence crystals of different crystals is large. If the uniform number of crystals is forcibly required, data is wasted to a certain extent. In order to utilize the data to the maximum, the previously obtained number of crystals per crystal is substituted into equation (4). Accordingly, equation (4) becomes the following form:
and traversing all crystals to obtain the detection efficiency.
The method reduces the complexity of the operation of workers, and does not need to manually and finely adjust the position of the bucket source again and again in order to strictly center the bucket source; and secondly, the inclusion of the experimental analysis method reduces the condition that the experiment needs to be carried out again because the bucket source is not well centered.
The invention greatly reduces the complexity of the operation of the detection efficiency measurement experiment, can effectively reduce the radiation of workers in the experiment and effectively reduce the experiment cost.
It should be understood that the above description of the specific embodiments of the present invention is provided for illustration only, and the purpose of the present invention is to enable those skilled in the art to understand the content of the present invention and to implement the present invention, but the present invention is not limited to the specific embodiments described above. It is intended to cover in the appended claims all such changes and modifications that are within the scope of this invention.
Claims (7)
1. A method of correcting for detection efficiency of a PET detector, comprising:
a1, acquiring scanning data of a PET system after a barrel source is placed;
a2, determining the actual position information of the barrel source in the PET system according to the scanning data;
a3, according to the actual position information, acquiring the barrelSource-related revision information and intermediate information for correcting detection efficiency; the revision information associated with the bucket source includes: correction factor SG representing the uneven detection efficiency caused by the influence of detector geometry on different LORs uivj ；SG uivj Is obtained in advance through experiments of a rotating line source;
correction factor FG representing the non-uniformity of detection efficiency due to the influence of the geometry of the phantom, i.e., the object to be detected, on different LORs uivj ；FG uivj The probability of being detected by each LOR at different positions in the FOV is determined by the field angles of two detectors corresponding to the LORs of different points in the FOV and is obtained based on system modeling of a detected object;
a4, obtaining correction values of all crystal detection efficiencies according to the correction information, the intermediate information, the scanning data, the actual position information of the barrel source and the intensity of the radioactive nuclide in the barrel source;
a41, actual detecting the number of cases E according to each LOR in the scanning data uivj The intensity of the radionuclide in the barrel source and the correction information are obtained based on the following formula I uivj ；
Wherein E is uivj The actual detection case number u, v represents the axial number of any two crystals, and i, j represents the annular number of the crystals;
a represents radionuclide intensity, SG uivj A correction factor representing a corresponding detector geometry affecting detection efficiency;
FG uivj a correction factor representing a corresponding bucket source geometry that affects detection efficiency;
a42, acquiring the corrected detection efficiency according to a formula II and a formula III;
wherein, N ui Represents the number of coincident crystals;
ε ui indicates the detection efficiency, ∈, of the crystal ui vj Indicating the detection efficiency of the crystal vj.
2. The method according to claim 1, wherein the step A2 comprises:
a21, based on the scanning data, adopting a selected reconstruction algorithm to obtain a reconstructed image of a bucket source;
and A22, acquiring the gravity center according to the reconstructed image, and obtaining the actual position information of the bucket source by using a gravity center method.
3. The method according to claim 2, wherein the step a21 comprises:
a21-1, carrying out non-attenuation correction reconstruction on the scanning data to obtain a PET activity distribution image I;
a21-2, calculating the average value of all pixels in the first PET activity distribution image, selecting a preset threshold value matched with the average value, and selecting pixels larger than the preset threshold value in the first PET activity distribution image;
giving attenuation coefficients corresponding to the selected pixels and the barrel source substances according to the priori knowledge to obtain first attenuation coefficient distribution;
a21-3, reconstructing the scanning data with attenuation correction by using the attenuation coefficient distribution to obtain a PET activity distribution image II;
repeating the process of the step A21-2 and the step A21-3 for P times to obtain a PET activity distribution output image subjected to accurate attenuation correction;
wherein the iteration P is determined according to the performance of the PET detector, the activity size of the radionuclide in the barrel source and/or the acquisition time.
4. The method according to claim 2, wherein said step a21 comprises:
a211, acquiring acquired data of the barrel source according to other imaging systems, and acquiring attenuation coefficient distribution corresponding to a reconstructed PET activity distribution output image;
and A212, reconstructing the scanning data with attenuation correction by using the attenuation coefficient distribution to obtain a PET activity distribution output image.
5. The method of claim 1,
the intermediate information for correcting the detection efficiency includes: number of crystals N ui 。
6. The method according to claim 5, wherein the obtaining of the intermediate information for correcting the detection efficiency in step A3 comprises:
a3-1, selecting LORs which pass through a barrel source part from LORs formed by any crystal and the crystal opposite to the crystal based on actual position information of the barrel source, wherein the LORs which pass through the barrel source part are LORs which are symmetrical with the diameter of a detector;
a3-2, determining effective LOR according to the selected LOR and the actual position information of the barrel source to obtain the number N of crystals ui ；
Alternatively, the first and second electrodes may be,
a31, based on the actual position information of the bucket source, selecting all LORs which pass through the bucket source and are symmetrical about the diameter of the bucket source;
a32, determining effective LOR according to the selected LOR and the actual position information of the barrel source to obtain the number N of crystals ui 。
7. A PET system comprising PET detectors which perform the method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910735219.9A CN110368009B (en) | 2019-08-09 | 2019-08-09 | Method for correcting detection efficiency of PET detector |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910735219.9A CN110368009B (en) | 2019-08-09 | 2019-08-09 | Method for correcting detection efficiency of PET detector |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110368009A CN110368009A (en) | 2019-10-25 |
| CN110368009B true CN110368009B (en) | 2023-03-28 |
Family
ID=68258758
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910735219.9A Active CN110368009B (en) | 2019-08-09 | 2019-08-09 | Method for correcting detection efficiency of PET detector |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110368009B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109077748B (en) * | 2018-08-31 | 2021-08-20 | 北京锐视康科技发展有限公司 | Accurate PET normalization correction method |
| CN111557675A (en) * | 2020-05-12 | 2020-08-21 | 上海联影医疗科技有限公司 | Correction method, device and system of medical imaging device |
| CN112807009A (en) * | 2021-02-02 | 2021-05-18 | 上海联影医疗科技股份有限公司 | Radioactive source positioning method, system, equipment, electronic device and storage medium |
| CN113069138B (en) * | 2021-03-23 | 2023-06-30 | 上海联影医疗科技股份有限公司 | Positron emission tomography device, coincidence efficiency detection method and normalization method |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7718954B2 (en) * | 2006-01-31 | 2010-05-18 | Koninklijke Philips Electronics N.V. | Component method and system for PET detector efficiency normalization |
| CN105894525B (en) * | 2016-04-29 | 2019-03-05 | 上海联影医疗科技有限公司 | Positron emission tomography detector correction method |
| CN109077748B (en) * | 2018-08-31 | 2021-08-20 | 北京锐视康科技发展有限公司 | Accurate PET normalization correction method |
-
2019
- 2019-08-09 CN CN201910735219.9A patent/CN110368009B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN110368009A (en) | 2019-10-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110368009B (en) | Method for correcting detection efficiency of PET detector | |
| JP4414410B2 (en) | Image reconstruction method | |
| US8098916B2 (en) | System and method for image-based attenuation correction of PET/SPECT images | |
| US8150112B2 (en) | Regional reconstruction of spatially distributed functions | |
| Tong et al. | Image reconstruction for PET/CT scanners: past achievements and future challenges | |
| Holly et al. | Single photon-emission computed tomography | |
| EP3224801B1 (en) | Multi-modality imaging system and method | |
| US8017914B2 (en) | Method and system for scatter correction | |
| US7324624B2 (en) | Shifted transmission mock for nuclear medical imaging | |
| US20100193697A1 (en) | Apparatus and methods for evaluating operation of pixelated detectors | |
| US20060266946A1 (en) | Fourier re-binning of time-of- flight positron emission tomography data | |
| US8415630B2 (en) | Apparatus and methods for determining a boundary of an object for positron emission tomography scatter correction | |
| US9684973B2 (en) | Systems and methods for selecting imaging data for principle components analysis | |
| JP2009528139A (en) | Local motion compensation based on list mode data | |
| EP1891899A1 (en) | Method and system for performing local tomography | |
| CN105125231A (en) | Method and device for eliminating positron emission tomography (PET) image ring artifacts | |
| EP2080166A2 (en) | Dirty isotope pet reconstruction | |
| EP1631844B1 (en) | Generating detector efficiency estimates for a oet scanner | |
| Perez et al. | Characterizing the contribution of cardiac and hepatic uptake in dedicated breast SPECT using tilted trajectories | |
| WO2023051719A1 (en) | Methods and systems for attenuation map generation | |
| Cuddy‐Walsh et al. | Patient‐specific estimation of spatially variant image noise for a pinhole cardiac SPECT camera | |
| CN108932740A (en) | A kind of normalization factor acquisition methods and medical imaging procedure | |
| CN108542409B (en) | Method for measuring noise equivalent counting rate of double-panel PET system | |
| JP3770914B2 (en) | Method of retrospectively determining the center of rotation of a scintillation camera detector from SPECT data obtained during a nuclear medicine examination | |
| Protonotarios et al. | The attenuated spline reconstruction technique for single |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |