Dead time correction and image reconstruction method, system and device and storage medium

Abstract

The invention discloses a dead time correction and image reconstruction method, system and device and a storage medium. The invention discloses a dead time correction method, system and device for positron emission computed tomography and a storage medium, and relates to the field of data processing. The dead time correction method comprises the following steps: acquiring a random coincidence event counting rate of each response line subset; traversing all the random coincidence event counting rates, and judging whether the random coincidence event counting rates are smaller than a minimum value of a random coincidence event counting rate range covered by a preset interpolation table one by one; if yes, assigning a dead time correction factor of a response line subset corresponding to the current random coincidence event counting rate as a preset value; and if not, obtaining a dead time correction factor of the response line subset corresponding to the current random coincidence event counting rate through a preset method. According to the dead time correction method, the dead time correction factors are associated with the response line subsets through division of the response line subsets, the more accurate dead time correction factors can be obtained, and better universality is achieved.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a dead time correction method, system, device and storage medium for positron emission tomography. Background Art

[0002] Positron emission tomography (PET) technology is one of the most advanced molecular imaging technologies in the world. It has many functions such as early cancer detection, efficacy evaluation, and pharmacokinetic observation. It can non-invasively, quantitatively, and dynamically evaluate the metabolic levels, biochemical reactions, and functional activities of various functional organs in the body by imaging compounds labeled with radionuclides in the body. It has high sensitivity and accuracy and plays an irreplaceable role in clinical medicine and drug development.

[0003] PET equipment includes detectors and back-end servers, etc. The detectors include scintillation crystals, photoelectric conversion devices and electronic readout devices. The gamma rays generated by the decay of radioactive drugs in the body to be tested are converted into light signals by scintillation crystals, and then converted into electrical signals by photoelectric conversion devices. The position (Position), energy (Energy), and time (Time) information of the gamma rays are accurately extracted in a digital way, and the digitized information is sent to the back-end server. The original waveform is restored by fitting, key information is extracted and processed, and used for PET image reconstruction to obtain the image of the body to be tested.

[0004] In an ideal data processing process, the true coincidence event count rate should change linearly with the activity of the radioactive source. However, in reality, due to the influence of detector performance, pulse stacking effect, coincidence strategy, etc., the count of true coincidence events will be lost to a certain extent. The higher the activity of the radioactive source, the more events will be lost. This phenomenon is called the dead time effect. Through practice, it is found that the dead time effect affects the system count rate. Therefore, it is very important to correctly correct the dead time effect to improve the system count rate performance.

[0005] The key to dead time correction is to obtain the count rate loss coefficient on the line of response (LOR) formed by each pair of detectors. At present, some studies use the global dead time correction method, which uses the global count rate to estimate the average count rate loss coefficient of all LORs; some studies use model-based methods to fit the parameters in the model with the collected data, and then use the obtained model to correct the dead time effect.

[0006] The global dead time correction method uses the average value of all LOR correction factors to correct the dead time effect, which cannot take into account the different dead time correction factors of detectors at different locations. However, due to the influence of the activity distribution in the field of view (FOV) or the radiation source outside the FOV, detectors at different locations may have different random events. At the same time, due to the limitation of detector performance, the distribution of true events is different, which leads to different dead time correction factors for detectors at different locations.

[0007] The model fitting method performs dead time effect correction based on the existing model. Its advantage is that the initial activity during acquisition does not need to be very high. Its disadvantage is that it is not universal. Due to different system structures or different system detection performances, the models corresponding to different systems may need to be adjusted. In addition, due to different conditions during acquisition, such as the material and shape of the model, the distribution of radiation sources, etc., the model fitted by the same system during one acquisition may not be suitable for correction during the next reconstruction.

[0008] The content of the background technology description is only for facilitating understanding of the relevant technology in this field and is not regarded as an admission of the prior art. Summary of the invention

[0009] Therefore, the present application intends to provide a dead time correction method, system, device and storage medium for positron emission tomography, which can solve at least one problem existing in the prior art.

[0010] In a first aspect, a dead time correction method is provided, comprising: obtaining a random coincidence event count rate of each response line subset; traversing all random coincidence event count rates, and determining one by one whether the random coincidence event count rate is less than a minimum value of a random coincidence event count rate range covered by a preset interpolation table; if so, specifying a dead time correction factor of the response line subset corresponding to the current random coincidence event count rate as a preset value; if not, obtaining the dead time correction factor of the response line subset corresponding to the current random coincidence event count rate by a preset method.

[0011] In an embodiment of the present application, obtaining the random coincidence event count rate of each response line subset includes: obtaining detection data of a positron emission tomography device, performing coincidence according to the detection data, and obtaining the random coincidence event count rate of each response line subset according to the coincidence result.

[0012] In an embodiment of the present application, performing conformity according to the detection data and obtaining the random conformity event count rate of each of the response line subsets according to the conformity result includes: performing conformity according to the detection data to obtain a delayed conformity data sinusoidal graph; and calculating the random conformity event count rate of each of the response line subsets according to the delayed conformity data sinusoidal graph and the acquisition time of the corresponding response line subset.

[0013] In the embodiment of the present application, the random coincidence event count rate of each response line subset is calculated according to the delayed coincidence data sinusoidal diagram and the acquisition time of the corresponding response line subset, including: calculating the random coincidence event count rate of the i-th response line subset according to the following formula (1):

[0014]

[0015] Where T is the acquisition duration, D k is the count on the kth response line in the delay conformance data sinusoid.

[0016] In an embodiment of the present application, before acquiring the random coincidence event count rate of each response line subset, the dead time correction method further includes: grouping all response lines according to position characteristics of the response lines to obtain a plurality of response line subsets.

[0017] In the embodiment of the present application, the method of grouping all the response lines according to the position characteristics of the response lines to obtain a plurality of response line subsets includes: × A ring formed by an array of m crystals is defined as a modular ring. When the crystals at both ends of response line a and the crystals at both ends of response line b belong to the same modular ring, response line a and response line b are divided into the same response line subset, wherein n and m are both positive integers, and at least one of n or m is greater than 2.

[0018] In an embodiment of the present application, the preset interpolation table can be obtained by the following steps: collecting detection experimental data of the radioactive source at different activities under experimental conditions, and performing data matching to obtain response lines; dividing all response lines into several response line subsets, and obtaining the random coincidence event count rate of each response line subset at each activity; obtaining the instantaneous coincidence event count rate of each response line subset at each activity; obtaining the true coincidence event detection count rate of each response line subset at each activity based on the instantaneous coincidence event count rate and the random coincidence event count rate; linearly extrapolating the true coincidence event detection count rate at the preset activity to obtain the ideal true coincidence event count rate of each response line subset at each activity; obtaining the dead time correction factor of each response line subset at each activity based on the corresponding true coincidence event detection count rate and the ideal true coincidence event count rate, and the dead time correction factor and the corresponding random coincidence event count rate constitute the preset interpolation table in a one-to-one correspondence.

[0019] In an embodiment of the present application, the collecting of detection experimental data of a radioactive source at different activities under experimental conditions includes: collecting detection experimental data of the same radioactive source multiple times within a predetermined time, and factors affecting the predetermined time include the initial activity and decay rate of the radioactive source.

[0020] In the embodiment of the present application, the true coincidence event detection count rate under the preset activity is regarded as the true coincidence event ideal count rate.

[0021] In the embodiment of the present application, the instant coincidence event count rate of each response line subset at each activity level is calculated, including: calculating the instant coincidence event count rate of the i-th response line subset at the j-th scan by the following formula (2):

[0022]

[0023] Among them, P kj is the number of instant coincidence events of the kth response line in the i-th response line subset at the jth scan, and T is the acquisition duration.

[0024] In the embodiment of the present application, the true coincidence event detection count rate of each response line subset at each activity is obtained according to the instant coincidence event count rate and the random coincidence event count rate, including: the true coincidence event detection count rate is calculated by the following formula (3):

[0025]

[0026] in, represents the instantaneous coincidence event count rate of the i-th response line subset at the j-th scan, represents the random coincidence event count rate of the i-th response line subset at the j-th scan.

[0027] In an embodiment of the present application, the linear extrapolation of the true coincidence event detection count rate under a preset activity to obtain the ideal true coincidence event count rate of each response line subset under each activity includes: taking the preset activity as the independent variable and the true coincidence event detection count rate as the dependent variable, performing least squares straight line fitting, and linearly extrapolating to the ideal true coincidence event count rate of each response line subset when the activity is greater than the preset activity, wherein the true coincidence event detection count rate of each response line subset at the preset activity is equivalent to its ideal true coincidence event count rate.

[0028] In the embodiment of the present application, the preset activity includes several activities less than a specific value, wherein the specific value A j It is calculated by the following formula (4):

[0029]

[0030] Among them, A 0 is the initial activity, T 1 / 2 is the half-life of the radioactive source, T j The time from the initial activity measurement to the start of the jth scan, T acq,j is the duration of the j-th scan.

[0031] In the embodiment of the present application, the dead time correction factor of each response line subset at each activity is obtained according to the corresponding true coincidence event detection count rate and the true coincidence event ideal count rate, including: the dead time correction factor is obtained by using the following formula (5):

[0032]

[0033] in, is the ideal count rate of true coincidence events of the i-th response line subset at the j-th scan, is the true coincidence event detection count rate of the i-th response line subset at the j-th scan.

[0034] In an embodiment of the present application, the dead time correction factor of the response line subset corresponding to the current random coincidence event count rate is obtained by a preset method, including: when the current random coincidence event count rate is within the random coincidence event count rate range covered by the preset interpolation table, determining whether there is a value in the preset interpolation table that is the same as the current random coincidence event count rate; if so, using the dead time correction factor corresponding to the value as the dead time correction factor of the corresponding response line subset; if not, obtaining the two values ​​R in the preset interpolation table that are closest to the current random coincidence event count rate. 1 and R2 and its corresponding dead time correction factor CF 1 and CF 2 , and the dead time correction factor CF of the corresponding response line subset is calculated using the following formula (6): DT :

[0035]

[0036] Among them, R ob is the current random coincidence event count rate.

[0037] In an embodiment of the present application, the dead time correction factor of the response line subset corresponding to the current random coincidence event count rate is obtained by a preset method, including: when the current random coincidence event count rate is greater than the random coincidence event count rate range covered by the preset interpolation table, obtaining the random coincidence event count rate R of the first acquisition and the second acquisition in the preset interpolation table 3 and R 4 and its corresponding dead time correction factor CF 3 and CF 4 , and the dead time correction factor CF of the corresponding response line subset is calculated using the following formula (7): DT :

[0038]

[0039] Among them, R ob is the current random coincidence event count rate.

[0040] In the embodiments of the present application,

[0041] The method of grouping all response lines according to the position characteristics of the response lines to obtain several response line subsets includes: taking a crystal ring composed of a single axial crystal as a grouping unit, and when the crystals at both ends of the response line are located in the same crystal ring, they are grouped into the same response line subset.

[0042] In the embodiment of the present application, all response lines are grouped according to the position characteristics of the response lines to obtain several response line subsets, including:

[0043] When the crystals are arranged in a crystal array, the response lines from the same two crystal arrays are grouped together.

[0044] In the embodiment of the present application, the radiation source includes 18 F. 22 Na, 68 At least one of Ge.

[0045] In a second aspect, a dead time correction system is provided, comprising: an acquisition unit for acquiring a random coincidence event count rate of each response line subset; a judgment unit configured to traverse all random coincidence event count rates and judge one by one whether the random coincidence event count rate is less than a minimum value of a random coincidence event count rate range covered by a preset interpolation table; a processing unit configured to specify a dead time correction factor of the response line subset corresponding to the current random coincidence event count rate as a preset value when the current random coincidence event count rate is less than the minimum value of the random coincidence event count rate range covered by the preset interpolation table, and to acquire the dead time correction factor of the response line subset corresponding to the current random coincidence event count rate by a preset method when the current random coincidence event count rate is greater than or equal to the minimum value of the random coincidence event count rate range covered by the preset interpolation table.

[0046] In an embodiment of the present application, the acquisition unit is configured to: acquire detection data of a positron emission tomography device, perform coincidence according to the detection data, and obtain a random coincidence event count rate of each response line subset according to the coincidence result.

[0047] In an embodiment of the present application, the acquisition unit is configured to: obtain a delayed coincidence data sinusoidal graph according to the detection data; and calculate the random coincidence event count rate of each response line subset according to the delayed coincidence data sinusoidal graph and the acquisition time of the corresponding response line subset.

[0048] In the embodiment of the present application, the acquisition unit calculates the random coincidence event count rate of the i-th response line subset according to the following formula (1):

[0049]

[0050] Where T is the acquisition duration, D k is the count on the kth response line in the delay conformance data sinusoid.

[0051] In the embodiment of the present application, the acquisition unit is further configured to group all response lines according to position characteristics of the response lines to obtain a plurality of response line subsets before acquiring the random coincidence event count rate of each response line subset.

[0052] In the embodiment of the present application, the acquisition unit is configured to obtain n × A ring formed by an array of m crystals is defined as a modular ring. When the crystals at both ends of response line a and the crystals at both ends of response line b belong to the same modular ring, response line a and response line b are divided into the same response line subset, wherein n and m are both positive integers, and at least one of n or m is greater than 2.

[0053] In an embodiment of the present application, the preset interpolation table can be obtained by the following steps: collecting detection experimental data of the radioactive source at different activities under experimental conditions, and performing data matching to obtain response lines; dividing all response lines into several response line subsets, and obtaining the random coincidence event count rate of each response line subset at each activity; obtaining the instantaneous coincidence event count rate of each response line subset at each activity; obtaining the true coincidence event detection count rate of each response line subset at each activity based on the instantaneous coincidence event count rate and the random coincidence event count rate; linearly extrapolating the true coincidence event detection count rate at the preset activity to obtain the ideal true coincidence event count rate of each response line subset at each activity; obtaining the dead time correction factor of each response line subset at each activity based on the corresponding true coincidence event detection count rate and the ideal true coincidence event count rate, and the dead time correction factor and the corresponding random coincidence event count rate constitute the preset interpolation table in a one-to-one correspondence.

[0054] In the embodiment of the present application, the step of collecting experimental detection data of a radioactive source at different activities under experimental conditions includes: collecting detection experimental data of the same radioactive source multiple times within a predetermined time, and the factors affecting the predetermined time include the initial activity and decay rate of the radioactive source.

[0055] In the embodiment of the present application, the true coincidence event detection count rate under the preset activity is regarded as the true coincidence event ideal count rate.

[0056] In the embodiment of the present application, the instant coincidence event count rate of each response line subset at each activity level is calculated, including: calculating the instant coincidence event count rate of the i-th response line subset at the j-th scan by the following formula (2):

[0057]

[0058] Among them, P kj is the number of instant coincidence events of the kth response line in the i-th response line subset at the jth scan, and T is the acquisition duration.

[0059] In the embodiment of the present application, the true coincidence event detection count rate of each response line subset at each activity is obtained according to the instant coincidence event count rate and the random coincidence event count rate, including: the true coincidence event detection count rate is calculated by the following formula (3):

[0060]

[0061] in, represents the instantaneous coincidence event count rate of the i-th response line subset at the j-th scan, represents the random coincidence event count rate of the i-th response line subset at the j-th scan.

[0062] In an embodiment of the present application, the linear extrapolation of the true coincidence event detection count rate under a preset activity to obtain the ideal true coincidence event count rate of each response line subset under each activity includes: taking the preset activity as the independent variable and the true coincidence event detection count rate as the dependent variable, performing least squares straight line fitting, and linearly extrapolating to the ideal true coincidence event count rate of each response line subset when the activity is greater than the preset activity, wherein the true coincidence event detection count rate of each response line subset at the preset activity is equivalent to its ideal true coincidence event count rate.

[0063] In the embodiment of the present application, the preset activity includes several activities less than a specific value, wherein the specific value A j It is calculated by the following formula (4):

[0064]

[0065] Among them, A 0 is the initial activity, T 1 / 2 is the half-life of the radioactive source, T j The time from the initial activity measurement to the start of the jth scan, T acq,j is the duration of the j-th scan.

[0066] In the embodiment of the present application, the dead time correction factor of each response line subset at each activity is obtained according to the corresponding true coincidence event detection count rate and the true coincidence event ideal count rate, including: the dead time correction factor is obtained by using the following formula (5):

[0067]

[0068] in, is the ideal count rate of true coincidence events of the i-th response line subset at the j-th scan, is the true coincidence event detection count rate of the i-th response line subset at the j-th scan.

[0069] In the embodiment of the present application, the processing unit is configured to: if the current random coincidence event count rate is within the random coincidence event count rate range covered by the preset interpolation table, determine whether there is a value in the preset interpolation table that is the same as the current random coincidence event count rate; if so, use the dead time correction factor corresponding to the value as the dead time correction factor of the corresponding response line subset; if not, obtain the two values ​​R in the preset interpolation table that are closest to the current random coincidence event count rate 1 and R 2 and its corresponding dead time correction factor CF1 and CF 2 , and the dead time correction factor CF of the corresponding response line subset is calculated using the following formula (6): DT :

[0070]

[0071] Among them, R ob is the current random coincidence event count rate.

[0072] In the embodiment of the present application, the processing unit is configured to: if the current random coincidence event count rate is greater than the random coincidence event count rate range covered by the preset interpolation table, then obtain the random coincidence event count rate R of the first acquisition and the second acquisition in the preset interpolation table 3 and R 4 and its corresponding dead time correction factor CF 3 and CF 4 , and the dead time correction factor CF of the corresponding response line subset is calculated using the following formula (7): DT :

[0073]

[0074] Among them, R ob is the current random coincidence event count rate.

[0075] In the embodiment of the present application, the acquisition unit is further configured to: use a crystal ring composed of a single axial crystal as a grouping unit, and when the crystals at both ends of the response line are located in the same crystal ring, they are grouped into the same response line subset.

[0076] In the embodiment of the present application, the acquisition unit is further configured to:

[0077] When the crystals are arranged in a crystal array, the response lines from the same two crystal arrays are grouped together.

[0078] In the embodiment of the present application, the radiation source includes 18 F. 22 Na, 68 At least one of Ge.

[0079] In a third aspect, an image reconstruction method is provided, comprising: during the image reconstruction process, performing dead time effect correction on coincident events according to the dead time correction method described in the first aspect.

[0080] In a fourth aspect, an image reconstruction system is provided, comprising the dead time correction system described in the second aspect.

[0081] In a fifth aspect, a positron emission tomography device is provided, the positron emission tomography device comprising the dead time correction system described in the second aspect or the image reconstruction method provided in the fourth aspect; and / or,

[0082] The positron emission tomography imaging device adopts the dead time correction method described in the first aspect to perform dead time correction or the image reconstruction method described in the third aspect to perform image reconstruction.

[0083] In a sixth aspect, a storage medium is provided, wherein the storage medium stores a computer program, wherein the computer program is configured to implement the dead time correction method described in the first aspect or the image reconstruction method described in the third aspect when executed.

[0084] In a seventh aspect, a computer program product is provided, comprising a computer program or instructions, wherein when the computer program or instructions are executed by a processor, the dead time correction method described in the first aspect or the image reconstruction method described in the third aspect is implemented.

[0085] The dead time correction method provided in the present application is a local correction method, which uses the interpolation method to obtain the dead time correction factor under each random coincidence event count rate, and associates the dead time correction factor with the response line subset by dividing the response line subset. It can be applied to the situation where the activity distribution inside the FOV is uneven or there is an activity distribution outside the FOV, and it does not rely on the model and complex external conditions. Compared with the existing global dead time correction method and the model-based fitting method, it can not only obtain a more accurate dead time correction factor, but also has better universality.

[0086] Part of the optional features and other effects of the embodiments of the present application are described below, and part of them can be understood by reading this document. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] The present application will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same reference numerals represent the same structure, wherein:

[0088] Figure 1 A flowchart of a dead time correction method according to an embodiment of the present application is shown;

[0089] Figure 2 A schematic structural diagram corresponding to a crystal ring grouping method according to an embodiment of the present application is shown;

[0090] Figure 3 A schematic structural diagram corresponding to a crystal ring grouping method according to an embodiment of the present application is shown;

[0091] Figure 4A schematic structural diagram corresponding to a crystal array grouping method according to an embodiment of the present application is shown;

[0092] Figure 5 A schematic structural diagram corresponding to a module ring grouping method according to an embodiment of the present application is shown;

[0093] Figure 6 A schematic structural diagram corresponding to a module ring grouping method according to an embodiment of the present application is shown;

[0094] Figure 7 A flowchart of obtaining a preset interpolation table according to an embodiment of the present application is shown;

[0095] Figure 8 A schematic diagram of a dead time correction method according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0096] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present application, so the present application is not limited by the specific embodiments disclosed below.

[0097] It should be noted that when an element is referred to as "fixed to" another element, it can be directly on another element or there can also be a centered element. When an element is considered to be "connected" to another element, it can be directly connected to another element or there may be a centered element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are only for illustrative purposes. The described features, structures or characteristics can be combined in one or more embodiments in any suitable manner. In the following description, specific details are provided so as to provide a full understanding of the embodiments of the present application. However, those skilled in the art should be aware that, without one or more of these specific details, the technical solution of the present application can also be put into practice, or the technical solution of the present application can be realized by other modes, components, materials, devices or operations. In these cases, known structures, methods, devices, implementations, materials or operations will not be shown or described in detail.

[0098] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0099] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices. The term "and / or" or "and / or" includes any and all combinations of one or more related listed items.

[0100] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art. The terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the present application.

[0101] As mentioned above, in the technology known to the inventors, a global dead time correction method or a model fitting method is generally used to correct the dead time effect in the PET reconstruction process. The global dead time correction method does not take into account that detectors at different positions require different dead time correction factors, so the correction effect is not ideal. The model fitting method has poor universality and cannot be applied to different systems. In some cases, due to different conditions during acquisition, it cannot even be applied to the same system, and the work of building a model is also relatively difficult.

[0102] In this regard, the present application provides a dead time correction method, which can be applied to image reconstruction. Figure 1 As shown, the dead time correction method provided by the present application generally includes the following steps:

[0103] S110: Obtaining the random coincidence event count rate of each response line subset.

[0104] Since the existing dead time correction using a global dead time correction factor ignores the differences in dead time effects at different detection positions, different dead time correction factors should be used for different detection positions. Therefore, in this application, the random coincidence event count rate is obtained in units of LOR subsets.

[0105] Before step S110, a radioactive source needs to be injected into the body to be tested. The radioactive source decays in the body to be tested to produce positrons. The positrons combine and annihilate with the negative electrons in the body to be tested, and then a pair of gamma photons with the same energy but opposite flight directions are produced. The detector of the positron emission tomography device (PET device) can detect a large number of gamma photons. Specifically, the gamma rays are converted into optical signals by scintillation crystals, and then the optical signals are converted into electrical signals by photoelectric conversion devices. The position (Position), energy (Energy), and time (Time) information of the gamma rays are accurately extracted in a digital manner, and then the digitized information is sent to the back-end server. The back-end server restores the original waveform by fitting, extracts key information, and performs compliance processing. The details of the above content can be found in the prior art and will not be repeated here. Image reconstruction is performed after the compliance processing, and the dead time effect is corrected during image reconstruction.

[0106] Specifically, in one example, step S110 includes: acquiring detection data of a positron emission computed tomography device, performing matching according to the detection data, and obtaining a random matching event count rate of each of the response line subsets according to the matching result. According to the matching according to the detection data, all response lines and counts on each response line can be obtained.

[0107] In one embodiment, during image reconstruction, the detection data is matched to generate a delayed coincidence data sinusoid. A pixel value on the delayed coincidence data sinusoid is the number of delayed coincidence events on a specific LOR, and all pixels correspond to all LORs. The LOR is divided into y subsets, denoted as φ 1 ,φ 2 ,…,φ y , and simultaneously obtain the acquisition time corresponding to all LOR subsets, and then obtain the random coincidence event count rate in each LOR subset i according to the delayed coincidence data sinusoidal diagram and the corresponding acquisition time.

[0108] Regarding the LOR division method, in one example, all LORs may be grouped according to the location characteristics of the LORs to obtain several LOR subsets.

[0109] Due to the influence of the location of the radiation source, the activity of the radiation source, the acquisition time, etc., the values ​​of different LORs in the same sinusoidal data will have large differences. Ideally, the correction is made directly based on each LOR, but due to the small number of counts on each LOR, and some LORs may not have counts, in this case, directly counting the counts of each LOR has a large statistical error, and the influence of the activity distribution on the LORs at adjacent positions can be ignored. Therefore, this application adopts grouping of LORs, which can also be understood as downsampling the sinusoidal data so that the obtained data has less statistical noise. LORs in the same group share the same dead time correction factor, and the dead time correction factors of different groups are independent of each other.

[0110] In order to ensure that the downsampled data still retains the characteristics of the original data, the LORs are usually divided according to their position characteristics, such as the crystal ring grouping method, the block pair grouping method and the block ring grouping method.

[0111] For the crystal ring grouping method, a crystal ring consisting of a single axial crystal is used as the grouping unit, where the axial direction is parallel to the axial direction of the detector. For all LORs, consider which crystal ring the two crystals at both ends belong to. If two LORs belong to the same crystal ring, the two LORs are grouped together. For details, see Figure 2 , wherein 300 represents a test object, crystal 110 and crystal 210 and several crystals not shown in the figure constitute a crystal ring, crystal 120 and crystal 220 and several crystals not shown in the figure constitute a crystal ring, Figure 2 A total of 16 crystal rings are shown, such as Figure 2 In the example, different LORs need to be grouped. For any LOR, if the crystal rings corresponding to its two ends are the same, they are grouped into the same group. For example, Figure 2 The two LORs in the graph have one end located in the crystal ring 11 and the other end located in the crystal ring 12. The two LORs are grouped into the same group. Figure 2 In the example, 16×16 groups can be formed. Figure 3 , which shows three crystal rings, wherein two straight lines represent two LORs. If one end of the two LORs is located in crystal ring 1 (Crystal ring 1) and the other end is located in crystal ring 2 (Crystal ring 2), the two LORs are divided into the same group.

[0112] For the crystal array pair grouping method, Block is also the crystal array. In the PET system, a crystal array is connected to one or a group of photodetectors. Each crystal belongs to a crystal array. Therefore, the LORs corresponding to two crystals can be regarded as the LORs corresponding to two crystal arrays, and the LORs from the same two crystal arrays can be grouped together. For details, see Figure 4 In the same crystal ring, four crystals form a crystal array, of which 12 crystal arrays are shown. Figure 4 In the example of , if both ends of two LORs are located in crystal array 1 (Block 1) and crystal array 7 (Block 7) respectively, the two LORs are grouped into one group.

[0113] For the modular ring grouping method, a ring consisting of a single axial block is called a modular ring, where the axial direction is parallel to the axial direction of the detector. For all LORs, consider which modular ring the crystals at both ends belong to. If the crystals at both ends of two LORs belong to the same modular ring, the two LORs are grouped together. Figure 5 In the same module ring, the corresponding crystal array on any cross section passing through the center includes four crystals. Crystal array 110 and crystal array 120 and several crystal arrays not shown in the figure constitute a module ring. Crystal array 120 and crystal array 220 and several crystal arrays not shown in the figure constitute a module ring. Figure 5 The total number of modules is shown as 4: Figure 5 In the example, if one end of the two LORs is located in the same module ring, such as crystal array 2 (Block ring 2), and the other end is located in the same module ring, such as crystal array 3 (Block ring 3), then the two LORs are divided into the same group. Figure 5 In the example, 4×4 groups can be formed. Figure 6 , which shows three module rings, where two straight lines represent two LORs. If one end of the two LORs is located in the same Block, such as module ring 1 (Block ring 1), and the other end is also located in the same Block, such as module ring 2 (Block ring 2), then the two LORs are divided into the same group.

[0114] In practical applications, the accuracy of the crystal ring grouping method and the crystal array pair grouping method is higher than that of the module ring grouping method, but at the same time, the calculation amount of the crystal ring grouping method and the crystal array pair grouping method is very high. Therefore, in order to balance the accuracy and calculation amount, the module ring grouping method is preferably used to group the LORs. ×The ring formed by the overall m crystal arrays is defined as a modular ring. When the crystals at both ends of the response line a and the crystals at both ends of the response line b belong to the same modular ring, the response line a and the response line b are divided into the same response line subset, wherein n and m are both positive integers, and at least one of n or m is greater than 2.

[0115] In one example, the random coincidence event count rate of the i-th response line subset is calculated by the following formula (1):

[0116]

[0117] Where T is the acquisition duration, D k is the count on the kth response line in the delay conformance data sinusoid.

[0118] Furthermore, in the present application, after step S110, the dead time correction method further includes:

[0119] S120: traverse all random coincidence event count rates, and determine one by one whether the random coincidence event count rate is less than the minimum value of the random coincidence event count rate range covered by the preset interpolation table.

[0120] The preset interpolation table may be obtained in advance through experiments.

[0121] Specifically, in one example of the present application, Figure 7 As shown, the preset interpolation table can be obtained by following the steps below:

[0122] S121: Collect detection experimental data of radioactive sources at different activities under experimental conditions, and obtain response lines by fitting the data.

[0123] Among them, the radiation source is required to emit only positrons and not gamma photons in the energy range of 400keV-650keV to avoid affecting positron annihilation. Typical radiation sources include 18 F. 22 Na, 68Ge et al., In addition, it is required that the initial activity of the radioactive source is not lower than the maximum activity of the PET device used during collection. In special cases, when the initial activity is lower than the maximum activity of the PET device used during collection, the linear extrapolation method proposed below can be used to make the data complete. The PET device used in this experiment requires the initial activity of the radioactive source to be 130MBq. The phantom used in the experiment can be cylindrical or in other shapes. The length direction of the phantom is parallel to the axial direction of the PET / CT. The length of the phantom is required to cover the axial field of view of the PET / CT device to ensure that the detectors at each position have sufficient data. Therefore, phantoms of different lengths are required for different systems. Fill the phantom with a radioactive source. In some cases, the radioactive source can be diluted, and the liquid used for dilution can be distilled water.

[0124] Step S121 further includes: collecting detection experimental data of the same radioactive source multiple times within a predetermined time. The predetermined time is also the total collection time, which can be set according to factors such as the initial activity and decay rate of the radioactive source. The total collection time includes the collection duration and the collection interval time. Generally speaking, as the collection proceeds, the activity of the radioactive source gradually decays. When the radioactive source is at a high activity, the data has an obvious dead time effect. When it decays to a value lower than a certain activity, it can be considered that there is no dead time effect. For example, it is found through testing that, for an existing small animal PET device, when the activity of the radioactive source is lower than 5MBq, the true coincidence event count rate changes linearly with the activity of the radioactive source, that is, there is no dead time effect. In order to estimate the ideal true coincidence event count rate under high activity, the dead time effect is corrected, and it is necessary to ensure that more than a certain number of the collected data have no dead time effect. For example, in one example, it is required that the data collected for the last five times have no dead time effect, and the collection time is, for example, 19 hours.

[0125] In one embodiment, for different systems, one of the strategies shown in the following Table 1 may be selected for data collection. Of course, those skilled in the art should understand that Table 1 is only shown as an example, and other collection strategies may also be used for data collection according to actual needs.

[0126] Table 1 Collection strategy table

[0127]

[0128] S122: Divide all response lines into a number of response line subsets, and obtain the random coincidence event count rate of each response line subset at each activity.

[0129] As for the LOR division method, one of the methods described in the subsequent step S110 may be used, and preferably, the modular ring grouping method is used for grouping, which will not be described in detail here.

[0130] In one example, the above formula (1) can be used to calculate the random coincidence event count rate of each LOR subset.

[0131] S123: Obtain the instant coincidence event count rate of each response line subset at each activity.

[0132] After step S121, a delayed coincidence data sinusoidal graph is generated while also generating an instantaneous coincidence data sinusoidal graph. A pixel value on the delayed coincidence data sinusoidal graph is the number of random coincidence events on a specific LOR, and a pixel value on the instantaneous coincidence data sinusoidal graph is the number of instantaneous coincidence events on a specific LOR. All pixels of the sinusoidal graph correspond to all LORs.

[0133] As mentioned above, as the acquisition proceeds, the activity of the radioactive source continues to decay. Through multiple acquisitions, experimental detection data at different activities of the radioactive source can be obtained, including relatively high and low activities. The instant coincidence event count rates derived therefrom are also at different activities.

[0134] In one example of the present application, the instant coincidence event count rate of the i-th response line subset at the j-th scan is calculated by the following formula (2):

[0135]

[0136] Among them, P kj is the number of instant coincidence events of the kth response line in the i-th response line subset at the jth scan, and T is the scan duration.

[0137] S124: Obtaining the true coincidence event detection count rate of each response line subset at each activity according to the instant coincidence event count rate and the random coincidence event count rate.

[0138] It can be understood that the various activities here also represent both high activities and low activities.

[0139] In one example of the present application, the true coincidence event detection count rate is calculated by the following formula (3):

[0140]

[0141] in, represents the instantaneous coincidence event count rate of the i-th response line subset at the j-th scan, is the random coincidence event count rate.

[0142] S125: According to the true coincidence event detection count rate at the preset activity, the ideal true coincidence event count rate of each response line subset at each activity is obtained by linearly extrapolating.

[0143] It should be particularly noted that the preset activity includes several activities that are less than a specific value. It can be understood that in order to ensure the accuracy of linear extrapolation, the activity values ​​included in the preset activity are required to reach a certain amount to ensure sufficient data volume. For example, it includes five activities that are less than a specific value. Taking an existing small animal PET as an example, the preset activity includes several activities that are less than 5MBq. Under these activities, the true coincidence event count rate changes linearly with the activity of the radiation source, and the true coincidence event detection count rate is the ideal true coincidence event count rate.

[0144] In step S125, regarding linear extrapolation, the preset activity can be used as the independent variable and the true coincidence event detection count rate can be used as the dependent variable to perform least squares straight line fitting, and linearly extrapolate to the ideal true coincidence event count rate of each response line subset when the activity is greater than the preset activity.

[0145] In one example, the specific value A can be calculated using the following formula (4): j :

[0146]

[0147] Among them, A 0 is the initial activity, T 1 / 2 is the half-life of the radioactive source, T j The time from the initial activity measurement to the start of the jth scan, T acq,j is the duration of the j-th scan.

[0148] S126: Obtain the dead time correction factor of each response line subset at each activity according to the corresponding true coincidence event detection count rate and the true coincidence event ideal count rate, and the dead time correction factor and the corresponding random coincidence event count rate form the preset interpolation table in a one-to-one correspondence.

[0149] It can be understood that in step S126, the true coincidence event detection count rate under the preset activity is regarded as the ideal true coincidence event count rate, and outside the preset activity, the ideal true coincidence event count rate is obtained by linear extrapolation of the corresponding true coincidence event detection count rate, and the random coincidence event count rate is obtained by step S122.

[0150] In one example of the present application, the dead time correction factor is obtained using the following formula (5):

[0151]

[0152] in, is the ideal count rate of true coincidence events of the i-th response line subset at the j-th scan, is the true coincidence event detection count rate of the i-th response line subset at the j-th scan.

[0153] Thus, according to the above formula (3) and formula (5), each LOR subset can generate a random coincidence event count rate -Dead time correction factor The value pair at each scan j Where i = 1, 2, ..., u, j = 1, 2, ..., v, u is the total number of acquisitions, all the value pairs The preset interpolation table is constructed.

[0154] It can be understood that, in this example, at the preset activity, since the true coincidence event detection count rate is regarded as the true coincidence event ideal count rate, that is, at the preset activity, the dead time correction factor Approximately equal to 1.

[0155] Further, in the present application, after step S120, the dead time correction method further includes:

[0156] S130: If the random coincidence event count rate is less than the minimum value of the random coincidence event count rate range covered by the preset interpolation table, specifying the dead time correction factor of the response line subset corresponding to the current random coincidence event count rate as a preset value.

[0157] The random coincidence event count rate range covered by the preset interpolation table basically covers all situations where dead time effects may occur. When the random coincidence event count rate of the LOR subset is less than the minimum value of the random coincidence event count rate range covered by the preset interpolation table, it means that the random coincidence event count rate of the LOR subset is already very small and can be ignored. In this case, when the preset interpolation table is obtained by the method described in steps S121 to S126 above, the preset value is 1, which means that there is no need to correct for the dead time effect.

[0158] S140: If the random coincidence event count rate is greater than or equal to the minimum value of the random coincidence event count rate range covered by the preset interpolation table, a dead time correction factor of the response line subset corresponding to the current random coincidence event count rate is obtained by a preset method.

[0159] Exemplarily, step S140 also includes at least three situations:

[0160] 1. If the current random coincidence event count rate is within the range of random coincidence event count rates covered by the preset interpolation table, and the preset interpolation table has a value that is exactly the same as the current random coincidence event count rate, then specify the dead time correction factor corresponding to the current random coincidence event count rate as the dead time correction factor of the corresponding response line subset;

[0161] 2. If the current random coincidence event count rate is within the range of random coincidence event count rates covered by the preset interpolation table, and there is no value in the preset interpolation table that is exactly the same as the current random coincidence event count rate, then obtain the two values ​​R in the preset interpolation table that are closest to the current random coincidence event count rate. 1 and R 2 and its corresponding dead time correction factor CF 1 and CF 2 , and the dead time correction factor CF of the corresponding response line subset is calculated using the following formula (6): DT :

[0162]

[0163] Among them, R ob is the current random coincidence event counting rate;

[0164] 3. If the current random coincidence event count rate is greater than the random coincidence event count rate range covered by the preset interpolation table, then obtain the random coincidence event count rate R of the first and second acquisitions in the preset interpolation table. 3 and R 4 and its corresponding dead time correction factor CF 3 and CF 4 , and the dead time correction factor CF of the corresponding response line subset is calculated using the following formula (7): DT :

[0165]

[0166] Among them, R ob is the current random coincidence event count rate.

[0167] Among them, the first and second acquisitions represent the two acquisitions with the highest radioactive source activity.

[0168] In an example, the relationship between the dead time correction factor and the random coincidence event count rate can be derived using the following process:

[0169]

[0170] in, and are the single event count rates of crystal i and crystal j, respectively, R ideal represents the ideal random coincidence event count rate, τ represents the coincidence time window;

[0171] It is required that τ cannot be too large. In one example, τ is slightly larger than the maximum flight time difference.

[0172] The single event count rate obeys the paralysis model or the non-paralysis model. ideal Expressed as S ob The function f(S ob ),but

[0173]

[0174] right As an approximation,

[0175]

[0176] By analogy with formula (11), we can get

[0177]

[0178] Right now

[0179]

[0180] in, is the detected single event count rate of scintillation crystal i, is the detected single event count rate of scintillation crystal j, R ob represents the random coincidence event count rate;

[0181] Substituting formula (15) into formula (13) and then into formula (12), we can get

[0182]

[0183] Since the dead time correction factor CF DT It is defined as the ratio of the ideal count rate of true coincidence events to the detected count rate of true coincidence events, that is,

[0184]

[0185] From this, we can find that the dead time correction factor can be regarded as a function only about the random coincidence event count rate. Since the specific expression of the function is unknown, an interpolation table of random coincidence event count rate-dead time correction factor can be established. Through sampling and interpolation methods, the dead time correction factor under full activity can be found.

[0186] The dead time correction method provided in the present application is a local correction method, which uses the interpolation method to obtain the dead time correction factor under each random coincidence event count rate, and associates the dead time correction factor with the response line subset by dividing the response line subset. It can be applied to the situation where the activity distribution inside the FOV is uneven or there is activity distribution outside the FOV, especially when correcting the uneven axial FOV dead time effect. It is effective and does not rely on models and complex external conditions. Compared with the existing global dead time correction method and model-based fitting method, it can not only obtain a more accurate dead time correction factor, but also has better universality.

[0187] In addition, the dead time correction method provided in this application innovatively uses an interpolation table to calculate the dead time correction factor. Through a decay experiment, the interpolation table of a specific system is obtained, and linear extrapolation is used to obtain the dead time correction factor outside the interpolation table. Compared with the method using a model, it avoids parameter fitting and is not constrained by system limitations. The effect has been verified on a variety of devices.

[0188] Corresponding to the above-mentioned dead time correction method, the present application also provides an image reconstruction method, which uses the dead time correction method provided by any of the above-mentioned examples to correct the dead time effect of the coincident events in the reconstruction process.

[0189] After the above dead time correction method traverses all LOR subsets to obtain the dead time correction factor for each response line, the dead time correction factor can be written as a vector D, whose diagonal elements D i is the dead time correction factor on the i-th LOR, similar to the vector N for normalization correction and the vector A for attenuation correction.

[0190] When applied, for example, to the existing OSEM reconstruction algorithm, the iterative formula of the OSEM reconstruction algorithm is:

[0191]

[0192] in and denote the value of voxel j in the nth and n+1th iterations, respectively, N i and A i denote the normalized corrected sine value and attenuation corrected sine value of the i-th LOR, y i represents the number of immediate matching events of the i-th LOR, S ij represents the system response matrix, R k is the random correction value of voxel k. If the dead time correction factor is introduced, the iterative formula can be correspondingly changed to:

[0193]

[0194] In a further example, the dead time correction factor can also be combined with the normalization factor, that is, let:

[0195] N' i =D i N i …………………………(10)

[0196] In this way, OSEM reconstruction can still use formula (8), just replace N i Replace with N' i That's it.

[0197] The image reconstruction method provided in the present application uses an interpolation method to obtain a dead time correction factor at each random coincidence event count rate, and applies it to reconstruction. The dead time correction factor is associated with a response line subset by dividing the response line subset. The method can be applied to situations where the activity distribution inside the FOV is uneven or there is an activity distribution outside the FOV, and the method is independent of models and complex external conditions. Compared with existing global dead time correction methods and model-based fitting methods, the method can not only obtain a more accurate dead time correction factor, but also has better universality.

[0198] Corresponding to the above dead time correction method, see Figure 8 The present application also provides a dead time correction system, including an acquisition unit 410, a judgment unit 420 and a processing unit 430. The acquisition unit 410 is used to acquire the random coincidence event count rate of each response line subset. The judgment unit 420 is used to traverse all random coincidence event count rates, and determine one by one whether the random coincidence event count rate is less than the minimum value of the random coincidence event count rate range covered by the preset interpolation table. The processing unit 430 is used to specify the dead time correction factor of the response line subset corresponding to the current random coincidence event count rate as a preset value when the random coincidence event count rate is less than the minimum value of the random coincidence event count rate range covered by the preset interpolation table, and to acquire the dead time correction factor of the response line subset corresponding to the current random coincidence event count rate by a preset method when the random coincidence event count rate is greater than or equal to the minimum value of the random coincidence event count rate range covered by the preset interpolation table.

[0199] Since the existing dead time correction using a global dead time correction factor ignores the differences in dead time effects at different detection positions, different dead time correction factors should be used for different detection positions. Therefore, in this application, the acquisition unit acquires the random coincidence event count rate in units of LOR subsets.

[0200] Before the acquisition unit 410 acquires the random coincidence event count rate, a radioactive source needs to be injected into the body to be tested. The radioactive source decays in the body to be tested to produce positrons. The positrons combine and annihilate with the negative electrons in the body to be tested, and then a pair of gamma photons with the same energy but opposite flight directions are generated; a large number of gamma photons are detected by the detector of the positron emission tomography device (PET device). Specifically, the gamma rays are converted into light signals by scintillation crystals, and then the light signals are converted into electrical signals by photoelectric conversion devices. The position (Position), energy (Energy), and time (Time) information of the gamma rays are accurately extracted in a digital manner, and then the digitized information is sent to the back-end server, and the original waveform is restored by fitting, key information is extracted, and coincidence processing is performed. The details of the above content can be found in the prior art and will not be repeated here. Image reconstruction is performed after coincidence processing, and dead time effect correction is performed during image reconstruction.

[0201] Specifically, in one example, the acquisition unit 410 is configured to acquire detection data of a positron emission tomography device, and perform coincidence based on the detection data to obtain a random coincidence event count rate of each of the response line subsets.

[0202] In one embodiment, during image reconstruction, the detection data is matched to generate a delayed coincidence data sinusoid. A pixel value on the delayed coincidence data sinusoid is the number of delayed coincidence events on a specific LOR, and all pixels correspond to all LORs. The LOR is divided into y subsets, denoted as φ 1 ,φ 2 ,…,φ y , and simultaneously obtain the acquisition time corresponding to all LOR subsets, and then obtain the random coincidence event count rate in each LOR subset i according to the delayed coincidence data sinusoidal diagram and the corresponding acquisition time.

[0203] Regarding the LOR division method, in one example, the acquisition unit can group all LORs according to the position characteristics of the LOR to obtain a plurality of LOR subsets before acquiring the random coincidence event count rate of each response line subset. The specific division method can refer to the embodiment of the dead time correction method above, which will not be repeated here.

[0204] Preferably, the LORs are grouped using a modular ring grouping method. Exemplarily, the acquisition unit 410 is configured to define a ring formed by an array of n×m crystals as a modular ring, and when the crystals at both ends of the response line a and the crystals at both ends of the response line b belong to the same modular ring, the response line a and the response line b are grouped into the same response line subset, wherein n and m are both positive integers, and at least one of n or m is greater than 2.

[0205] Exemplarily, the acquisition unit 410 calculates the random coincidence event count rate of the i-th response line subset according to the following formula (1):

[0206]

[0207] Where T is the acquisition duration, D k is the count on the kth response line in the delay conformance data sinusoid.

[0208] Regarding the method for obtaining the preset interpolation table, please refer to the contents recorded after step 120 of the above-mentioned dead time correction method, which will not be repeated here.

[0209] Preferably, according to the relationship between the random coincidence event count rate and the preset interpolation table, the processing unit is configured into at least three situations:

[0210] 1. If the current random coincidence event count rate is within the range of random coincidence event count rates covered by the preset interpolation table, and the preset interpolation table has a value that is exactly the same as the current random coincidence event count rate, then specify the dead time correction factor corresponding to the current random coincidence event count rate as the dead time correction factor of the corresponding response line subset;

[0211] 2. If the current random coincidence event count rate is within the range of random coincidence event count rates covered by the preset interpolation table, and there is no value in the preset interpolation table that is exactly the same as the current random coincidence event count rate, then obtain the two values ​​R in the preset interpolation table that are closest to the current random coincidence event count rate. 1 and R 2 and its corresponding dead time correction factor CF 1 and CF 2 , and the dead time correction factor CF of the corresponding response line subset is calculated using the following formula (6): DT :

[0212]

[0213] Among them, R ob is the current random coincidence event counting rate;

[0214] 3. If the current random coincidence event count rate is greater than the random coincidence event count rate range covered by the preset interpolation table, then obtain the random coincidence event count rate R of the first and second acquisitions in the preset interpolation table. 3 and R 4 and its corresponding dead time correction factor CF 3 and CF 4 , and the dead time correction factor CF of the corresponding response line subset is calculated using the following formula (7):DT :

[0215]

[0216] Among them, R ob is the current random coincidence event count rate.

[0217] Among them, the first and second acquisitions represent the two acquisitions with the highest radioactive source activity.

[0218] The derivation process of the relationship between the dead time correction factor and the random coincidence event count rate can also refer to the contents recorded in the embodiment of the above dead time correction method, which will not be repeated here. Through derivation, it can be found that the dead time correction factor can be regarded as a function only about the random coincidence event count rate. Since the specific expression of the function is unknown, we can establish an interpolation table of the random coincidence event count rate-dead time correction factor, and find the dead time correction factor under full activity by sampling and interpolation.

[0219] The dead time correction system provided by the present application adopts a local correction method and uses an interpolation method to obtain a dead time correction factor under each random coincidence event count rate. The dead time correction factor is associated with a response line subset by dividing the response line subset. It can be applicable to situations where the activity distribution inside the FOV is uneven or there is an activity distribution outside the FOV, and it does not rely on models and complex external conditions. Compared with the existing global dead time correction method and model-based fitting method, it can not only obtain a more accurate dead time correction factor, but also has better universality.

[0220] In some embodiments, the present application further provides an image reconstruction system, which includes the dead time correction system provided in the above embodiments, and can also use the dead time correction method provided in the above embodiments to correct the dead time effect in the reconstruction process.

[0221] After the above dead time correction method traverses all LOR subsets to obtain the dead time correction factor for each response line, the dead time correction factor can be written as a vector D, whose diagonal elements D i is the dead time correction factor on the i-th LOR, similar to the vector N for normalization correction and the vector A for attenuation correction.

[0222] When applied, for example, to the existing OSEM reconstruction algorithm, the iterative formula of the OSEM reconstruction algorithm is:

[0223]

[0224] in and denote the value of voxel j in the nth and n+1th iterations, respectively, Ni and A i denote the normalized corrected sine value and attenuation corrected sine value of the i-th LOR, y i represents the number of immediate matching events of the i-th LOR, S ij represents the system response matrix, R k is the random correction value of voxel k. If the dead time correction factor is introduced, the iterative formula can be correspondingly changed to:

[0225]

[0226] In a further example, the dead time correction factor can also be combined with the normalization factor, that is, let:

[0227] N' i =D i N i …………………………(10)

[0228] In this way, OSEM reconstruction can still use formula (8), just replace N i Replace with N' i That's it.

[0229] The image reconstruction system provided by the present application uses an interpolation method to obtain a dead time correction factor at each random coincidence event count rate, and applies it to reconstruction. The dead time correction factor is associated with a response line subset by dividing the response line subset. This system can be applied to situations where the activity distribution inside the FOV is uneven or there is an activity distribution outside the FOV, and is independent of models and complex external conditions. Compared with existing global dead time correction methods and model-based fitting methods, this system can not only obtain a more accurate dead time correction factor, but also has better universality.

[0230] In addition, the dead time correction system provided by the present application innovatively uses an interpolation table to calculate the dead time correction factor. Through a decay experiment, the interpolation table of a specific system is obtained, and linear extrapolation is used to obtain the dead time correction factor outside the interpolation table. Compared with the method of using a model, it avoids parameter fitting and is not constrained by system limitations. The effect has been verified on a variety of devices.

[0231] In some embodiments, the present application also provides a positron emission tomography device, which may include a dead time correction system or an image reconstruction system provided by any one of the above embodiments; in addition, the positron emission tomography device adopts a dead time correction method provided by any one of the above embodiments to perform dead time correction or an image reconstruction method to perform image reconstruction.

[0232] Through the description of the above implementation methods, it is easy for those skilled in the art to understand that the example implementation methods described here can be implemented by software or by combining software with necessary hardware. The technical solution according to the implementation methods of the present application can be embodied in the form of a software product or a computer program product, which can be stored in a computer-readable storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including a number of computer program instructions to enable a computing device (which can be a personal computer, a server, or a network device, etc.) to execute the above method according to the implementation methods of the present application.

[0233] Software products and computer program products may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0234] Computer readable storage media may include data signals propagated in baseband or as part of a carrier wave, wherein readable program codes are carried. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program codes contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination thereof.

[0235] Program code for performing the operations of the present application may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as C or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0236] The computer-readable medium carries one or more program instructions. When the one or more program instructions are executed by a device, the computer-readable medium implements the aforementioned functions.

[0237] Through the description of the above implementation methods, it is easy for those skilled in the art to understand that the example implementation methods described here can be implemented by software or by combining software with necessary hardware. The technical solution according to the implementation method of the present application can be embodied in the form of a software product, which can be stored in a computer-readable storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including a number of computer program instructions to enable a computing device (which can be a personal computer, a server, or a network device, etc.) to execute the above method according to the implementation method of the present application.

[0238] Those skilled in the art will appreciate that the above modules can be distributed in the device according to the description of the embodiment, or can be changed accordingly and only used in one or more devices different from the embodiment. The multiple modules of the above embodiments can be combined into one module, or one module can be further divided into multiple sub-modules.

[0239] Although the present application provides the method operation steps as described in the above embodiments or flow charts, more or fewer operation steps may be included in the method based on routine or no creative labor. In the steps where there is no necessary causal relationship logically, the execution order of these steps is not limited to the execution order provided in the embodiments of the present application.

[0240] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0241] The embodiments of the present application are described in detail above. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and its core idea of ​​the present application. At the same time, changes or deformations made by those skilled in the art based on the ideas of the present application, the specific implementation methods and the scope of application of the present application, all belong to the scope of protection of the present application. In summary, the content of this specification should not be construed as a limitation on the present application.

Claims

1. A dead time correction method, characterized in that: The dead time correction method comprises: Obtain the random coincidence event count rate for each response line subset; Traversing all random coincidence event count rates, and determining one by one whether the random coincidence event count rate is less than a minimum value of the random coincidence event count rate range covered by a preset interpolation table; If so, specifying the dead time correction factor of the subset of response lines corresponding to the current random coincidence event count rate as a preset value; If not, the dead time correction factor of the response line subset corresponding to the current random coincidence event count rate is obtained by a preset method.

2. The dead time correction method according to claim 1, characterized in that: The obtaining of the random coincidence event count rate of each response line subset comprises: Detection data of a positron emission tomography device is acquired, coincidence is performed according to the detection data, and a random coincidence event count rate of each of the response line subsets is obtained according to the coincidence result.

3. The dead time correction method according to claim 2, characterized in that: The step of obtaining a random coincidence event count rate of each of the response line subsets by performing coincidence according to the detection data comprises: Performing matching according to the detection data to obtain a delayed matching data sinusoidal diagram; The random coincidence event count rate of each response line subset is calculated according to the delayed coincidence data sinusoidal diagram and the acquisition time of the corresponding response line subset.

4. The dead time correction method according to claim 3, characterized in that: The step of calculating the random coincidence event count rate of each response line subset according to the delayed coincidence data sinusoidal diagram and the acquisition time of the corresponding response line subset comprises: The random coincidence event count rate of the i-th response line subset is calculated according to the following formula (1): Where T is the acquisition duration, D k is the count on the kth response line in the delay conformance data sinusoid.

5. The dead time correction method according to claim 1, characterized in that: Before obtaining the random coincidence event count rate of each response line subset, the dead time correction method further includes: All response lines are grouped according to the position characteristics of the response lines to obtain several response line subsets.

6. The dead time correction method according to claim 5, characterized in that: The method of grouping all the response lines according to the position characteristics of the response lines to obtain several response line subsets includes: Will be n × A ring formed by an array of m crystals is defined as a modular ring. When the crystals at both ends of response line a and the crystals at both ends of response line b belong to the same modular ring, response line a and response line b are divided into the same response line subset, wherein n and m are both positive integers, and at least one of n or m is greater than 2.

7. The dead time correction method according to claim 1, characterized in that: The preset interpolation table can be obtained by following the steps below: Collect detection experimental data of radioactive sources at different activities under experimental conditions, and obtain response lines by fitting the data; Divide all response lines into several response line subsets, and obtain the random coincidence event counting rate of each response line subset at each activity; Obtain the instant coincidence event count rate of each response line subset at each activity; Obtaining a true coincidence event detection count rate of each response line subset at each activity according to the instant coincidence event count rate and the random coincidence event count rate; According to the true coincidence event detection count rate under the preset activity, the ideal true coincidence event count rate of each response line subset under each activity is obtained by linear extrapolation; The dead time correction factor of each response line subset at each activity is obtained according to the corresponding true coincidence event detection count rate and the true coincidence event ideal count rate, and the dead time correction factor and the corresponding random coincidence event count rate form the preset interpolation table in a one-to-one correspondence.

8. The dead time correction method according to claim 7, characterized in that: The method of collecting detection experimental data of the radioactive source at different activities under experimental conditions includes: The detection experimental data of the same radioactive source are collected multiple times within a predetermined time, and the factors affecting the predetermined time include the initial activity and decay rate of the radioactive source.

9. The dead time correction method according to claim 7, characterized in that: The true coincidence event detection count rate under the preset activity is regarded as the true coincidence event ideal count rate.

10. The dead time correction method according to claim 7, characterized in that: The calculation of the instant coincidence event count rate of each response line subset at different activities includes: The instant coincidence event count rate of the i-th response line subset at the j-th scan is calculated by the following formula (2): Among them, P kj is the number of instant coincidence events of the kth response line in the i-th response line subset at the jth scan, and T is the acquisition duration.

11. The dead time correction method according to claim 7, characterized in that: The obtaining of the true coincidence event detection count rate of each response line subset at different activities according to the instant coincidence event count rate and the delayed coincidence event count rate comprises: The true coincidence event detection count rate is calculated by the following formula (3): in, represents the instantaneous coincidence event count rate of the i-th response line subset at the j-th scan, represents the random coincidence event count rate of the i-th response line subset at the j-th scan.

12. The dead time correction method according to claim 7, characterized in that: The method of obtaining the ideal true coincidence event count rate of each response line subset at different activities by linearly extrapolating the true coincidence event detection count rate at the preset activity includes: The preset activity is used as the independent variable and the true coincidence event detection count rate is used as the dependent variable. Least squares straight line fitting is performed and linearly extrapolated to the ideal true coincidence event count rate of each response line subset when the activity is greater than the preset activity, wherein the true coincidence event detection count rate of each response line subset is equal to its ideal true coincidence event count rate at the preset activity.

13. The dead time correction method according to claim 12, characterized in that: The preset activity includes several activities less than a specific value, wherein the specific value A j It is calculated by the following formula (4): Where A0 is the initial activity, T 1 / 2 is the half-life of the radioactive source, T j The time from the initial activity measurement to the start of the jth scan, T acq，j is the duration of the j-th scan.

14. The dead time correction method according to claim 7, characterized in that: The dead time correction factor of each response line subset at different activities is obtained according to the corresponding true coincidence event detection count rate and the true coincidence event ideal count rate, including: The dead time correction factor is obtained using the following formula (5): in, is the ideal count rate of true coincidence events of the i-th response line subset at the j-th scan, is the true coincidence event detection count rate of the i-th response line subset at the j-th scan.

15. The dead time correction method according to claim 1, characterized in that: The method of obtaining the dead time correction factor of the response line subset corresponding to the current random coincidence event count rate by a preset method includes: when the current random coincidence event count rate is within the random coincidence event count rate range covered by the preset interpolation table, determining whether there is a value in the preset interpolation table that is the same as the current random coincidence event count rate; if so, using the dead time correction factor corresponding to the value as the dead time correction factor of the corresponding response line subset; if not, obtaining two values ​​R1 and R2 in the preset interpolation table that are closest to the current random coincidence event count rate and their corresponding dead time correction factors CF1 and CF2, and using the following formula (6) to calculate the dead time correction factor CF of the corresponding response line subset. DT : Among them, R ob is the current random coincidence event count rate.

16. The dead time correction method according to claim 1, characterized in that: The dead time correction factor of the response line subset corresponding to the current random coincidence event count rate is obtained by a preset method, including: when the current random coincidence event count rate is greater than the random coincidence event count rate range covered by the preset interpolation table, the random coincidence event count rates R3 and R4 collected for the first time and the second time in the preset interpolation table and their corresponding dead time correction factors CF3 and CF4 are obtained, and the dead time correction factor CF of the corresponding response line subset is calculated by the following formula (7): DT : Among them, R ob is the current random coincidence event count rate.

17. The dead time correction method according to claim 5, characterized in that: The method of grouping all the response lines according to the position characteristics of the response lines to obtain several response line subsets includes: The crystal ring composed of a single axial crystal is used as the grouping unit. When the crystals at both ends of the response line are located in the same crystal ring, they are divided into the same response line subset.

18. The dead time correction method according to claim 5, characterized in that: The method of grouping all the response lines according to the position characteristics of the response lines to obtain several response line subsets includes: When the crystals are arranged in a crystal array, the response lines from the same two crystal arrays are grouped together.

19. The dead time correction method according to claim 7, characterized in that: The radiation source includes 18 F. 22 Na, 68 At least one of Ge.

20. A dead time correction system, characterized in that: The dead time correction system comprises: an acquisition unit configured to acquire a random coincidence event count rate for each response line subset; A judgment unit configured to traverse all random coincidence event count rates and judge one by one whether the random coincidence event count rate is less than a minimum value of the random coincidence event count rate range covered by a preset interpolation table; The processing unit is configured to designate a dead time correction factor of a response line subset corresponding to a current random coincidence event count rate as a preset value when the current random coincidence event count rate is less than a minimum value of a random coincidence event count rate range covered by a preset interpolation table, and to obtain the dead time correction factor of a response line subset corresponding to a current random coincidence event count rate by a preset method when the current random coincidence event count rate is greater than or equal to a minimum value of a random coincidence event count rate range covered by a preset interpolation table.

21. The dead time correction system according to claim 20, characterized in that: The acquisition unit is configured to: acquire detection data of a positron emission tomography device, perform coincidence according to the detection data, and obtain a random coincidence event count rate of each response line subset according to the coincidence result.

22. The dead time correction system according to claim 21, characterized in that: The acquisition unit is configured as follows: Performing matching according to the detection data to obtain a delayed matching data sinusoidal diagram; The random coincidence event count rate of each response line subset is calculated according to the delayed coincidence data sinusoidal diagram and the acquisition time of the corresponding response line subset.

23. The dead time correction system according to claim 22, characterized in that: The acquisition unit calculates the random coincidence event count rate of the i-th response line subset according to the following formula (1): Where T is the acquisition duration, D k is the count on the kth response line in the delay conformance data sinusoid.

24. The dead time correction system according to claim 20, characterized in that: The acquisition unit is further configured to group all the response lines according to position characteristics of the response lines to obtain a plurality of response line subsets before acquiring the random coincidence event count rate of each response line subset.

25. The dead time correction system according to claim 24, characterized in that: The acquisition unit is configured to obtain n × A ring formed by an array of m crystals is defined as a modular ring. When the crystals at both ends of response line a and the crystals at both ends of response line b belong to the same modular ring, response line a and response line b are divided into the same response line subset, wherein n and m are both positive integers, and at least one of n or m is greater than 2.

26. The dead time correction system according to claim 20, characterized in that: The preset interpolation table can be obtained by following the steps below: Collect detection experimental data of radioactive sources at different activities under experimental conditions, and obtain response lines by fitting the data; Divide all response lines into several response line subsets, and obtain the random coincidence event counting rate of each response line subset at each activity; Obtain the instant coincidence event count rate of each response line subset at each activity; Obtaining a true coincidence event detection count rate of each response line subset at each activity according to the instant coincidence event count rate and the random coincidence event count rate; According to the true coincidence event detection count rate under the preset activity, the ideal true coincidence event count rate of each response line subset under each activity is obtained by linear extrapolation; The dead time correction factor of each response line subset at each activity is obtained according to the corresponding true coincidence event detection count rate and the true coincidence event ideal count rate, and the dead time correction factor and the corresponding random coincidence event count rate form the preset interpolation table in a one-to-one correspondence.

27. The dead time correction system according to claim 26, characterized in that: The method of collecting detection experimental data of the radioactive source at different activities under experimental conditions includes: The detection experimental data of the same radioactive source are collected multiple times within a predetermined time, and the factors affecting the predetermined time include the initial activity and decay rate of the radioactive source.

28. The dead time correction system according to claim 26, characterized in that: The true coincidence event detection count rate under the preset activity is regarded as the true coincidence event ideal count rate.

29. The dead time correction system according to claim 26, characterized in that: The calculation of the instant coincidence event count rate of each response line subset at different activities includes: The instant coincidence event count rate of the i-th response line subset at the j-th scan is calculated by the following formula (2): Among them, P kj is the number of instant coincidence events of the kth response line in the i-th response line subset at the jth scan, and T is the acquisition duration.

30. The dead time correction system according to claim 26, characterized in that: The obtaining of the true coincidence event detection count rate of each response line subset at different activities according to the instant coincidence event count rate and the delayed coincidence event count rate comprises: The true coincidence event detection count rate is calculated by the following formula (3): in, represents the instantaneous coincidence event count rate of the i-th response line subset at the j-th scan, represents the random coincidence event count rate of the i-th response line subset at the j-th scan.

31. The dead time correction system according to claim 26, characterized in that: The method of obtaining the ideal true coincidence event count rate of each response line subset at different activities by linearly extrapolating the true coincidence event detection count rate at the preset activity includes: The preset activity is used as the independent variable and the true coincidence event detection count rate is used as the dependent variable. Least squares straight line fitting is performed and linearly extrapolated to the ideal true coincidence event count rate of each response line subset when the activity is greater than the preset activity, wherein the true coincidence event detection count rate of each response line subset is equal to its ideal true coincidence event count rate at the preset activity.

32. The dead time correction system according to claim 31, characterized in that: The preset activity includes several activities less than a specific value, wherein the specific value A j It is calculated by the following formula (4): Where A0 is the initial activity, T 1 / 2 is the half-life of the radioactive source, T j The time from the initial activity measurement to the start of the jth scan, T acq,j is the duration of the j-th scan.

33. The dead time correction system according to claim 26, characterized in that: The dead time correction factor of each response line subset at different activities is obtained according to the corresponding true coincidence event detection count rate and the true coincidence event ideal count rate, including: The dead time correction factor is obtained using the following formula (5): in, is the ideal count rate of true coincidence events of the i-th response line subset at the j-th scan, is the true coincidence event detection count rate of the i-th response line subset at the j-th scan.

34. The dead time correction system according to claim 20, characterized in that: The processing unit is configured to: If the current random coincidence event count rate is within the range of random coincidence event count rates covered by the preset interpolation table, determining whether there is a value in the preset interpolation table that is the same as the current random coincidence event count rate; If so, taking the dead time correction factor corresponding to the value as the dead time correction factor for the corresponding response line subset; If not, obtain the two values ​​R1 and R2 closest to the current random coincidence event count rate in the preset interpolation table and their corresponding dead time correction factors CF1 and CF2, and use the following formula (6) to calculate the dead time correction factor CF of the corresponding response line subset: DT : Among them, R ob is the current random coincidence event count rate.

35. The dead time correction system of claim 20, wherein: The processing unit is configured to: If the current random coincidence event count rate is greater than the random coincidence event count rate range covered by the preset interpolation table, the random coincidence event count rates R3 and R4 of the first and second acquisitions in the preset interpolation table and their corresponding dead time correction factors CF3 and CF4 are obtained, and the dead time correction factor CF of the corresponding response line subset is calculated using the following formula (7): DT : Among them, R ob is the current random coincidence event count rate.

36. The dead time correction system of claim 24, wherein: The acquisition unit is further configured to: The crystal ring composed of a single axial crystal is used as the grouping unit. When the crystals at both ends of the response line are located in the same crystal ring, they are divided into the same response line subset.

37. The dead time correction system of claim 24, wherein: The acquisition unit is further configured to: When the crystals are arranged in a crystal array, the response lines from the same two crystal arrays are grouped together.

38. The dead time correction system of claim 20, wherein: The radiation source includes 18 F. 22 Na, 68 At least one of Ge.

39. An image reconstruction method, characterized in that: include: During image reconstruction, the dead time effect correction is performed on coincident events according to the dead time correction method described in any one of claims 1 to 19.

40. An image reconstruction system, characterized in that: A dead time correction system comprising any one of claims 20 to 38.

41. A positron emission tomography device, characterized in that: The positron emission tomography imaging device comprises the dead time correction system according to any one of claims 20 to 38 or the image reconstruction system according to claim 40; and / or, The positron emission tomography imaging device adopts the dead time correction method described in any one of claims 1 to 19 to perform dead time correction or the image reconstruction method described in claim 39 to perform image reconstruction.

42. A storage medium, characterized in that The storage medium stores a computer program, and the computer program is configured to implement the dead time correction method according to any one of claims 1 to 19 or the image reconstruction method according to claim 39 when executed.

43. A computer program product, characterized in that It comprises a computer program or an instruction, characterized in that when the computer program or the instruction is executed by a processor, it implements the dead time correction method described in any one of claims 1 to 19 or implements the image reconstruction method described in claim 39.

Images