CN105125231A - Method and device for eliminating positron emission tomography (PET) image ring artifacts - Google Patents

Abstract

The invention provides a method and device for eliminating positron emission tomography (PET) image ring artifacts. The method is applied to correcting each conforming line of response (LOR) in PET data. The method comprises the steps: for the LORs, determining two conforming detector modules BLOCK corresponding to the LORs; for each BLOCK, according to the counting rate of the BLOCK, obtaining a BLOCK pulse accumulation correction factor corresponding to the BLOCK under the counting rate; multiplying the BLOCK pulse accumulation correction factors of the two BLOCKs, and obtaining a local pulse accumulation correction factor corresponding to the LORs; and according to the local pulse accumulation correction factor, performing local pulse accumulation correction on the LORs. According to the method, the ring artifacts of PET images are eliminated.

Description

A kind of minimizing technology of PET image ring artifact and device

Technical field

The application relates to medical imaging technology, particularly a kind of minimizing technology of PET image ring artifact and device.

Background technology

Positron emission computed tomography (PositronEmissionTomography, PET) be a kind of utilization to organism (such as, human body) the isotope-labeled compound of inner injection positron radioactivity, and measure their spatial distributions in vivo and the detection technique of time response in vitro, have highly sensitive, accuracy good, the feature of accurate positioning, clinical medicine has higher using value, such as, can be applied to the diagnosis etc. of lesion detection, cerebrovascular disease.PET or PET/CT equipment, by detecting the ray sent from organism, obtains through reconstruction the PET image reflecting organism each tissue metabolism situation.

System is when detecting these rays, affect by factors such as environment residing for the design of detector system geometry, detector crystal type, system, the number of rays of the actual reception of system is not identical with the number of rays that organism is launched, and therefore needs to correct scanning the PET data obtained before carrying out PET image reconstruction.Conventional Data correction comprises random correction, normalization correction, counting loss correction, scatter correction, correction for attenuation etc.But after above-mentioned Data correction, PET image still may exist ring artifact, particularly when high count rate imaging, more may there is ring artifact in PET image.

Summary of the invention

In view of this, the application provides a kind of minimizing technology and device of PET image ring artifact, to eliminate the ring artifact of PET image.

Particularly, the application is achieved by the following technical solution:

First aspect, provides a kind of minimizing technology of PET image ring artifact, and described method is applied to and meets line of response LOR to the every bar in PET data and correct; Described method comprises:

For described LOR, determine that corresponding two of described LOR meet detector module BLOCK;

For each BLOCK, according to the counting rate of described BLOCK, obtain the BLOCK pulse pile-up correction factor that described BLOCK under described counting rate is corresponding;

The BLOCK pulse pile-up correction factor of described two BLOCK is multiplied, obtains local pulse corresponding to described LOR and pile up correction factor;

Pile up correction factor according to described local pulse, local pulse is carried out to described LOR and piles up correction.

Second aspect, provides a kind of removal device of PET image ring artifact, and described application of installation corrects in meeting line of response LOR to the every bar in PET data; Described device comprises:

BLOCK determination module, for for described LOR, determines that corresponding two of described LOR meet detector module BLOCK;

BLOCK factor determination module, for for each BLOCK, according to the counting rate of BLOCK, obtains the BLOCK pulse pile-up correction factor that described BLOCK under described counting rate is corresponding;

LOR factor determination module, for being multiplied by the BLOCK pulse pile-up correction factor of described two BLOCK, obtaining local pulse corresponding to described LOR and piling up correction factor;

LOR correction module, for piling up correction factor according to described local pulse, carrying out local pulse to described LOR and piling up correction.

The minimizing technology of the PET image ring artifact that the application provides and device, by the BLOCK pulse pile-up correction factor according to two BLOCK corresponding to LOR, obtain local pulse corresponding to this LOR and pile up correction factor, in order to carry out pulse pile-up correction to LOR, thus eliminate the ring artifact of PET image.

Accompanying drawing explanation

Fig. 1 is a kind of PET Data correction flow chart shown in the application one exemplary embodiment;

Fig. 2 is a kind of PET sniffer schematic diagram shown in the application one exemplary embodiment;

Fig. 3 is that schematic diagram set up by the BLOCK Averagefactor model shown in the application one exemplary embodiment;

Fig. 4 is that schematic diagram set up by the average single event model of BLOCK shown in the application one exemplary embodiment;

Fig. 5 is that the photon shown in the application one exemplary embodiment is to detection schematic diagram;

Fig. 6 is that the LOR shown in the application one exemplary embodiment records schematic diagram;

Fig. 7 is the flow chart of the minimizing technology of the PET image ring artifact shown in the application one exemplary embodiment;

Fig. 8 is the ring artifact removal effect comparison diagram shown in the application one exemplary embodiment;

Fig. 9 is the control appliance structural representation shown in the application one exemplary embodiment;

Figure 10 is the structural representation of the removal device of a kind of PET image ring artifact shown in the application one exemplary embodiment;

Figure 11 is the structural representation of the removal device of the another kind of PET image ring artifact shown in the application one exemplary embodiment.

Detailed description of the invention

Here will be described exemplary embodiment in detail, its sample table shows in the accompanying drawings.When description below relates to accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawing represents same or analogous key element.Embodiment described in following exemplary embodiment does not represent all embodiments consistent with the application.On the contrary, they only with as in appended claims describe in detail, the example of apparatus and method that some aspects of the application are consistent.

PET will launch the radionuclide of positron (as F-18 etc., can positron radionuclide be called for short) labelling is on the compound that can participate in bio-tissue blood flow or metabolic process, this compound can be called tracer, and by this tracer injection in organisms, allow organism carry out PET video picture within the scope of the effective field of view of PET.In this process, positron radionuclide in tracer can discharge positron e+, the positron e+ discharged moves after a segment distance in vivo, can bury in oblivion with the negatron e-in surrounding, produce the γ photon of a pair energy equal (511KeV), the direction of propagation (about 180 degree) on the contrary.Utilize the sniffer of PET system, this γ photon pair can be detected, and then analyze the existence of positron, and rebuild the PET image of reflection organism each tissue metabolism situation, obtain tracer by inspection organism in CONCENTRATION DISTRIBUTION, doctor can judge the focus of the diseases such as cancer accordingly.

In order to obtain image clearly, needed to correct scanning the PET data obtained before carrying out PET image reconstruction, conventional Data correction comprises random correction, normalization correction, counting loss correction, scatter correction, correction for attenuation etc.And the method that the disclosure provides also be applied to image reconstruction before the PET Data correction stage, for eliminating the ring artifact of PET image rebuild and obtain.Because ring artifact is mainly owing to there occurs pulse pile-up phenomenon (because pulse pile-up may cause counting loss on the one hand when detecting γ photon pair, counting position mistake may be caused on the other hand, therefore easily in PET image, there is ring artifact), therefore the correction of this elimination ring artifact is called " local pulse is piled up and corrected " by the disclosure.See the example of Fig. 1, can see, after collection clinical data, in the process of the PET Data correction before image reconstruction, the disclosure needs to pile up through above-mentioned local pulse to correct.

The minimizing technology of following explanation PET image ring artifact of the present disclosure, the method has related to service acquisition phase and the clinical scanning stage of PET.Wherein, service acquisition phase mainly carries out system calibration and trimming process, such as normalization correction, counting loss correction etc.By the correction data that service acquisition phase collects, come the relevant calibration of computing system and correction factor.The factor pair clinical scan data mainly obtained by service acquisition phase in the clinical scanning stage carries out calibrating and correcting, and finally obtains desirable PET image.Specific to the disclosure, local pulse is carried out for PET data and piles up timing, will use " local pulse accumulation correction factor ", and the determination of this factor needs to use some parameters of serving acquisition phase and obtaining.Therefore, in the following description, first will be described in the parameter how service acquisition phase is determined to need to use, then be described in the clinical scanning stage and how to utilize the local pulse that obtains to pile up correction factor to correct.

Service acquisition phase: need the pulse pile-up correction factor determining that each detector module Block is corresponding respectively in this stage, namely determine each " Block pulse pile-up correction factor ".

As shown in Figure 2, simply illustrate in PET system for detecting the right sniffer of γ photon.As shown in Figure 2, the sniffer 200 of PET system generally comprises the multiple gauging rings 20 along axis arrangement, and each gauging ring 20 is assembled by multiple detector module 21, and this detector module is " Block " in the disclosure.Each detector module 21 can be made up of scintillation crystal and photomultiplier tube, scintillation crystal can absorb γ photon, and the light photon of some is produced according to the energy of γ photon, the visible light signal that scintillation crystal produces then is converted into the signal of telecommunication and exports by photomultiplier tube, such as, be converted to pulse and export.Above-mentionedly detect the event that a γ photon incides a scintillation crystal and can be referred to as " single event ".

For single Block, present disclose provides " Block Averagefactor model ", for calculating a parameter " Averagefactor BlockAvgFactor " corresponding to each Block according to this model, additionally provide " the average single event model of Block ", for calculating another parameter corresponding to each Block " average single event BlockAvgSingle " according to this model.Again according to these two parameters of BlockAvgFactor and BlockAvgSingle, calculate " the Block pulse pile-up correction factor " of corresponding Block.

See the example of Fig. 3, Block Averagefactor model is set up according to regular correction factor.Composition graphs 1 also can be seen, before local pulse piles up correction, PET system can be carried out normalization in advance and be corrected, and in the process that normalization corrects, each regular correction factor is calculated according to the regular correction data collected, the CBN shown in following formula (1) (Component-basednormalization, the normalization based on factorization method corrects) factorization method model such as can be used to calculate each correction factor.Calculate the process of regular correction factor according to regular correction data, can perform in a conventional manner, no longer describe in detail.

NC _uivj=ε _uiε _vjb _ub _vc _uimodDc _vjmodDd _uvrkf _uvg _uvr... ... .... formula (1)

Wherein, ε _uiand ε _vjdetector crystal efficiency factor (comprise crystal intrinsic efficiency part and the dead time affects part), b _uand b _vthe axial segments side factor, c _uimodDand c _vjmodDthe cross section block side factor, d _uvrkcrystal interference factor, f _uvaxial geometrical factor, g _uvrbe radial geometric factor, D is block detector cross section crystal number.U and v represents the ring at two detector crystal places respectively, i and j represents two positions of detector crystal in the ring of place that LOR is corresponding respectively, r represents LOR (LineOfResponse, line of response) radial position, and k represents the relative position of LOR in Block detector.

And the disclosure calculates regular correction factor under certain counting rate, and by the regular correction factor of use part, set up Block Averagefactor model, namely by the partial parameters for calculating regular correction factor, such as, Summing Factor cross section block side, the axial segments side factor, sets up Block Averagefactor model.Such as, Summing Factor cross section block side, the axial segments side factor obtained during normalization can be used to correct, the above-mentioned axial segments side Summing Factor cross section block side factor corresponding to the crystal of same position in all Block is sued for peace and is normalized, the following formula of Modling model (2):

${\mathrm{BlockAvgFactor}}_{a,t}=\frac{1}{Norm}\underset{a=u\mathrm{mod}M,t=i\mathrm{mod}D}{\Σ}{b}_u{c}_{ui\mathrm{mod}D}$ ... ... formula (2)

Wherein, a is crystal axial location mark in Block, and t is crystal cross-section location mark in Block, M is the axial crystal number of Block, and D is Block cross section crystal number, and Norm is normalization coefficient, such as, this Norm can be by all BlockAvgFactor _a,taverage, to ensure that BlockAvgFactor average is for 1; Mod in formula (2) represents modulo operation, such as, and 6mod5=1.In addition, in above-mentioned formula (2), the axial segments side factor is with b _ufor example, the cross section block side factor employs c _uimodD, but be not limited thereto in concrete enforcement, such as, can also b be used _vreplace b _u, and use c _vjmodDreplace c _uimodD.

As above be the foundation of Block Averagefactor model, according to this model, the parameter BlockAvgFactor of each Block can be obtained.

Fig. 4 illustrates the foundation of the average single event model of Block, and the average single event model of Block is that the counting loss correction data obtained according to service acquisition phase obtains.

Such as, the counting loss correction data under multiple counting rate can be gathered by high activity decayed source model, and set up the average single event model of Block according to this counting loss correction data.This counting loss correction data can for the single event data (single event here refers to the single photon counting that crystal receives) directly gathered, or, can also be carry out to meeting event the single event data that inverse process obtains.Block average single event model comprises many group factors, every group factor obtains under same counting rate (same to single pass), namely be the crystal count summation on position same in all Block be also finally normalized, following formula (3) is the average single event model of Block:

${\mathrm{BlockAvgSingle}}_{a,t}^c=\frac{1}{Norm}\underset{a=u\mathrm{mod}M,t=i\mathrm{mod}D}{\Σ}{S}_{ui}^c$ ... ... .... formula (3)

Wherein, a is crystal axial location mark in Block, and t is crystal cross-section location mark in Block, c is counting rate, and M is the axial crystal number of Block, and D is Block cross section crystal number, Norm is normalization coefficient, ensures that BlockAvgSingle average is 1, S _uithat system directly gathers single event data or meets the single event data that event inverse process obtains.

According to above-mentioned formula (2) and formula (3), these two parameters of BlockAvgFactor and BlockAvgSingle can be calculated, on this basis, BlockAvgFactor and BlockAvgSingle can be calculated with certain functional relationship, obtain Block pulse pile-up correction factor.Shown in following formula (4):

${\mathrm{BlockPileupFactor}}_{a,t}^c=F({\mathrm{BlockAvgSingle}}_{a,t}^c,{\mathrm{BlockAvgFactor}}_{a,t})$

... ... .... formula (4)

Wherein, be the Block pulse pile-up correction factor under certain counting rate c, F is certain functional relationship.Such as, Block pulse pile-up correction factor can be the ratio of Block Averagefactor model and the average single event model of Block, such as, and F (A, B)=A/B.

In addition, it should be noted that, in example of the present disclosure, can by the calculating of formula (4), be arranged on service acquisition phase to complete, so, in service acquisition phase, each Block obtained in detector is distinguished corresponding Block pulse pile-up correction factor, follow-up directly searching in the clinical scanning stage uses this Block pulse pile-up correction factor.Optionally, also the calculating of formula (4) can be placed on the execution of clinical scanning stage, can set flexibly in force.

The clinical scanning stage: need to determine that every bar meets two BLOCK corresponding to line of response LOR respectively in this stage, and according to " the Block pulse pile-up correction factor " of these two BLOCK, obtain local pulse corresponding to this LOR and pile up correction factor, corrected by this factor pair LOR.Wherein, line of response LOR: tracer positron annihilation can occur in vivo, the gamma photons being mutually 180 degree for a pair can be produced simultaneously, this can be received by two crystal gamma photons simultaneously, and these two crystal can determine a line, be line of response (LOR), when these two crystal receive gamma photons simultaneously, be determined and there occurs a positron annihilation on this line, so on this line, bury in oblivion number of times increase once.

Wherein, Fig. 5 illustrates and meets line of response (LineofResponse, LOR) relation and between above-mentioned Block, as shown in Figure 5, multiple gauging ring 20 forms an inner space, also show an organism 30 be in the inner space that gauging ring forms, the γ photon that the positron annihilation events 31 occurred in this inner space produces is to 32, incide along contrary direction on a pair detection block module Block21, detected detector module 21 by this, then this can be called LOR to the connecting line between Block.The PET data gathered when clinical scanning, meet line LOR by each bar detected and record, and as the example of Fig. 6, have recorded many LOR50.

In example of the present disclosure, clinical data is being carried out to local pulse accumulation timing, namely will correct every bar LOR, Fig. 7 illustrates the flow process of the minimizing technology of PET image ring artifact of the present disclosure, and this flow process can be applied to every bar LOR:

In step 701, for every bar LOR, determine that corresponding two of this LOR meet detector module BLOCK.This LOR, when recording LOR, can be that the line between which two Block carries out record, so easily can obtain two Block corresponding to LOR by PET system.

After obtaining two Block corresponding to LOR, in a step 702, for each Block, obtain the Block pulse pile-up correction factor BlockPileupFactor that this Block is corresponding.More specifically, what LOR was corresponding is two crystal, these two crystal belong to two Block respectively, when obtaining Block pulse pile-up correction factor BlockPileupFactor, and according to the counting rate (average single event counting rate) of the Block at the position of crystal and crystal place.

Shown in formula (4), such as, Block pulse pile-up correction factor BlockPileupFactor calculates according to BlockAvgSingle and BlockAvgFactor, wherein, according to formula (2) and formula (3), each crystal in Block is a corresponding factor respectively, and BlockAvgFactor can have nothing to do with counting rate, can be the factor corresponding with " crystal that LOR is corresponding "; And BlockAvgSingle can according to the counting rate c of Block, search should the BlockAvgSingle of counting rate c, namely corresponding with " counting rate of the Block at the crystal that LOR is corresponding+crystal place " this two factors factor.After obtaining BlockAvgSingle and BlockAvgFactor, these two factors can be calculated Block pulse pile-up correction factor BlockPileupFactor according to certain function, such as, can using the ratio of BlockAvgSingle/BlockAvgFactor as Block pulse pile-up correction factor BlockPileupFactor.

In step 703, after the Block pulse pile-up correction factor BlockPileupFactor obtaining two Block corresponding to LOR respectively, according to the Block pulse pile-up correction factor BlockPileupFactor of these two Block, local pulse corresponding to this LOR can be obtained and pile up correction factor.Such as, can calculate according to following formula (5):

${\mathrm{LORPileupFactor}}_{uivj}={\mathrm{BlockPileupFactor}}_{u\mathrm{mod}eM,i\mathrm{mod}eD}^{C_A}*{\mathrm{BlockPileupFactor}}_{v\mathrm{mod}eM,j\mathrm{mod}eD}^{C_B}$

... ... .... formula (5)

Wherein, C _aand C _bthe counting rate of two Block that LOR is corresponding, LORPileupFactor _uivjthat the local pulse that this LOR is corresponding piles up correction factor.

In step 704, pile up correction factor according to the local pulse obtained in step 703, local pulse is carried out to described LOR and piles up correction.Such as, at timing, can be pile up by local pulse the data that correction factor is multiplied by LOR record.

The minimizing technology of the PET image ring artifact in disclosure example, Block pulse pile-up correction factor is calculated by the Block corresponding to LOR, and then obtain local pulse accumulation correction factor corresponding to LOR, LOR is corrected, the problem of pulse pile-up can be solved, remove the ring artifact of PET image, improve the quality of PET image; Further, in the calculating of correction factor, by only to the part normalization correction factor modeling under certain counting rate, complexity that normalization factor modeling under multiple counting rate is caused and computation time is avoided; By setting up functional relationship to Block Averagefactor model and the average single event model of Block, after can avoiding upgrading regular correction factor, local pulse is piled up process and is upgraded mismatch problem between rear regular correction process, and when being convenient to serve, normalization corrects regular update.

After using method of the present disclosure to correct PET data, through experimental verification, effect is better, shown in Figure 8, the left side of Fig. 8 is for before correction, and right side is for after correction, comparison diagram before and after correcting can obviously be seen, the removal effect of the ring artifact of PET image is better.

For scanning PET (or PET/CT) scanning device obtaining PET image, a PET system can be formed by multiple equipment such as scanning gantry, examinating couch, computer system, operating consoles.Wherein, the inside of scanning gantry is provided with the detector rings for scanning.And the PET Data correction of the removal of PET image ring artifact of the present disclosure, be for scanning collection data after data processing stage, such as, can be performed by the data processing software installed in computer systems, which.As shown in Figure 9, method of the present disclosure can be performed by control appliance 91, and this control appliance 91 can comprise processor 910, communication interface 920, memorizer 930, bus 940.Processor 910, communication interface 920, memorizer 930 completes mutual communication by bus 940.

Wherein, can store the removal logical order of PET image ring artifact in memorizer 930, this memorizer can be such as nonvolatile memory (non-volatilememory).Processor 910 can call the removal logical order of the PET image ring artifact in execute store 930, to perform the minimizing technology of above-mentioned PET image ring artifact.

If the function of the removal logical order of PET image ring artifact using the form of SFU software functional unit realize and as independently production marketing or use time, can be stored in a computer read/write memory medium.Based on such understanding, the part of the part that technical scheme of the present disclosure contributes to prior art in essence in other words or this technical scheme can embody with the form of software product, this computer software product is stored in a storage medium, comprising some instructions in order to make a computer equipment (can be personal computer, server, or the network equipment etc.) perform all or part of step of method described in each embodiment of the present invention.And aforesaid storage medium comprises: USB flash disk, portable hard drive, read only memory (ROM, Read-OnlyMemory), random access memory (RAM, RandomAccessMemory), magnetic disc or CD etc. various can be program code stored medium.

The removal logical order of above-mentioned PET image ring artifact, can be called " removal device of PET image ring artifact ", this device can be divided into each functional module.As shown in Figure 10, this device can comprise: BLOCK determination module 1001, BLOCK factor determination module 1002, LOR factor determination module 1003 and LOR correction module 1004.

BLOCK determination module 1001, for for described LOR, determines that corresponding two of described LOR meet detector module BLOCK;

BLOCK factor determination module 1002, for for each BLOCK, according to the counting rate of described BLOCK, obtains the BLOCK pulse pile-up correction factor that described BLOCK under described counting rate is corresponding;

LOR factor determination module 1003, for being multiplied by the BLOCK pulse pile-up correction factor of described two BLOCK, obtaining local pulse corresponding to described LOR and piling up correction factor;

LOR correction module 1004, for piling up correction factor according to described local pulse, carrying out local pulse to described LOR and piling up correction.

Further, BLOCK factor determination module 1002, for calculating described BLOCK pulse pile-up correction factor according to BLOCK Averagefactor model corresponding to described BLOCK and BLOCK average single event model.

As shown in figure 11, BLOCK factor determination module 1002 can comprise: Averagefactor submodule 1101, average single event submodule 1102 and integrated treatment submodule 1103.

Averagefactor submodule 1101, under a counting rate, according to Summing Factor cross section block side, the axial segments side factor in regular correction factor, sets up BLOCK Averagefactor model.

Average single event submodule 1102, for according to the counting loss correction data under multiple counting rate, sets up the average single event model of BLOCK.

Integrated treatment submodule 1103, for according to described BLOCK Averagefactor model and the average single event model of BLOCK, calculates described BLOCK pulse pile-up correction factor.

Such as, integrated treatment submodule 1103, for using the ratio of described BLOCK Averagefactor model and the average single event model of BLOCK as BLOCK pulse pile-up correction factor.

Further, counting loss correction data, comprising: the single event data of collection, or, carry out to meeting event the single event data that inverse process obtains.

The removal device of the PET image ring artifact in disclosure example, Block pulse pile-up correction factor is calculated by the Block corresponding to LOR, and then obtain local pulse accumulation correction factor corresponding to LOR, LOR is corrected, the problem of pulse pile-up can be solved, remove the ring artifact of PET image, improve the quality of PET image.

The foregoing is only the preferred embodiment of the application, not in order to limit the application, within all spirit in the application and principle, any amendment made, equivalent replacements, improvement etc., all should be included within scope that the application protects.

Claims (11)

1. a minimizing technology for PET image ring artifact, is characterized in that, described method is applied to and meets line of response LOR to the every bar in PET data and correct; Described method comprises:

For described LOR, determine that corresponding two of described LOR meet detector module BLOCK;

For each BLOCK, according to the counting rate of described BLOCK, obtain the BLOCK pulse pile-up correction factor that described BLOCK under described counting rate is corresponding;

The BLOCK pulse pile-up correction factor of described two BLOCK is multiplied, obtains local pulse corresponding to described LOR and pile up correction factor;

Pile up correction factor according to described local pulse, local pulse is carried out to described LOR and piles up correction.

2. method according to claim 1, is characterized in that, described BLOCK pulse pile-up correction factor calculates according to BLOCK Averagefactor model corresponding to described BLOCK and the average single event model of BLOCK.

3. method according to claim 2, is characterized in that, described BLOCK pulse pile-up correction factor is the ratio of described BLOCK Averagefactor model and the average single event model of BLOCK.

4. method according to claim 2, is characterized in that, described BLOCK Averagefactor model is under a counting rate, sets up according to Summing Factor cross section block side, the axial segments side factor in regular correction factor.

5. method according to claim 2, is characterized in that, the average single event model of described BLOCK, is to set up according to the counting loss correction data under multiple counting rate.

6. method according to claim 5, is characterized in that,

Described counting loss correction data, comprising: the single event data of collection, or, carry out to meeting event the single event data that inverse process obtains.

7. a removal device for PET image ring artifact, is characterized in that, described application of installation corrects in meeting line of response LOR to the every bar in PET data; Described device comprises:

BLOCK determination module, for for described LOR, determines that corresponding two of described LOR meet detector module BLOCK;

BLOCK factor determination module, for for each BLOCK, according to the counting rate of described BLOCK, obtains the BLOCK pulse pile-up correction factor that described BLOCK under described counting rate is corresponding;

LOR factor determination module, for being multiplied by the BLOCK pulse pile-up correction factor of described two BLOCK, obtaining local pulse corresponding to described LOR and piling up correction factor;

LOR correction module, for piling up correction factor according to described local pulse, carrying out local pulse to described LOR and piling up correction.

8. device according to claim 7, is characterized in that,

Described BLOCK factor determination module, for calculating described BLOCK pulse pile-up correction factor according to BLOCK Averagefactor model corresponding to described BLOCK and BLOCK average single event model.

9. device according to claim 8, is characterized in that, described BLOCK factor determination module, comprising:

Averagefactor submodule, under a counting rate, according to Summing Factor cross section block side, the axial segments side factor in regular correction factor, sets up described BLOCK Averagefactor model;

Average single event submodule, for according to the counting loss correction data under multiple counting rate, sets up the average single event model of described BLOCK;

Integrated treatment submodule, for according to described BLOCK Averagefactor model and the average single event model of BLOCK, calculates described BLOCK pulse pile-up correction factor.

10. device according to claim 9, is characterized in that,

Described integrated treatment submodule, for using the ratio of described BLOCK Averagefactor model and the average single event model of BLOCK as BLOCK pulse pile-up correction factor.

11. devices according to claim 9, is characterized in that,

Described counting loss correction data, comprising: the single event data of collection, or, carry out to meeting event the single event data that inverse process obtains.