CN103211608A - Calculation method of aperture weighting in spiral CT (computed tomography) image reconstruction based on GPU (graphic processing unit) acceleration - Google Patents

Calculation method of aperture weighting in spiral CT (computed tomography) image reconstruction based on GPU (graphic processing unit) acceleration Download PDF

Info

Publication number
CN103211608A
CN103211608A CN2012100141994A CN201210014199A CN103211608A CN 103211608 A CN103211608 A CN 103211608A CN 2012100141994 A CN2012100141994 A CN 2012100141994A CN 201210014199 A CN201210014199 A CN 201210014199A CN 103211608 A CN103211608 A CN 103211608A
Authority
CN
China
Prior art keywords
aperture
spiral
image reconstruction
numbered
gpu
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN2012100141994A
Other languages
Chinese (zh)
Other versions
CN103211608B (en
Inventor
全国涛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shanghai United Imaging Healthcare Co Ltd
Original Assignee
Shanghai United Imaging Healthcare Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai United Imaging Healthcare Co Ltd filed Critical Shanghai United Imaging Healthcare Co Ltd
Priority to CN201210014199.4A priority Critical patent/CN103211608B/en
Publication of CN103211608A publication Critical patent/CN103211608A/en
Application granted granted Critical
Publication of CN103211608B publication Critical patent/CN103211608B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Apparatus For Radiation Diagnosis (AREA)

Abstract

The invention provides a calculation method of aperture weighting in spiral CT (computed tomography) image reconstruction based on GPU (graphic processing unit) acceleration. The calculation method comprises the following steps of a, determining aperture weighting values of all symmetrical rays of each voxel; b, carrying out the summation of the aperture weighting values of all symmetrical rays of each voxel, and obtaining the aperture weighting value of the voxel; and c, carrying out normalization calculation on the aperture weighting values of all voxels, and obtaining the aperture weighting value after the normalization of each voxel. According to the aperture weighting method, non-smoothness of detection data of an edge detector unit is effectively solved, and a conditional statement is not needed, but the accurate calculation of the aperture weighting and normalized value in multi-layered spiral CT image reconstruction can be realized, thus the quality of a reconstructed image and the efficiency of the reconstruction calculation method are improved.

Description

Computational methods based on aperture weighting in the spiral CT image reconstruction of GPU acceleration
Technical field
The present invention relates to the Medical Image Processing technology, relate in particular to a kind of computational methods based on aperture weighting in the spiral CT image reconstruction of GPU acceleration.
Background technology
The image reconstruction speed of multi-layer spiral CT is a focus of CT technical research always, and more common technology can be divided into three types in the existing accelerated method:
First type for utilizing the CPU cluster to carry out speed-up computation, as the patent No. is to disclose a kind of fast parallel cone beam reconstruction system and method that uses one or more microprocessor CPUs in the Chinese patent of CN02805089.4.But this method is not mentioned the reconstruction model that multi-layer spiral CT is relevant just at cone beam reconstruction, and CPU costs an arm and a leg, and it is too high relatively to reach reasonable acceleration effect cost.
Second type for utilizing this class specific calculations hardware device of FPGA to quicken, as: disclosed method in the article " High Speed CT Image Reconstruction using FPGA " (International Journal of Computer Applications (0975 – 8887) Volume 22 – No.4, May 2011).Yet this method is also just mentioned basic CT reconstruct, does not relate to the multi-layer spiral CT correlation technique, and this specific calculations hardware of FPGA algorithm design well after transplantability not strong, serious to the dependence of hardware.
The third type is quickened for utilizing graphic process unit (GPU).The cost of this mode has reduced with respect to the CPU cluster a lot, and the floating-point disposal ability of existing GPU is far above CPU, and its needs only is a video card, transplantability and compatible all better than this specific calculations hardware device of FPGA, the acceleration that therefore utilizes GPU to carry out the CT restructing algorithm can be issued to reconstruct speed faster in condition cheaply.Patent at this aspect is also many, is that the Chinese patent of CN200810114478.1 discloses a kind of CT parallel reconstructing system and formation method as the patent No., utilizes the GPU cluster to carry out CT reconstruct; The patent No. is the GPU accelerated method that the Chinese patent of CN200810113846.0 discloses a kind of CT image reconstruction, but this two kinds of methods and be not suitable for the multi-layer spiral CT image reconstruction do not comprise the computational problem to the aperture weighting function yet.In addition, disclose a kind of graphic process unit of utilizing though the patent No. is the Chinese patent of CN200910248774.5, realized the method for 3 D back projection, this method is applicable to the multi-layer spiral CT image reconstruction, but the calculating of also not mentioned aperture weighted value, and normalized computational problem.
In the multi-layer spiral CT image reconstruction, the calculating of aperture weighting numerical value all has very important influence to the efficient of restructing algorithm and the quality of reconstructed image.In the Spiral CT scan process, patient moves along being parallel to detector direction always, a mutation process that for several detecting units at detector edge, can exist detectable signal to grow out of nothing, and then influence detects the concordance of data.Therefore need to add an aperture weighting function and come level and smooth this process of growing out of nothing, thereby reduce arc pseudo-shadow, the importance that in article " Weighted FBP-a simple approximate 3D FBP algorithm for multi-slice spiral CT with good dose usage for arbitrary pitch " (Phys. Med. Biol. 49 (2004) 2209 – 2218) the aperture weighting function is calculated has detailed description.But this article just carries out the theory analysis on basis to this weighting function, how not to relate to specific implementation, does not mention that more what use is made of GPU quickens.
Therefore, the necessary computational methods that a kind of aperture weighting in the multi-layer spiral CT image reconstruction that graphic process unit (GPU) is quickened is provided, can solve the unsmooth property of edge detector unit detection data, and do not need conditional statement just can accurately calculate aperture weighting and normalized value in the multi-layer spiral CT image reconstruction, improve the quality of reconstructed image and the efficient of restructing algorithm.
Summary of the invention
The object of the present invention is to provide a kind of computational methods based on aperture weighting in the spiral CT image reconstruction of GPU acceleration, can effectively solve the unsmooth property of edge detector unit detection data, and do not need conditional statement just can accurately calculate aperture weighting and normalized value in the multi-layer spiral CT image reconstruction, thereby improve the quality of reconstructed image and the efficient of restructing algorithm.
For achieving the above object, the present invention adopts following technical scheme: a kind of computational methods based on aperture weighting in the spiral CT image reconstruction of GPU acceleration comprise the steps: that a. determines the aperture weighted value of all symmetrical rays of each voxel
Figure 971668DEST_PATH_IMAGE001
B. to the aperture weighted value of each all symmetrical ray of voxel
Figure 482546DEST_PATH_IMAGE001
Carry out addition, obtain the aperture weighted value of this voxel; C. the aperture weighted value of all voxels is carried out normalization and calculate, obtain the aperture weighted value after each voxel normalization
Figure 874213DEST_PATH_IMAGE002
Further, described aperture weighted value
Figure 10796DEST_PATH_IMAGE001
Be calculated as follows;
Figure 308703DEST_PATH_IMAGE003
（1）
Wherein, Q is the parameter of control smoothness of curve, and q is the coordinate of detector in the Z direction,
Figure 426701DEST_PATH_IMAGE004
Be current projection angle, N is row's number of detector,
Figure 869446DEST_PATH_IMAGE005
Be the coordinate of the voxel after the reconstruct in X, Y, Z direction.
Further, described detector the coordinate q of Z direction by formula (2) calculate:
Figure 996670DEST_PATH_IMAGE006
（2）
Wherein,
Figure 143225DEST_PATH_IMAGE007
Be current projection angle
Figure 245173DEST_PATH_IMAGE004
Under scanning bed physical location on the z direction, t is the coordinate of probe access direction:
Figure 611432DEST_PATH_IMAGE008
, R is the distance of x-ray source and center of rotation,
Figure 778234DEST_PATH_IMAGE009
Be the scanning bed distance that in a circle scanning, moves,
Figure 343076DEST_PATH_IMAGE010
,
Figure 584569DEST_PATH_IMAGE011
Be the value of center of rotation at the pairing port number of channel direction projection.
Further, described spiral CT is a multi-layer spiral CT, and calculating detector is divided into the symmetrical ray of odd-numbered and the symmetrical ray of even-numbered with the symmetrical ray that differs 180 ° or 360 ° in the helical scanning, and calculates respectively when the coordinate q of Z direction.
Further, described even-numbered is the coordinate of the symmetrical ray of n in detector z direction
Figure 172545DEST_PATH_IMAGE012
Calculate by following formula (3), wherein n is the integer more than or equal to 0;
Figure 454622DEST_PATH_IMAGE013
（3）
Initial value
Figure 375435DEST_PATH_IMAGE014
Calculate by following formula (4):
Figure 84765DEST_PATH_IMAGE015
（4）
Figure 160038DEST_PATH_IMAGE016
Be that two neighbouring even-numbered are numbered the spacing between the symmetrical ray, calculate by following formula (5):
Figure 728029DEST_PATH_IMAGE017
（5）。
Further, described odd-numbered is the coordinate of the symmetrical ray of n in detector z direction
Figure 18196DEST_PATH_IMAGE018
Calculate by following formula (6), wherein n is the integer more than or equal to 0;
Figure 147695DEST_PATH_IMAGE019
（6）
Initial value
Figure 211728DEST_PATH_IMAGE020
Calculate by following formula (7):
Figure 835607DEST_PATH_IMAGE021
（7）
Figure 229548DEST_PATH_IMAGE022
Be two spacings between the adjacent odd-numbered symmetry ray, calculate by following formula (8):
Figure 280681DEST_PATH_IMAGE023
（8）。
Further, described even-numbered n scope is
Figure 834940DEST_PATH_IMAGE024
Further, described odd-numbered n scope is
Figure 183882DEST_PATH_IMAGE025
Further, the aperture weighted value after the described normalization
Figure 199373DEST_PATH_IMAGE002
Calculate by following formula (11):
Figure 155828DEST_PATH_IMAGE026
（11）
Wherein, i is a positive integer, and 2n is the quantity of passing these all symmetrical rays of voxel.
Further, the aperture weighted value after the described normalization
Figure 755306DEST_PATH_IMAGE002
Be stored in the shared video memory of GPU.
The present invention contrasts prior art following beneficial effect: the computational methods of aperture weighting in the spiral CT image reconstruction that quickens based on GPU provided by the invention effectively solve the unsmooth property of edge detector unit detection data by the method for aperture weighting.In addition, the present invention is divided into the symmetrical ray that differs 180 ° or 360 ° in the multi-layer spiral CT helical scanning the symmetrical ray separate computations of the symmetrical ray and the even-numbered of odd-numbered, further improve the efficient of restructing algorithm, solve the high flow rate characteristic that GPU handles conditional statement.
Description of drawings
Fig. 1 is the flow chart of steps that the present invention is based on the computational methods of aperture weighting in the spiral CT image reconstruction that GPU quickens.
Fig. 2 is the sketch map that concerns of the coordinate of aperture weighted value in the spiral CT image reconstruction of the present invention and detector Z direction.
Fig. 3 is the arrangement sketch map of symmetrical ray in the multi-layer helical CT scan of the present invention.
Fig. 4 be in the embodiment of the invention multi-layer spiral CT to the reformatted slices figure of actual die body.
The specific embodiment
The invention will be further described below in conjunction with drawings and Examples.
Fig. 1 is the flow chart of steps of the computational methods of aperture weighting in the multi-layer spiral CT image reconstruction of the present invention, and Fig. 2 is the sketch map that concerns of the coordinate of aperture weighted value in the multi-layer spiral CT image reconstruction of the present invention and detector Z direction.
See also Fig. 1, the computational methods based on aperture weighting in the multi-layer spiral CT image reconstruction of GPU acceleration provided by the invention comprise the steps:
At first, in step S101, determine the aperture weighted value of all symmetrical rays of each tissue points by following formula (1)
Figure 406473DEST_PATH_IMAGE001
Figure 525739DEST_PATH_IMAGE003
（1）
The aperture weighted value
Figure 167942DEST_PATH_IMAGE001
For be multiplied by a coefficient on each probe unit of detector Z direction, this coefficient is smaller or equal to 1, and the value of central passage is 1, and the value of edge gateway is the numerical value less than 1.In the above-mentioned formula (1), Q is the parameter of control smoothness of curve, and q is the coordinate of detector in the Z direction,
Figure 739869DEST_PATH_IMAGE004
Be current projection angle, N is row's number of detector, i.e. the number of plies of multi-layer spiral CT,
Figure 994395DEST_PATH_IMAGE005
Be the coordinate of i voxel after the reconstruct in X, Y, Z direction; Detector is in the coordinate q and the aperture weighted value of Z direction
Figure 420697DEST_PATH_IMAGE001
Relation as shown in Figure 2.
In the reconstruct of multi-layer spiral CT 3-D view, through behind the data rearrangement, described detector can calculate by following formula (2) at the coordinate q of Z direction:
Figure 984533DEST_PATH_IMAGE006
（2）
In the above-mentioned formula (2),
Figure 785700DEST_PATH_IMAGE027
Be the coordinate of i voxel at the z axle,
Figure 93184DEST_PATH_IMAGE007
Be current projection angle
Figure 373993DEST_PATH_IMAGE004
Under scanning bed physical location on the z direction, t is the coordinate of probe access direction:
Figure 859463DEST_PATH_IMAGE008
, R is the distance of x-ray source and center of rotation,
Figure 405982DEST_PATH_IMAGE009
Be the scanning bed distance that in a circle scanning, moves,
Figure 500846DEST_PATH_IMAGE010
,
Figure 134696DEST_PATH_IMAGE011
Be the value of center of rotation at the pairing port number of channel direction projection.
Owing to have many symmetrical rays in the helical scanning, and the symmetrical ray that differs 180 ° and 360 ° all there are differences in spacing and initial position, thus symmetrical ray is divided into the symmetrical ray of odd-numbered and the symmetrical ray of even-numbered, and calculate respectively.
Described even-numbered is the coordinate of the symmetrical ray of n in detector z direction
Figure 40335DEST_PATH_IMAGE012
Calculate by following formula (3), wherein n is the integer more than or equal to 0;
（3）
Initial value Calculate by following formula (4):
Figure 526308DEST_PATH_IMAGE015
（4）
Figure 789800DEST_PATH_IMAGE016
Be that two neighbouring even-numbered are numbered the spacing between the symmetrical ray, calculate by following formula (5):
（5）。
Similarly, described odd-numbered is the coordinate of the symmetrical ray of n in detector z direction
Figure 517551DEST_PATH_IMAGE018
Calculate by following formula (6), wherein n is the integer more than or equal to 0:
Figure 174928DEST_PATH_IMAGE019
（6）
Initial value
Figure 609321DEST_PATH_IMAGE020
Calculate by following formula (7):
Figure 368460DEST_PATH_IMAGE021
（7）
Figure 359550DEST_PATH_IMAGE022
Be two spacings between the adjacent odd-numbered symmetry ray, calculate by following formula (8):
Figure 386281DEST_PATH_IMAGE023
（8）。
Scope strange in order to determine, even-numbered n, also the mode according to odd even is carried out in the calculated hole diameters weights, at first analyzes the situation of even-numbered, and the initial value position of supposing even-numbered is No. 0 symmetrical ray, as shown in Figure 3.
Because the detector origin coordinates is 0, the termination coordinate is N, and N is the number of plies of multi-layer spiral CT, therefore can be calculated as follows according to these parameters and obtain the detector original position
Figure 804624DEST_PATH_IMAGE028
Positional value in-2 to+2 these sequences:
Figure 985813DEST_PATH_IMAGE029
（9）
With the detector final position
Figure 764282DEST_PATH_IMAGE030
Positional value in this sequence:
Figure 396252DEST_PATH_IMAGE031
（10）。
Can accurately calculate the numbering of each symmetrical ray thus
Figure 1808DEST_PATH_IMAGE032
, For being not less than
Figure 239071DEST_PATH_IMAGE028
Smallest positive integral,
Figure 756507DEST_PATH_IMAGE034
For being not more than Maximum integer, obtain this value back and just can accurately calculate the symmetrical ray of each even number in the formula (3)
Figure 908320DEST_PATH_IMAGE012
Value without any need for decision operation, only needs twice and adds/subtract and twice division.
In like manner, can calculate odd-numbered symmetry ray
Figure 795636DEST_PATH_IMAGE025
, obtain this value back and just can accurately calculate each odd-numbered symmetry ray in the formula (6)
Figure 402198DEST_PATH_IMAGE018
Value.
Then, in step S102, to the aperture weighted value of all symmetrical rays of each voxel of calculating among the step S101
Figure 582512DEST_PATH_IMAGE001
Carry out addition, obtain the aperture weighted value of this voxel
Figure 274525DEST_PATH_IMAGE035
At last, in step S103,, carry out normalization by following formula (11) and calculate, obtain the aperture weighted value after each tissue points normalization the aperture weighted value of all tissue points of obtaining among the step S102
Figure 634705DEST_PATH_IMAGE002
:
Figure 345041DEST_PATH_IMAGE026
（11）
Wherein, i is a positive integer, and 2n is the quantity of passing these all symmetrical rays of voxel.
Because these aperture weights
Figure 712569DEST_PATH_IMAGE002
Value for the probe unit of all Z directions all is identical, therefore only need be when program begins according to the number of plies of detector Z direction, calculate and once get final product, because each GPU thread all needs to transfer this one piece of data and conducts interviews, so consider the video memory that GPU is dissimilar, the difference of its access efficiency, present embodiment is with the aperture weights that calculate
Figure 642610DEST_PATH_IMAGE002
Be stored in the shared video memory of GPU.Shared video memory is arranged in the core processor of each GPU, belongs to on-chip memory cell, and its access speed is the fastest in a few class GPU video memorys.
To illustrate and verify effect of the present invention with concrete data and reconstructed image quality below.
The data that obtain with the scanning of actual die body are carried out 3-D view reconstruct and are obtained 64 slice maps, a slice map of getting wherein compares, wherein one group of data has not only been carried out normalized for there being the calculated hole diameters weighting, and other one group of data computation aperture weighting and carry out normalized.The parameters that this slice map uses is:
Figure 42367DEST_PATH_IMAGE036
, 64 Layer Detection devices, detector are at the spacing 0.625mm of z direction, and reconstruct visual field size is 210mm, and light source is 570mm to the center of rotation distance, and reconstruct thickness is 0.625, Q-value is 0.8.
Fig. 4 is the reformatted slices figure of multi-layer spiral CT to actual die body, and Fig. 4 a is at the reformatted slices figure that does not have to obtain under the calculated hole diameters weighting condition, the reformatted slices figure of Fig. 4 b for obtaining under calculated hole diameters weighting condition.Fig. 4 c is the differential chart of Fig. 4 a and Fig. 4 b.
As can be seen from Figure 4, there is a little arc pseudo-shadow in the upper left corner of Fig. 4 a, can it is evident that in Fig. 4 c the slice map that two distinct methods obtain has very significant diversity.Pseudo-shadow and not obvious among Fig. 4 a, this be since in the actual data that obtain noise level than higher, covered part owing to do not add the arc pseudo-shadow that the aperture weighted calculation is brought, this part concealed pseudo-shadow still can clearly reflect in the differential chart 4c of two methods.These computational methods that aperture weighting in spiral CT image reconstruction that quickens based on GPU provided by the invention also just has been described have very significant effects for the picture quality of actual multi-layer spiral CT reconstruct.
Use concrete data that the computational efficiency based on aperture weighting in the multi-layer spiral CT reconstruct of GPU acceleration provided by the invention is described below.
The parameter of using among experiment condition and the above-mentioned Fig. 4 reformatted slices figure is identical, the experiment condition of the contrast groups that adopts and GPU code and computational methods provided by the invention are in full accord, and unique difference is that the calculated hole diameters weighting has adopted the judgement statement to calculate in the contrast groups.Shown in the following form of computational efficiency:
The reconstruct number of voxel 512x512x1 512x512x16 512x512x32 512x512x64
Contrast groups (millisecond) 796 5425 10847 21719
The inventive method (millisecond) 677 3081 6153 12378
The computational methods that can clearly find out aperture weighting the spiral CT image reconstruction that quickens based on GPU provided by the invention from last table have improved nearly one times in computational efficiency, and this is because computational methods provided by the invention have been given up fully and judged statement and can accurately calculate the quantity of the needed symmetrical ray of tissue points of each reconstruct and the value of its aperture weighting.
To sum up, the computational methods of aperture weighting in the spiral CT image reconstruction that quickens based on GPU provided by the invention effectively solve the unsmooth property of edge detector unit detection data by the method for aperture weighting.In addition, the present invention is divided into the symmetrical ray that differs 180 ° or 360 ° in the multi-layer spiral CT helical scanning the symmetrical ray separate computations of the symmetrical ray and the even-numbered of odd-numbered, further improve the efficient of restructing algorithm, solve the high flow rate characteristic that GPU handles conditional statement.
Though the present invention discloses as above with preferred embodiment; right its is not in order to qualification the present invention, any those skilled in the art, without departing from the spirit and scope of the present invention; when can doing a little modification and perfect, so protection scope of the present invention is when with being as the criterion that claims were defined.

Claims (10)

1. the computational methods based on aperture weighting in the spiral CT image reconstruction of GPU acceleration is characterized in that, comprise the steps:
A. determine the aperture weighted value of all symmetrical rays of each voxel
B. to the aperture weighted value of each all symmetrical ray of voxel
Figure 275909DEST_PATH_IMAGE001
Carry out addition, obtain the aperture weighted value of this voxel;
C. the aperture weighted value of all voxels is carried out normalization and calculate, obtain the aperture weighted value after each voxel normalization
Figure 76506DEST_PATH_IMAGE002
2. the computational methods based on aperture weighting in the spiral CT image reconstruction of GPU acceleration as claimed in claim 1 is characterized in that described aperture weighted value
Figure 363131DEST_PATH_IMAGE001
Be calculated as follows;
（1）
Wherein, Q is the parameter of control smoothness of curve, and q is the coordinate of detector in the Z direction,
Figure 456781DEST_PATH_IMAGE004
Be current projection angle, N is row's number of detector, Be the coordinate of the voxel after the reconstruct in X, Y, Z direction.
3. the computational methods of aperture weighting is characterized in that in the spiral CT image reconstruction that quickens based on GPU as claimed in claim 2, described detector the coordinate q of Z direction by formula (2) calculate:
Figure 585722DEST_PATH_IMAGE006
（2）
Wherein, Be current projection angle
Figure 894661DEST_PATH_IMAGE004
Under scanning bed physical location on the z direction, t is the coordinate of probe access direction: , R is the distance of x-ray source and center of rotation,
Figure 238890DEST_PATH_IMAGE009
Be the scanning bed distance that in a circle scanning, moves,
Figure 353608DEST_PATH_IMAGE010
,
Figure 589548DEST_PATH_IMAGE011
Be the value of center of rotation at the pairing port number of channel direction projection.
4. the computational methods based on aperture weighting in the spiral CT image reconstruction of GPU acceleration as claimed in claim 3, it is characterized in that, described spiral CT is a multi-layer spiral CT, calculating detector is when the coordinate q of Z direction, the symmetrical ray that differs 180 ° or 360 ° in the helical scanning is divided into the symmetrical ray of odd-numbered and the symmetrical ray of even-numbered, and calculates respectively.
5. the computational methods based on aperture weighting in the spiral CT image reconstruction of GPU acceleration as claimed in claim 4 is characterized in that described even-numbered is the coordinate of the symmetrical ray of n in detector z direction
Figure 38984DEST_PATH_IMAGE012
Calculate by following formula (3), wherein n is the integer more than or equal to 0;
Figure 221835DEST_PATH_IMAGE013
（3）
Initial value
Figure 174747DEST_PATH_IMAGE014
Calculate by following formula (4):
Figure 781922DEST_PATH_IMAGE015
（4）
Figure 453075DEST_PATH_IMAGE016
Be that two neighbouring even-numbered are numbered the spacing between the symmetrical ray, calculate by following formula (5):
Figure 439616DEST_PATH_IMAGE017
（5）。
6. the computational methods based on aperture weighting in the spiral CT image reconstruction of GPU acceleration as claimed in claim 4, it is characterized in that, described odd-numbered is that the symmetrical ray of n calculates by following formula (6) at the coordinate of detector z direction, and wherein n is the integer more than or equal to 0;
Figure 325664DEST_PATH_IMAGE018
（6）
Initial value
Figure 355937DEST_PATH_IMAGE019
Calculate by following formula (7):
Figure 265118DEST_PATH_IMAGE020
（7）
Figure 39039DEST_PATH_IMAGE021
Be two spacings between the adjacent odd-numbered symmetry ray, calculate by following formula (8):
Figure 794241DEST_PATH_IMAGE022
（8）。
7. the computational methods based on aperture weighting in the spiral CT image reconstruction of GPU acceleration as claimed in claim 5 is characterized in that described even-numbered n scope is
Figure 746148DEST_PATH_IMAGE023
8. the computational methods based on aperture weighting in the spiral CT image reconstruction of GPU acceleration as claimed in claim 6 is characterized in that described odd-numbered n scope is
Figure 391893DEST_PATH_IMAGE024
9. as described each the computational methods of claim 1～8, it is characterized in that the aperture weighted value after the described normalization based on aperture weighting in the spiral CT image reconstruction of GPU acceleration
Figure 454658DEST_PATH_IMAGE002
Calculate by following formula (11):
（11）
Wherein, i is a positive integer, and 2n is the quantity of passing these all symmetrical rays of voxel.
10. the computational methods based on aperture weighting in the spiral CT image reconstruction of GPU acceleration as claimed in claim 1 is characterized in that the aperture weighted value after the described normalization Be stored in the shared video memory of GPU.
CN201210014199.4A 2012-01-18 2012-01-18 Computational methods based on the spiral CT image reconstruction mesoporous weighting that GPU accelerates Active CN103211608B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210014199.4A CN103211608B (en) 2012-01-18 2012-01-18 Computational methods based on the spiral CT image reconstruction mesoporous weighting that GPU accelerates

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210014199.4A CN103211608B (en) 2012-01-18 2012-01-18 Computational methods based on the spiral CT image reconstruction mesoporous weighting that GPU accelerates

Publications (2)

Publication Number Publication Date
CN103211608A true CN103211608A (en) 2013-07-24
CN103211608B CN103211608B (en) 2016-08-03

Family

ID=48810023

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210014199.4A Active CN103211608B (en) 2012-01-18 2012-01-18 Computational methods based on the spiral CT image reconstruction mesoporous weighting that GPU accelerates

Country Status (1)

Country Link
CN (1) CN103211608B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109640830A (en) * 2016-07-14 2019-04-16 医视特有限公司 Focus ultrasonic based on precedent
US10482632B2 (en) 2017-04-28 2019-11-19 Uih America, Inc. System and method for image reconstruction
CN115051164A (en) * 2022-06-21 2022-09-13 中山大学 Broadband circularly polarized horn antenna based on accelerating spiral super-elliptic double ridges

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1174364A (en) * 1996-05-02 1998-02-25 西门子公司 Image reproduction method of computer tomograph
CN1489977A (en) * 2002-09-23 2004-04-21 Method and device for arranging images of movable objects for CT
CN1742296A (en) * 2003-02-05 2006-03-01 西门子公司 Voxel-driven spiral reconstruction for cone-beam computer tomography
CN101283913A (en) * 2008-05-30 2008-10-15 首都师范大学 GPU acceleration method of CT image reconstruction
US20090161820A1 (en) * 2006-02-08 2009-06-25 Siemens Aktiengesellschaft Method for increasing the quality of computer tomographic recording series by projection data processing and CT system having a computation unit
CN102110310A (en) * 2009-12-25 2011-06-29 东软飞利浦医疗设备系统有限责任公司 Method for realizing three-dimensional back projection by graphics processor

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1174364A (en) * 1996-05-02 1998-02-25 西门子公司 Image reproduction method of computer tomograph
CN1489977A (en) * 2002-09-23 2004-04-21 Method and device for arranging images of movable objects for CT
CN1742296A (en) * 2003-02-05 2006-03-01 西门子公司 Voxel-driven spiral reconstruction for cone-beam computer tomography
US20090161820A1 (en) * 2006-02-08 2009-06-25 Siemens Aktiengesellschaft Method for increasing the quality of computer tomographic recording series by projection data processing and CT system having a computation unit
CN101283913A (en) * 2008-05-30 2008-10-15 首都师范大学 GPU acceleration method of CT image reconstruction
CN102110310A (en) * 2009-12-25 2011-06-29 东软飞利浦医疗设备系统有限责任公司 Method for realizing three-dimensional back projection by graphics processor

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
JOHAN S ET AL,: "Regularized iterative weighted filtered backprojection for helical cone-beam CT", 《MEDICAL PHYSICS》 *
KARL S ETAL,: "Weighted FBP - a simple approximate 3D FBP algorithm for multi-slice spiral CT with good dose usage for arbitrary pitch", 《PHYS. MED. BIOL.》 *
P KOKEN ET AL,: "Aperture weighted cardiac reconstruction for cone-beam CT", 《PHYS. MED. BIOL.》 *
P KOKEN ET AL,: "Aperture weighted cardiac reconstruction for cone-beam CT", 《PHYS. MED. BIOL.》, 6 July 2006 (2006-07-06) *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109640830A (en) * 2016-07-14 2019-04-16 医视特有限公司 Focus ultrasonic based on precedent
US10482632B2 (en) 2017-04-28 2019-11-19 Uih America, Inc. System and method for image reconstruction
US11062487B2 (en) 2017-04-28 2021-07-13 Shanghai United Imaging Healthcare Co., Ltd. System and method for image reconstruction
US11455756B2 (en) 2017-04-28 2022-09-27 Shanghai United Imaging Healthcare Co., Ltd. System and method for image reconstruction
CN115051164A (en) * 2022-06-21 2022-09-13 中山大学 Broadband circularly polarized horn antenna based on accelerating spiral super-elliptic double ridges
CN115051164B (en) * 2022-06-21 2023-06-27 中山大学 Broadband circular polarization horn antenna based on acceleration spiral super-elliptic double ridges

Also Published As

Publication number Publication date
CN103211608B (en) 2016-08-03

Similar Documents

Publication Publication Date Title
CN103732147B (en) X-ray computed tomograohy apparatus and image rebuilding method
US8873819B2 (en) Method for sorting CT image slices and method for constructing 3D CT image
CN103310458B (en) In conjunction with the medical image elastic registrating method of convex closure coupling and multiple dimensioned hierarchical policy
CN101065771A (en) Planar angular visualization of the bronchial tree
CN108257196A (en) A kind of blood vessel based on cardiac CT image stretches method for reconstructing
CN102831627A (en) PET (positron emission tomography) image reconstruction method based on GPU (graphics processing unit) multi-core parallel processing
Hamill et al. Respiratory‐gated CT as a tool for the simulation of breathing artifacts in PET and PET/CT
CN103211608A (en) Calculation method of aperture weighting in spiral CT (computed tomography) image reconstruction based on GPU (graphic processing unit) acceleration
CN108520542A (en) A kind of matched method for reconstructing of PET/CT data phase
Engelke et al. Computer-assisted detection of pulmonary embolism: performance evaluation in consensus with experienced and inexperienced chest radiologists
CN105488798B (en) SAR image method for measuring similarity based on point set contrast
CN109285176A (en) A kind of cerebral tissue dividing method cut based on regularization figure
CN107403465A (en) City scenarios sectional plan method for reconstructing based on structure priori and deep learning
US20140056499A1 (en) Apparatus and method for generating image using correction model
Carvalho et al. Low-dose, large-angled cone-beam helical CT data reconstruction using algebraic reconstruction techniques
Kohli et al. Compensation for distance-dependent resolution in cardiac-perfusion SPECT: impact on uniformity of wall counts and wall thickness
CN101887591B (en) Cone beam CT (computed tomography) fast reconstruction method based on rectangular bounding box
Guo et al. A computerized scheme for lung nodule detection in multiprojection chest radiography
Pintea et al. Medical image processing: A brief survey and a new theoretical hybrid ACO model
Wang et al. New calculation method for exact length weighting factor in cone‐beam computed tomography
Tavares et al. Lung movement determination in temporal sequences of MR images using Hough transform and interval arithmetics
Ting-ting et al. Research on medical image registration based on QPSO and powell algorithm
Tingting et al. Reconstruction of the attenuated Radon transform in π-scheme short-scan SPECT
CN101529472B (en) Imaging system, imaging method and computer program for imaging a region of interest
Moriconi et al. Application and evaluation of wavelet-based surface reconstruction for contour propagation in radiotherapy

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CP01 Change in the name or title of a patent holder

Address after: Room 1098, building 3, 333 Puhui Road, Jiading Industrial Zone, Jiading District, Shanghai, 201203

Patentee after: Shanghai Lianying Medical Technology Co., Ltd

Address before: Room 1098, building 3, 333 Puhui Road, Jiading Industrial Zone, Jiading District, Shanghai, 201203

Patentee before: SHANGHAI UNITED IMAGING HEALTHCARE Co.,Ltd.

CP01 Change in the name or title of a patent holder
CP02 Change in the address of a patent holder

Address after: 201807 2258 Chengbei Road, Jiading District, Shanghai

Patentee after: Shanghai Lianying Medical Technology Co.,Ltd.

Address before: Room 1098, building 3, 333 Puhui Road, Jiading Industrial Zone, Jiading District, Shanghai, 201203

Patentee before: Shanghai Lianying Medical Technology Co.,Ltd.

CP02 Change in the address of a patent holder