CN109741411A - Low dosage PET image reconstruction method, device, equipment and medium based on gradient field - Google Patents
Low dosage PET image reconstruction method, device, equipment and medium based on gradient field Download PDFInfo
- Publication number
- CN109741411A CN109741411A CN201811525022.4A CN201811525022A CN109741411A CN 109741411 A CN109741411 A CN 109741411A CN 201811525022 A CN201811525022 A CN 201811525022A CN 109741411 A CN109741411 A CN 109741411A
- Authority
- CN
- China
- Prior art keywords
- image
- reconstruction
- pet
- gradient
- pet image
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Nuclear Medicine (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The present invention is applicable in medicine PET technical field of imaging, provide a kind of low dosage PET image reconstruction method based on gradient field, device, equipment and medium, this method comprises: according to the collected data for projection of PET device and the sytem matrix of the PET device, image reconstruction is carried out to the PET image to be reconstructed initialized in advance by PET image reconstruction algorithm, obtain original reconstruction PET image, according to original reconstruction PET image, the image reconstruction equation constructed in advance and the gradient field characteristics of image constructed in advance are chosen by equation using Lagrange multiplier and carry out combined optimization solution, obtain the corresponding Object reconstruction PET image of original reconstruction PET image, to improve the reconstruction speed of low dosage PET image, and reduce the artifact degree of reconstruction image, and then improve low dosage PET figure As the picture quality rebuild.
Description
Technical field
The invention belongs to medicine PET technical field of imaging more particularly to a kind of low dosage PET image weights based on gradient field
Construction method, device, equipment and medium.
Background technique
Positron emission tomography (Positron Emission Tomography, abbreviation PET) is a kind of emission type
Imaging technique (Emission Tomography, abbreviation ET), it is shown not by the method that radiopharmaceutical is injected in vivo
With the metabolic situation of tissue.PET technology be after computer tomography (Computed Tomography, abbreviation CT) and
It is applied to a kind of clinical New video skill after magnetic resonance imaging (Magnetic Resonance Imaging, abbreviation MRI)
Art, PET technology are shown in the fields such as oncology, cardiovascular disease, the nervous system disease research and new drug development research
Brilliant performance out.
In PET imaging, radiopharmaceutical is actually a molecular vehicle, it depends on specific physiological tissue or pathology
Process.Radioactive substance purposive distribution in human body under the leading of drug.The purpose of PET imaging is actually to obtain
In the distribution map of inside of human body, its working principle is radioactive substance: by some radioactive nucleus elements, such as O-15, C-11, N-
Then the label such as 13 and F-18 is inputted by modes such as arm vein injections and is examined on the compound needed for body metabolism
In person's body.During participating in metabolism in vivo, radioactive nucleus element decays labeled compound, releases positive electron (band
The electronics of one positive charge), positive electron is buried in oblivion with surrounding (negatively charged) electronics, and generating two energy is 511keV's
Gammaphoton.This point-blank projects photon in the opposite direction, can be detected using external gamma camera
All photons of specific region radiation, then design certain algorithm, so that it may which approximation obtains radioactive substance in inside of human body
Distribution situation.
The radiopharmaceutical as used in being checked in PET can the personnel to the close contact drug generate radiation, and
The probability for being suffered from cancer by the personnel radiated can be much higher than normal person, while the consumption of radiopharmaceutical is in the cost that PET is checked
Occupy certain weight proportion.Therefore, according to International Commission for Radiological Protection (International Commission on
Radiological Protection, abbreviation ICRP) propose reasonable employment low dosage (As Low As Reasonably
Achievable, abbreviation ALARA) principle, in PET clinical diagnosis, to meet clinical demand with the acquisition of the smallest dosage
Image reduces the dose of radiation to patient as far as possible.
However, when the measurement data sampled to low dosage carries out PET image reconstruction, existing traditional PET image
The speed of algorithm for reconstructing reconstruction image is slow, so that reconstruction image generates motion artifacts, these artifacts will will have a direct impact on doctor
Raw Diagnosis behavior.
Summary of the invention
The low dosage PET image reconstruction method that the purpose of the present invention is to provide a kind of based on gradient field, device, equipment and
Medium, it is intended to solve that a kind of effective low dosage PET image reconstruction method can not be provided due to the prior art, lead to low dosage
PET image reconstruction speed is slow and the problem of reconstructed image quality difference.
On the one hand, the low dosage PET image reconstruction method based on gradient field that the present invention provides a kind of, the method includes
Following step:
When receiving the reconstruction request of low dosage PET image, obtain through the collected data for projection of PET device, and
Obtain the sytem matrix of the PET device；
According to the data for projection and the sytem matrix, by preset PET image reconstruction algorithm to initial in advance
The PET image to be reconstructed changed carries out image reconstruction, obtains original reconstruction PET image；
According to the original reconstruction PET image, using Lagrange multiplier by the image reconstruction equation constructed in advance and in advance
The gradient field characteristics of image first constructed chooses equation and carries out combined optimization solution, and it is corresponding to obtain the original reconstruction PET image
Object reconstruction PET image.
On the other hand, the present invention provides a kind of low dosage PET image reconstruction device based on gradient field, described device packet
It includes:
Parameter acquiring unit, for when receiving the reconstruction request of low dosage PET image, acquisition to be adopted by PET device
The data for projection collected, and obtain the sytem matrix of the PET device；
Original reconstruction unit, for passing through preset PET image weight according to the data for projection and the sytem matrix
It builds algorithm and image reconstruction is carried out to the PET image to be reconstructed initialized in advance, obtain original reconstruction PET image；And
Reconstruction image obtaining unit, for according to the original reconstruction PET image, using Lagrange multiplier by preparatory structure
The image reconstruction equation built and the gradient field characteristics of image constructed in advance choose equation and carry out combined optimization solution, obtain described first
Starting weight builds the corresponding Object reconstruction PET image of PET image.
On the other hand, the present invention also provides a kind of calculating equipment, including memory, processor and it is stored in described deposit
In reservoir and the computer program that can run on the processor, the processor are realized such as when executing the computer program
Step described in the above-mentioned low dosage PET image reconstruction method based on gradient field.
On the other hand, the present invention also provides a kind of computer readable storage medium, the computer readable storage mediums
It is stored with computer program, such as the above-mentioned low dosage PET based on gradient field is realized when the computer program is executed by processor
Step described in image rebuilding method.
The present invention passes through PET image according to the collected data for projection of PET device and the sytem matrix of the PET device
Algorithm for reconstructing carries out image reconstruction to the PET image to be reconstructed initialized in advance, original reconstruction PET image is obtained, according to initial
PET image is rebuild, it is using Lagrange multiplier that the image reconstruction equation constructed in advance and the gradient area image constructed in advance is special
Sign chooses equation and carries out combined optimization solution, the corresponding Object reconstruction PET image of original reconstruction PET image is obtained, to improve
The reconstruction speed of low dosage PET image, and the artifact degree of reconstruction image is reduced, and then improve low dosage PET image weight
The picture quality built.
Detailed description of the invention
Fig. 1 is the implementation process for the low dosage PET image reconstruction method based on gradient field that the embodiment of the present invention one provides
Figure；
Fig. 2 is to be iterated solution to Lagrange's equation using Bregman alternative manner in the embodiment of the present invention one
Implementation flow chart；
Fig. 3 is the structural representation of the low dosage PET image reconstruction device provided by Embodiment 2 of the present invention based on gradient field
Figure；
Fig. 4 is the preferred structure of the low dosage PET image reconstruction device provided by Embodiment 2 of the present invention based on gradient field
Schematic diagram；
Fig. 5 is the another preferred of the low dosage PET image reconstruction device provided by Embodiment 2 of the present invention based on gradient field
Structural schematic diagram；And
Fig. 6 is the structural schematic diagram for the calculating equipment that the embodiment of the present invention three provides.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
Specific implementation of the invention is described in detail below in conjunction with specific embodiment:
Embodiment one:
Fig. 1 shows the realization of the low dosage PET image reconstruction method based on gradient field of the offer of the embodiment of the present invention one
Process, for ease of description, only parts related to embodiments of the present invention are shown, and details are as follows:
In step s101, when receiving the reconstruction request of low dosage PET image, acquisition is collected by PET device
Data for projection, and obtain the sytem matrix of PET device.
The embodiment of the present invention is suitable for Medical Image Processing platform, system or equipment, such as personal computer, server
Deng.When receiving the request rebuild to low dosage PET image, obtains and acquired under the conditions of low dosage by PET device
The undersampling projection data arrived, and the sytem matrix of PET device is obtained, which is the geometry according to PET device
What information was calculated.
In step s 102, according to data for projection and sytem matrix, by preset PET image reconstruction algorithm to preparatory
The PET image to be reconstructed of initialization carries out image reconstruction, obtains original reconstruction PET image.
In embodiments of the present invention, according to data for projection and sytem matrix, pass through preset PET image reconstruction algorithm pair
The PET image to be reconstructed initialized in advance carries out the iterative operation of preset times, to carry out image weight to PET image to be reconstructed
It builds, obtains original reconstruction PET image, wherein PET image to be reconstructed is two dimensional image, and preset PET image reconstruction algorithm is most
Maximum-likelihood it is expected maximum algorithm (Maximum Likelihood Expectation Maximized, abbreviation MLEM) or orderly
Subset expectation maximization algorithm (Ordered Subset Expectation Maximization, abbreviation OSEM) or maximum
Posterior probability algorithm (Maximum A Posterior, MAP).
When initializing PET image to be reconstructed, as illustratively, the pixel value of PET image to be reconstructed is initialized to
Zero.
In step s 103, according to original reconstruction PET image, the image reconstruction that will be constructed in advance using Lagrange multiplier
Equation and the gradient field characteristics of image constructed in advance choose equation and carry out combined optimization solution, obtain original reconstruction PET image pair
The Object reconstruction PET image answered.
In embodiments of the present invention, the image reconstruction equation y that will be constructed in advance using Lagrange multiplieru=GuM and in advance
The gradient field characteristics of image of building chooses equationIt carries out
Simultaneous obtains corresponding Lagrange's equation The Lagrange's equation is optimized again, is finally obtained initial
Rebuild the corresponding Object reconstruction PET image of PET image, wherein yuFor data for projection, GuFor sytem matrix, m is PET to be reconstructed figure
As (namely Object reconstruction PET image), v1For preset weight parameter, RlMatrix is extracted for image block, i.e., according to the RlFrom gradient
L image block is extracted in image ω, D is the eigenmatrix of gradient image ω, αlFor first extracted from gradient image ω
The corresponding feature vector of image block, ω(i)Indicate the corresponding horizontal/vertical gradient image of original reconstruction PET image, i ∈ { 1,2 }
Indicate the direction (horizontal/vertical) of the gradient image ω, L indicates the control coefrficient to the degree of rarefication of feature vector.
In embodiments of the present invention, it is preferable that 5 are set by the control coefrficient L of feature vector degree of rarefication, thus preferably
Reduce the noise of the PET image of the eigenmatrix and feature vector rarefaction representation by learning.
Using Lagrange multiplier by the image reconstruction equation constructed in advance and the gradient field characteristics of image constructed in advance
When choosing equation progress combined optimization solution, it is preferable that using Bregman alternative manner to by image reconstruction equation and gradient field
The Lagrange's equation that characteristics of image chooses equations simultaneousness is iterated solution, to improve the reconstruction speed of PET image.
It is further preferred that using Bregman alternative manner by Lagrange's equation be decomposed into gradient image renewal function,
Iteration error correction function, PET image reconstruction function and feature extraction function, to gradient image renewal function, iteration mistake
Poor correction function, PET image reconstruction function and feature extraction function are iterated solution, obtain original reconstruction PET image pair
The Object reconstruction PET image answered to further improve the reconstruction speed of PET image, and improves the target weight rebuild and obtained
Build the picture quality of PET image.
Preferably, it is to the feature extraction function that Lagrange's equation decomposesIt obtains
Eigenmatrix D and feature vector αlIt is used as the initial value being updated in next round iteration to gradient image ω, thus will
Original reconstruction PET image is transformed into gradient field from image area, carries out feature learning to gradient area image, to pass through the spy learnt
Matrix and feature vector rarefaction representation original reconstruction PET image are levied, to reduce the noise in original reconstruction PET image, in turn
Improve the reconstruction effect of subsequent PET image.Wherein, k is current iteration number.
Preferably, it is to the gradient image renewal function that Lagrange's equation decomposesThe gradient image ω of acquisition is used as next
The initial value that Object reconstruction PET image m is rebuild in wheel iteration, to reduce the Object reconstruction PET image of subsequent reconstruction
Artifact degree.Wherein, v2For the preset weight for being used to control iteration error in Bregman iteration, b changes for Bregman
The error correction value in generation.
In embodiments of the present invention, it is further preferred that by v2It is set as 1, to further decrease the mesh of subsequent reconstruction
Indicated weight builds the artifact degree of PET image.
Preferably, iteration error correction function Lagrange's equation decomposed
The error correction value b of acquisition is used as the initial value rebuild in next round iteration to Object reconstruction PET image m, thus
Improve the picture quality for rebuilding obtained PET image.
Preferably, PET image reconstruction function Lagrange's equation decomposedωkFor the gradient image of kth time iteration, bkFor kth time
The error correction value of iteration, to improve the picture quality for rebuilding obtained PET image.
As shown in Figure 2, it is preferable that by following step realize using Bregman alternative manner to Lagrange's equation into
Row iteration solves:
In step s 201, according to preset initial characteristics matrix and preset initial characteristics vector, gradient image is used
Renewal function is updated the corresponding gradient image of original reconstruction PET image.
In step S202, according to the error correction value that updated gradient image and iteration error correction function obtain,
Using PET image reconstruction function gradient image is restored to image area from gradient field, obtains Object reconstruction PET image.
In step S203, judge whether current iteration number reaches preset iteration threshold.
In embodiments of the present invention, when current iteration number reaches preset iteration threshold (for example, 50 times), step is executed
Rapid S204, otherwise, go to step S205.
In step S204, Object reconstruction PET image is exported.
In step S205, original reconstruction PET image is set by Object reconstruction PET image, and according to preset image
Block extracts matrix, and the image block of corresponding number is extracted from the corresponding gradient image of original reconstruction PET image, and gradient image includes
Horizontal gradient image and vertical gradient image.
In embodiments of the present invention, original reconstruction PET image is transformed into gradient field from image area first, obtains horizontal ladder
Image and vertical gradient image are spent, matrix is extracted according to preset image block, respectively from horizontal gradient image and vertical gradient map
The horizontal image block and vertical image block of corresponding number are extracted as in.
In step S206, feature learning is carried out to image block, until the corresponding feature square of gradient image that study obtains
Battle array feature vector corresponding with image block meets feature extraction function.
In embodiments of the present invention, feature learning is carried out to extracting horizontal image block and vertical image block respectively, until
Learn the obtained corresponding horizontal/vertical eigenmatrix of horizontal/vertical gradient image and the corresponding water of horizontal/vertical image block
Flat/vertical features vector meets feature extraction function, wherein each column of eigenmatrix feature corresponding with each image block to
Amount corresponds.
In step S207, by eigenmatrix and feature vector be respectively set to initial characteristics matrix and initial characteristics to
Amount increases current iteration number 1 time, and the S201 that gos to step, and continues next round iteration, to rebuild PET image.
S201- step S207 realization through the above steps is iterated solution to Lagrange's equation, to obtain just starting weight
It builds the corresponding Object reconstruction PET image of PET image and improves low dosage PET image to reduce the artifact degree of reconstruction image
The picture quality of reconstruction.
In embodiments of the present invention, according to the collected data for projection of PET device and the sytem matrix of the PET device,
Image reconstruction is carried out to the PET image to be reconstructed initialized in advance by PET image reconstruction algorithm, obtains original reconstruction PET figure
Picture, according to original reconstruction PET image, using Lagrange multiplier by the image reconstruction equation constructed in advance and the ladder constructed in advance
It spends area image Feature Selection equation and carries out combined optimization solution, obtain the corresponding Object reconstruction PET figure of original reconstruction PET image
Picture to improve the reconstruction speed of low dosage PET image, and reduces the artifact degree of reconstruction image, and then improve low dose
Measure the picture quality of PET image reconstruction.
Embodiment two:
Fig. 3 shows the structure of the low dosage PET image reconstruction device provided by Embodiment 2 of the present invention based on gradient field,
For ease of description, only parts related to embodiments of the present invention are shown, including:
Parameter acquiring unit 31, for when receiving the reconstruction request of low dosage PET image, acquisition to pass through PET device
Collected data for projection, and obtain the sytem matrix of PET device；
Original reconstruction unit 32, for passing through preset PET image reconstruction algorithm according to data for projection and sytem matrix
Image reconstruction is carried out to the PET image to be reconstructed initialized in advance, obtains original reconstruction PET image；And
Reconstruction image obtaining unit 33, for that will be constructed in advance using Lagrange multiplier according to original reconstruction PET image
Image reconstruction equation and the gradient field characteristics of image that constructs in advance choose equation and carry out combined optimization solution, obtain original reconstruction
The corresponding Object reconstruction PET image of PET image.
As shown in Figure 4, it is preferable that reconstruction image obtaining unit 33 includes:
Unit 331 is iteratively solved, for using Bregman alternative manner to special by image reconstruction equation and gradient area image
The Lagrange's equation that sign chooses equations simultaneousness is iterated solution.
It is further preferred that iterative solution unit 331 includes:
Equation decomposition unit 3311, for Lagrange's equation to be decomposed into gradient image using Bregman alternative manner
Renewal function, iteration error correction function, PET image reconstruction function and feature extraction function, to update letter to gradient image
Number, iteration error correction function, PET image reconstruction function and feature extraction function are iterated solution, obtain original reconstruction
The corresponding Object reconstruction PET image of PET image.
It is further preferred that as shown in figure 5, equation decomposition unit 3311 includes:
Gradient image updating unit 51, for making according to preset initial characteristics matrix and preset initial characteristics vector
The corresponding gradient image of original reconstruction PET image is updated with gradient image renewal function；
PET image reconstruction unit 52, the mistake for being obtained according to updated gradient image and iteration error correction function
Gradient image is restored to image area using PET image reconstruction function from gradient field, obtains Object reconstruction PET by poor corrected value
Image；
The number of iterations judging unit 53, for judging whether current iteration number reaches preset iteration threshold；
PET image output unit 54 exports Object reconstruction PET image with then then；
Image block extraction unit 55, for otherwise, setting original reconstruction PET image, and root for Object reconstruction PET image
Matrix is extracted according to preset image block, the image block of corresponding number is extracted from the corresponding gradient image of original reconstruction PET image,
Gradient image includes horizontal gradient image and vertical gradient image；
Feature learning unit 56, for carrying out feature learning to image block, until the gradient image that study obtains is corresponding
Eigenmatrix and the corresponding feature vector of image block meet feature extraction function；And
Parameter set unit 57, for eigenmatrix and feature vector to be respectively set to initial characteristics matrix and initial spy
Vector is levied, and triggers gradient image updating unit 51, continues next round iteration, to rebuild PET image.
In embodiments of the present invention, each unit of the low dosage PET image reconstruction device based on gradient field can be by corresponding
Hardware or software unit realize that each unit can be independent soft and hardware unit, also can integrate as a soft and hardware unit,
Herein not to limit the present invention.Specifically, the embodiment of each unit can refer to the description of previous embodiment one, herein no longer
It repeats.
Embodiment three:
Fig. 6 shows the structure of the calculating equipment of the offer of the embodiment of the present invention three, for ease of description, illustrates only and this
The relevant part of inventive embodiments.
The calculating equipment 6 of the embodiment of the present invention includes processor 60, memory 61 and is stored in memory 61 and can
The computer program 62 run on processor 60.The processor 60 is realized above-mentioned based on gradient field when executing computer program 62
Low dosage PET image reconstruction method embodiment in step, such as step S101 to S103 shown in FIG. 1.Alternatively, processor
The function of each unit in above-mentioned each Installation practice, such as unit 31 to 33 shown in Fig. 3 are realized when 60 execution computer program 62
Function.
In embodiments of the present invention, according to the collected data for projection of PET device and the sytem matrix of the PET device,
Image reconstruction is carried out to the PET image to be reconstructed initialized in advance by PET image reconstruction algorithm, obtains original reconstruction PET figure
Picture, according to original reconstruction PET image, using Lagrange multiplier by the image reconstruction equation constructed in advance and the ladder constructed in advance
It spends area image Feature Selection equation and carries out combined optimization solution, obtain the corresponding Object reconstruction PET figure of original reconstruction PET image
Picture to improve the reconstruction speed of low dosage PET image, and reduces the artifact degree of reconstruction image, and then improve low dose
Measure the picture quality of PET image reconstruction.
The calculating equipment of the embodiment of the present invention can be personal computer, server.Processor 60 is held in the calculating equipment 6
The step of realizing when realizing the low dosage PET image reconstruction method based on gradient field when row computer program 62 can refer to aforementioned side
The description of method embodiment, details are not described herein.
Example IV:
In embodiments of the present invention, a kind of computer readable storage medium is provided, which deposits
Computer program is contained, which realizes the above-mentioned low dosage PET image weight based on gradient field when being executed by processor
Step in construction method embodiment, for example, step S101 to S103 shown in FIG. 1.Alternatively, the computer program is held by processor
The function of each unit in above-mentioned each Installation practice, such as the function of unit 31 to 33 shown in Fig. 3 are realized when row.
In embodiments of the present invention, according to the collected data for projection of PET device and the sytem matrix of the PET device,
Image reconstruction is carried out to the PET image to be reconstructed initialized in advance by PET image reconstruction algorithm, obtains original reconstruction PET figure
Picture, according to original reconstruction PET image, using Lagrange multiplier by the image reconstruction equation constructed in advance and the ladder constructed in advance
It spends area image Feature Selection equation and carries out combined optimization solution, obtain the corresponding Object reconstruction PET figure of original reconstruction PET image
Picture to improve the reconstruction speed of low dosage PET image, and reduces the artifact degree of reconstruction image, and then improve low dose
Measure the picture quality of PET image reconstruction.
The computer readable storage medium of the embodiment of the present invention may include can carry computer program code any
Entity or device, recording medium, for example, the memories such as ROM/RAM, disk, CD, flash memory.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (14)
1. a kind of low dosage PET image reconstruction method based on gradient field, which is characterized in that the method includes the following steps:
When receiving the reconstruction request of low dosage PET image, obtain through the collected data for projection of PET device, and obtain
The sytem matrix of the PET device；
According to the data for projection and the sytem matrix, by preset PET image reconstruction algorithm to initializing in advance
PET image to be reconstructed carries out image reconstruction, obtains original reconstruction PET image；
According to the original reconstruction PET image, using Lagrange multiplier by the image reconstruction equation constructed in advance and preparatory structure
The gradient field characteristics of image built chooses equation and carries out combined optimization solution, obtains the corresponding target of the original reconstruction PET image
Rebuild PET image.
2. the method as described in claim 1, which is characterized in that the image reconstruction side that will be constructed in advance using Lagrange multiplier
Journey and the gradient field characteristics of image constructed in advance choose the step of equation carries out combined optimization solution, comprising:
Equations simultaneousness is chosen to by described image Reconstructed equation and the gradient field characteristics of image using Bregman alternative manner
Lagrange's equation is iterated solution.
3. method according to claim 2, which is characterized in that using Bregman alternative manner to the Lagrange's equation
The step of being iterated solution, comprising:
The Lagrange's equation is decomposed by gradient image renewal function using Bregman alternative manner, iteration error corrects
Function, PET image reconstruction function and feature extraction function, with to gradient image renewal function, iteration error correction function,
PET image reconstruction function and feature extraction function are iterated solution, obtain the corresponding mesh of the original reconstruction PET image
Indicated weight builds PET image.
4. method as claimed in claim 3, which is characterized in that using Bregman alternative manner to the Lagrange's equation
The step of being iterated solution, comprising:
According to preset initial characteristics matrix and preset initial characteristics vector, using the gradient image renewal function to described
The corresponding gradient image of original reconstruction PET image is updated；
According to the error correction value that the updated gradient image and the iteration error correction function obtain, using described
The gradient image is restored to image area by PET image reconstruction function from gradient field, obtains Object reconstruction PET image；
Judge whether current iteration number reaches preset iteration threshold；
It is then, to export the Object reconstruction PET image；
Otherwise, the original reconstruction PET image is set by the Object reconstruction PET image, and is mentioned according to preset image block
Matrix is taken, the image block of corresponding number, the gradient image are extracted from the corresponding gradient image of the original reconstruction PET image
Including horizontal gradient image and vertical gradient image；
Feature learning is carried out to described image block, until the corresponding eigenmatrix of the gradient image and the figure that study obtains
As the corresponding feature vector of block meets the feature extraction function；
The eigenmatrix and described eigenvector are respectively set to the initial characteristics matrix and the initial characteristics vector,
And jump to the step of being updated using the gradient image renewal function to the gradient image.
5. method as claimed in claim 3, which is characterized in that the feature extraction function isAnd s.t. | | αl||0≤ L, wherein
RlMatrix is extracted for described image block, i.e., according to the RlL image block is extracted from the gradient image ω, D is the feature
Matrix, αlFor the corresponding feature vector of first of image block, ω(i)Indicate the corresponding horizontal/vertical of the original reconstruction PET image
Gradient image, i ∈ { 1,2 } indicate the direction (horizontal/vertical) of the gradient image ω, and L is indicated to the dilute of described eigenvector
The control coefrficient of degree is dredged, k is the current iteration number.
6. method as claimed in claim 3, which is characterized in that the gradient image renewal functionWherein, v2It indicates to control in Bregman iteration
The weight of iteration error processed, b are the error correction value of Bregman iteration.
7. method as claimed in claim 3, which is characterized in that the iteration error correction function
8. method as claimed in claim 3, which is characterized in that the PET image reconstruction functionyuFor the data for projection, GuFor the sytem matrix, m
For the PET image to be reconstructed, v1For preset weight parameter, ωkFor the gradient image of kth time iteration, bkFor kth time iteration
Error correction value.
9. a kind of low dosage PET image reconstruction device based on gradient field, which is characterized in that described device includes:
Parameter acquiring unit, for when receiving the reconstruction request of low dosage PET image, acquisition to be collected by PET device
Data for projection, and obtain the sytem matrix of the PET device；
Original reconstruction unit, for being calculated by preset PET image reconstruction according to the data for projection and the sytem matrix
Method carries out image reconstruction to the PET image to be reconstructed initialized in advance, obtains original reconstruction PET image；And
Reconstruction image obtaining unit will be constructed using Lagrange multiplier in advance for according to the original reconstruction PET image
Image reconstruction equation and the gradient field characteristics of image constructed in advance choose equation and carry out combined optimization solution, obtain the just starting weight
Build the corresponding Object reconstruction PET image of PET image.
10. device as claimed in claim 9, which is characterized in that the reconstruction image obtaining unit includes:
Unit is iteratively solved, for using Bregman alternative manner to by described image Reconstructed equation and the gradient area image
The Lagrange's equation of Feature Selection equations simultaneousness is iterated solution.
11. device as claimed in claim 10, which is characterized in that the iterative solution unit includes:
Equation decomposition unit is updated for the Lagrange's equation to be decomposed into gradient image using Bregman alternative manner
Function, iteration error correction function, PET image reconstruction function and feature extraction function, with to gradient image renewal function,
Iteration error correction function, PET image reconstruction function and feature extraction function are iterated solution, obtain the just starting weight
Build the corresponding Object reconstruction PET image of PET image.
12. device as claimed in claim 11, which is characterized in that the equation decomposition unit includes:
Gradient image updating unit is used for according to preset initial characteristics matrix and preset initial characteristics vector, using described
Gradient image renewal function is updated the corresponding gradient image of the original reconstruction PET image；
PET image reconstruction unit, for what is obtained according to the updated gradient image and the iteration error correction function
The gradient image is restored to image area using the PET image reconstruction function from gradient field, obtains mesh by error correction value
Indicated weight builds PET image；
The number of iterations judging unit, for judging whether current iteration number reaches preset iteration threshold；
PET image output unit, with then then, exporting the Object reconstruction PET image；
Image block extraction unit, for otherwise, setting the original reconstruction PET image for the Object reconstruction PET image, and
Matrix is extracted according to preset image block, corresponding number is extracted from the corresponding gradient image of the original reconstruction PET image
Image block, the gradient image include horizontal gradient image and vertical gradient image；
Feature learning unit, for carrying out feature learning to described image block, until the gradient image that study obtains is corresponding
Eigenmatrix and the corresponding feature vector of described image block meet the feature extraction function；And
Parameter set unit, for by the eigenmatrix and described eigenvector be respectively set to the initial characteristics matrix and
The initial characteristics vector, and trigger the gradient image updating unit and execute using the gradient image renewal function to described
The step of gradient image is updated.
13. a kind of calculating equipment, including memory, processor and storage are in the memory and can be on the processor
The computer program of operation, which is characterized in that the processor realizes such as claim 1 to 8 when executing the computer program
The step of any one the method.
14. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In when the computer program is executed by processor the step of any one of such as claim 1 to 8 of realization the method.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811525022.4A CN109741411B (en) | 2018-12-13 | 2018-12-13 | Low-dose PET image reconstruction method, device, equipment and medium based on gradient domain |
| PCT/CN2019/071098 WO2020118828A1 (en) | 2018-12-13 | 2019-01-10 | Gradient domain-based low-dose pet image reconstruction method and apparatus, device, and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811525022.4A CN109741411B (en) | 2018-12-13 | 2018-12-13 | Low-dose PET image reconstruction method, device, equipment and medium based on gradient domain |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109741411A true CN109741411A (en) | 2019-05-10 |
| CN109741411B CN109741411B (en) | 2023-04-28 |
Family
ID=66359015
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811525022.4A Active CN109741411B (en) | 2018-12-13 | 2018-12-13 | Low-dose PET image reconstruction method, device, equipment and medium based on gradient domain |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109741411B (en) |
| WO (1) | WO2020118828A1 (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116503570A (en) * | 2023-06-29 | 2023-07-28 | 聚时科技(深圳)有限公司 | Three-dimensional reconstruction method and related device for image |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112365593B (en) * | 2020-11-12 | 2024-03-29 | 江苏赛诺格兰医疗科技有限公司 | PET image reconstruction method and system |
| CN112614204A (en) * | 2020-12-29 | 2021-04-06 | 哈尔滨理工大学 | Capacitance tomography image reconstruction method based on improved least square |
| CN112766395B (en) * | 2021-01-27 | 2023-11-28 | 中国地质大学（北京） | Image matching method and device, electronic equipment and readable storage medium |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090161933A1 (en) * | 2007-12-20 | 2009-06-25 | Guang-Hong Chen | Method for dynamic prior image constrained image reconstruction |
| CN103247061A (en) * | 2013-02-05 | 2013-08-14 | 南方医科大学 | Augmented lagrangian iterative reconstruction method of X-ray image and CI image |
| CN103559728A (en) * | 2013-10-31 | 2014-02-05 | 南方医科大学 | PET image maximum posterior reconstruction method based on united prior model with dissection function |
| CN104408758A (en) * | 2014-11-12 | 2015-03-11 | 南方医科大学 | Low-dose processing method of energy spectrum CT image |
| CN104574416A (en) * | 2015-01-27 | 2015-04-29 | 南方医科大学 | Low-dose energy spectrum CT image denoising method |
| CN104657950A (en) * | 2015-02-16 | 2015-05-27 | 浙江大学 | Dynamic PET (positron emission tomography) image reconstruction method based on Poisson TV |
| CN104751429A (en) * | 2015-01-27 | 2015-07-01 | 南方医科大学 | Dictionary learning based low-dosage energy spectrum CT image processing method |
| US20150287223A1 (en) * | 2014-04-04 | 2015-10-08 | The Board Of Trustees Of The University Of Illinois | Highly accelerated imaging and image reconstruction using adaptive sparsifying transforms |
| CN106056644A (en) * | 2016-05-24 | 2016-10-26 | 深圳先进技术研究院 | Data processing method and data processing device for CT scanning |
| CN106056554A (en) * | 2016-06-01 | 2016-10-26 | 南昌大学 | Magnetic-resonance fast imaging method based on gradient-domain convolution sparse coding |
| CN106530258A (en) * | 2016-11-22 | 2017-03-22 | 哈尔滨工业大学 | Fast iterative magnetic resonance image reconstruction method based on high-order total variation regularization |
| CN106683144A (en) * | 2016-12-30 | 2017-05-17 | 上海联影医疗科技有限公司 | Image iteration reconstruction method and device |
| CN107358653A (en) * | 2017-08-15 | 2017-11-17 | 北京数字精准医疗科技有限公司 | Imaging reconstruction method and device |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8879811B2 (en) * | 2012-03-28 | 2014-11-04 | Siemens Aktiengesellschaft | Alternating direction of multipliers method for parallel MRI reconstruction |
| US9600910B2 (en) * | 2014-01-08 | 2017-03-21 | Rensselaer Polytechnic Institute | Attenuation map reconstruction from TOF PET data |
| CN107845120B (en) * | 2017-09-27 | 2020-11-27 | 深圳先进技术研究院 | PET image reconstruction method, system, terminal and readable storage medium |
-
2018
- 2018-12-13 CN CN201811525022.4A patent/CN109741411B/en active Active
-
2019
- 2019-01-10 WO PCT/CN2019/071098 patent/WO2020118828A1/en active Application Filing
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090161933A1 (en) * | 2007-12-20 | 2009-06-25 | Guang-Hong Chen | Method for dynamic prior image constrained image reconstruction |
| CN103247061A (en) * | 2013-02-05 | 2013-08-14 | 南方医科大学 | Augmented lagrangian iterative reconstruction method of X-ray image and CI image |
| CN103559728A (en) * | 2013-10-31 | 2014-02-05 | 南方医科大学 | PET image maximum posterior reconstruction method based on united prior model with dissection function |
| US20150287223A1 (en) * | 2014-04-04 | 2015-10-08 | The Board Of Trustees Of The University Of Illinois | Highly accelerated imaging and image reconstruction using adaptive sparsifying transforms |
| CN104408758A (en) * | 2014-11-12 | 2015-03-11 | 南方医科大学 | Low-dose processing method of energy spectrum CT image |
| CN104751429A (en) * | 2015-01-27 | 2015-07-01 | 南方医科大学 | Dictionary learning based low-dosage energy spectrum CT image processing method |
| CN104574416A (en) * | 2015-01-27 | 2015-04-29 | 南方医科大学 | Low-dose energy spectrum CT image denoising method |
| CN104657950A (en) * | 2015-02-16 | 2015-05-27 | 浙江大学 | Dynamic PET (positron emission tomography) image reconstruction method based on Poisson TV |
| CN106056644A (en) * | 2016-05-24 | 2016-10-26 | 深圳先进技术研究院 | Data processing method and data processing device for CT scanning |
| CN106056554A (en) * | 2016-06-01 | 2016-10-26 | 南昌大学 | Magnetic-resonance fast imaging method based on gradient-domain convolution sparse coding |
| CN106530258A (en) * | 2016-11-22 | 2017-03-22 | 哈尔滨工业大学 | Fast iterative magnetic resonance image reconstruction method based on high-order total variation regularization |
| CN106683144A (en) * | 2016-12-30 | 2017-05-17 | 上海联影医疗科技有限公司 | Image iteration reconstruction method and device |
| CN107358653A (en) * | 2017-08-15 | 2017-11-17 | 北京数字精准医疗科技有限公司 | Imaging reconstruction method and device |
Non-Patent Citations (2)
| Title |
|---|
| 梁栋: "快速磁共振成像", 《2018第二十届全国波谱学学术年会会议论文摘要集》 * |
| 胡战利 等: "低剂量口腔CT成像***关键技术与成像方法研究", 《集成技术》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116503570A (en) * | 2023-06-29 | 2023-07-28 | 聚时科技(深圳)有限公司 | Three-dimensional reconstruction method and related device for image |
| CN116503570B (en) * | 2023-06-29 | 2023-11-24 | 聚时科技(深圳)有限公司 | Three-dimensional reconstruction method and related device for image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109741411B (en) | 2023-04-28 |
| WO2020118828A1 (en) | 2020-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Gong et al. | PET image reconstruction using deep image prior | |
| US11164345B2 (en) | System and method for generating attenuation map | |
| Gong et al. | Iterative PET image reconstruction using convolutional neural network representation | |
| CN107137102B (en) | PET imaging system and multi-modal medical image processing system | |
| US11576628B2 (en) | Full dose PET image estimation from low-dose PET imaging using deep learning | |
| CN109741411A (en) | Low dosage PET image reconstruction method, device, equipment and medium based on gradient field | |
| Tong et al. | Image reconstruction for PET/CT scanners: past achievements and future challenges | |
| EP2340524B1 (en) | Method for generation of attenuation map in pet-mr | |
| CN109215093A (en) | Low dosage PET image reconstruction method, device, equipment and storage medium | |
| CN106846430B (en) | Image reconstruction method | |
| US20110103669A1 (en) | Completion of Truncated Attenuation Maps Using MLAA | |
| CN107638188A (en) | Image attenuation bearing calibration and device | |
| Gong et al. | Direct reconstruction of linear parametric images from dynamic PET using nonlocal deep image prior | |
| Cheng et al. | Applications of artificial intelligence in nuclear medicine image generation | |
| CN108932741A (en) | Dynamic PET parameter imaging method, device, system and computer readable storage medium | |
| US20210304457A1 (en) | Using neural networks to estimate motion vectors for motion corrected pet image reconstruction | |
| CN107610198A (en) | PET image attenuation correction method and device | |
| US10217250B2 (en) | Multi-view tomographic reconstruction | |
| Karakatsanis et al. | Generalized 3D and 4D motion compensated whole-body PET image reconstruction employing nested EM deconvolution | |
| CN102949201A (en) | Method for phase correlated tomographic image reconstruction from data measured by medical imaging, involves performing phase-correlated iterative image reconstruction which is initialized with non-phase related image volume | |
| Zhong et al. | Quantitative 4D-PET reconstruction for small animal using SMEIR-reconstructed 4D-CBCT | |
| CN110189387A (en) | PET image bearing calibration method, apparatus, computer equipment and storage medium | |
| CN112634147B (en) | PET image noise reduction method, system, device and medium for self-supervision learning | |
| US20230069017A1 (en) | Systems and methods for image reconstruction | |
| US11663758B2 (en) | Systems and methods for motion estimation in PET imaging using AI image reconstructions |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |