CN110197486A - X-ray grating integrated imaging information extracting method, system and storage medium - Google Patents
X-ray grating integrated imaging information extracting method, system and storage medium Download PDFInfo
- Publication number
- CN110197486A CN110197486A CN201910593224.0A CN201910593224A CN110197486A CN 110197486 A CN110197486 A CN 110197486A CN 201910593224 A CN201910593224 A CN 201910593224A CN 110197486 A CN110197486 A CN 110197486A
- Authority
- CN
- China
- Prior art keywords
- displacement curve
- kernel function
- background
- ohject
- order
- 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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus for radiation diagnosis, e.g. combined with radiation therapy equipment
- A61B6/50—Clinical applications
- A61B6/502—Clinical applications involving diagnosis of breast, i.e. mammography
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N23/00—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00
- G01N23/02—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material
- G01N23/04—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material and forming images of the material
- G01N23/041—Phase-contrast imaging, e.g. using grating interferometers
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N23/00—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00
- G01N23/02—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material
- G01N23/06—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material and measuring the absorption
- G01N23/083—Investigating or analysing materials by the use of wave or particle radiation, e.g. X-rays or neutrons, not covered by groups G01N3/00 – G01N17/00, G01N21/00 or G01N22/00 by transmitting the radiation through the material and measuring the absorption the radiation being X-rays
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2223/00—Investigating materials by wave or particle radiation
- G01N2223/03—Investigating materials by wave or particle radiation by transmission
- G01N2223/04—Investigating materials by wave or particle radiation by transmission and measuring absorption
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2223/00—Investigating materials by wave or particle radiation
- G01N2223/10—Different kinds of radiation or particles
- G01N2223/101—Different kinds of radiation or particles electromagnetic radiation
- G01N2223/1016—X-ray
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2223/00—Investigating materials by wave or particle radiation
- G01N2223/40—Imaging
- G01N2223/401—Imaging image processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20048—Transform domain processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20048—Transform domain processing
- G06T2207/20056—Discrete and fast Fourier transform, [DFT, FFT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30068—Mammography; Breast
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Pathology (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Medical Informatics (AREA)
- Radiology & Medical Imaging (AREA)
- General Physics & Mathematics (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biomedical Technology (AREA)
- Toxicology (AREA)
- Theoretical Computer Science (AREA)
- Dentistry (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Biophysics (AREA)
- High Energy & Nuclear Physics (AREA)
- Optics & Photonics (AREA)
- Quality & Reliability (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Analysing Materials By The Use Of Radiation (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
A kind of X-ray grating integrated imaging information extracting method, system and storage medium, method includes: to carry out translation transformation to background displacement curve maximum value is made to be located at center, while to ohject displacement curvilinear translation same distance；Background displacement curve after translation transformation is expressed as cosine function model；Convolution relation based on background displacement curve and ohject displacement curve, obtain the zeroth order square of small angle scatter distributions in Fourier space, normalization first moment and normalization second-order moment around mean, respectively correspond three kinds of absorption, phase contrast and dark field contrast informations, wherein, the kernel function for normalizing first moment and normalizing in second-order moment around mean is broad sense kernel function, is obtained based on the integral characteristic that displacement curve has.This method has greatly widened the selection range and freedom degree of kernel function, additionally provides bigger freedom degree under the premise of having cracking calculating speed as guaranteeing directly to calculate multistage Moment Methods with original kernel function to pick out for the most suitable kernel function of practical application.
Description
Technical field
The disclosure belongs to x-ray imaging technology field, is related to a kind of X-ray grating integrated imaging information extracting method, is
System and storage medium.
Background technique
X-ray grating integrated imaging technology, the partial structurtes that can be realized on micron even submicron order are differentiated, and are to biography
The fine supplement of system X-ray attenuation imaging technique.The technology can extract three kinds of absorption, phase contrast and dark field contrast letters simultaneously
Breath is suitable for low atomic number, materials of low density, especially for the biologic soft tissue structure including mammary gland.In order to obtain
It obtains since the information that small refraction and scattering etc. include occurs in substance for ray, grating integrated imaging technology usually passes through phase
The mode of stepping acquires data.Phase stepping mode is referred to one block of grating in grating integrated imaging along perpendicular to light
The direction of grizzly bar line is equidistantly displaced within one or more periods, is not being placed object respectively and is being placed two kinds of situations of object
Lower progress, to obtain background displacement curve and ohject displacement curve.
After obtaining background displacement curve and ohject displacement curve, grating integrated imaging technology mainly passes through two songs of analysis
The difference of line (can also extract more contrast informations, only current physical message is more to extract three kinds of contrast informations
It is clear that three kinds).
Summary of the invention
(1) technical problems to be solved
Present disclose provides a kind of X-ray grating integrated imaging information extracting method, system and storage mediums, at least portion
Decompose technical problem certainly set forth above.
(2) technical solution
According to one aspect of the disclosure, a kind of X-ray grating integrated imaging information extracting method is provided, comprising: obtain
Take background displacement curve and corresponding ohject displacement curve；Translation transformation is carried out to background displacement curve to make during maximum value is located at
Heart position, while to ohject displacement curvilinear translation same distance；Background displacement curve after translation transformation is expressed as cosine letter
Exponential model；Background displacement curve and small-angle scattering distribution after ohject displacement curve after translation transformation to be expressed as to translation transformation
Convolution；Based on above-mentioned convolution relation, small angle scatter distributions in Fourier space are obtained according to the Fourier transform property of convolution
Zeroth order square, normalization first moment and normalization second-order moment around mean, respectively correspond absorption, phase contrast and dark field three kinds of contrasts letter
Breath, wherein the kernel function in normalization first moment and normalization second-order moment around mean is broad sense kernel function, which is base
What the integral characteristic having in ohject displacement curve and background displacement curve obtained, the integral characteristic are as follows: ohject displacement curve or
The inner product of background displacement curve and any one function only includes zeroth order item and single order item in its Fourier space.
In some embodiments of the present disclosure, the expression-form of the broad sense kernel function are as follows: original kernel function and include second order
And the sum of the arbitrary function of the above Fourier space ingredient of second order.
In some embodiments of the present disclosure, X-ray grating integrated imaging information extracting method, further includes: optimum option
The step of kernel function.
In some embodiments of the present disclosure, the step of kernel function described in the optimum option in, include: pressing down for noise
System carrys out kernel function described in optimum option, wherein for noise suppressed come the method for kernel function described in optimum option are as follows: by parent
Function carries out Fourier expansion, and the kernel function ingredient for choosing zero and first order is optimal kernel function.
In some embodiments of the present disclosure, the method for obtaining background displacement curve and corresponding ohject displacement curve
One or more comprising following method: background displacement curve is obtained based on phase stepping mode and corresponding ohject displacement is bent
Line, the mode successively exposed based on the incoherent x-ray source of distribution perpendicular to grating fringe direction obtain background displacement curve
With corresponding ohject displacement curve, and based on alternating expression or tilting grating and be parallel to grating fringe direction distribution it is non-
The mode that coherent X-ray source successively exposes obtains background displacement curve and corresponding ohject displacement curve.
A kind of X-ray grating integrated imaging system, the system include: number another aspect of the present disclosure provides
According to module is obtained, for obtaining background displacement curve and corresponding ohject displacement curve；Translation transformation operation module, for back
Scape displacement curve carries out translation transformation and maximum value is made to be located at center, while to ohject displacement curvilinear translation same distance；
Function representation module, for the background displacement curve after translation transformation to be expressed as cosine function model；And for that will translate
Transformed ohject displacement curve is expressed as the convolution of the background displacement curve after translation transformation and small-angle scattering distribution；Parse mould
Block obtains small angle scatter distributions in Fourier space according to the Fourier transform property of convolution for being based on above-mentioned convolution relation
Zeroth order square, normalization first moment and normalization second-order moment around mean, respectively correspond absorption, phase contrast and dark field three kinds of contrasts letter
Breath, wherein the kernel function in normalization first moment and normalization second-order moment around mean is broad sense kernel function, which is base
What the integral characteristic having in ohject displacement curve and background displacement curve obtained, the integral characteristic are as follows: ohject displacement curve or
The inner product of background displacement curve and any one function only includes zeroth order item and single order item in its Fourier space.
In some embodiments of the present disclosure, the expression-form of the broad sense kernel function are as follows: original kernel function and include two
The sum of the arbitrary function of rank and the above Fourier space ingredient of second order.
In some embodiments of the present disclosure, X-ray grating integrated imaging system, further includes: kernel optimization module is used
The kernel function described in optimum option.
In some embodiments of the present disclosure, the kernel optimization module is for noise suppressed come core described in optimum option
Function chooses the kernel function ingredient of zero and first order as optimal core by the way that original kernel function is carried out Fourier expansion
Function.
According to the another aspect of the disclosure, a kind of storage medium is provided, can be read, be stored thereon with by computer
Executable instruction, the instruction are subsequently can by computer device and execute any method that the disclosure refers to.
(3) beneficial effect
It can be seen from the above technical proposal that the X-ray grating integrated imaging information extracting method of disclosure offer, system
And storage medium, it has the advantages that
Maximum value is made to be located at center by carrying out translation transformation to background displacement curve, while to corresponding object
Displacement curve translate same distance, this method can be suitable for X-ray grating integrated imaging system, and the application propose
Information extracting method is GD-MMA method, which is expanded on the basis of D-MMA, (includes based on displacement curve
Background displacement curve and ohject displacement curve) integral characteristic that has to calculate in D-MMA method f (φ) or s (φ) with
Another kernel function h (φ) (for example be h for phase contrastP(φ)=φ is h for dark fieldD(φ)=φ2) inner product expansion be
The inner product form for calculating f (φ) or s (φ) and broad sense kernel function, has greatly widened the selection range and freedom of kernel function
Degree additionally provides bigger freedom degree under the premise of guaranteeing has cracking calculating speed as D-MMA to pick out needle
To kernel function most suitable in practical application.
Detailed description of the invention
Fig. 1 is the X-ray grating integrated imaging information extracting method flow chart according to shown in one embodiment of the disclosure.
Fig. 2 is tri- kinds of method contrast schematic diagrams of DB-MMA, D-MMA and GD-MMA according to shown in one embodiment of the disclosure.
Fig. 3 is the X-ray grating integrated imaging system block diagram according to shown in one embodiment of the disclosure.
The background of Fig. 4 output for (a) X-ray grating integrated imaging system according to shown in one embodiment of the disclosure and (b)
Displacement curve and ohject displacement curve synoptic diagram.
Fig. 5 is the experimental data acquired on Talbot type system according to shown in one embodiment of the disclosure, is led to respectively
Cross absorption, phase contrast and dark field hum pattern that tetra- kinds of methods of GD-MMA of FCA, DB-MMA, D-MMA and noise optimization extract.
[symbol description]
3-X radial grating integrated imaging system；
31- data acquisition module；32- translation transformation operation module；
33- function representation module；34- parsing module.
Specific embodiment
In existing X-ray grating integrated imaging information extracting method, main analysis method can be divided into two major classes:
First is that Fourier's constituent analysis (Fourier Component Analysis, FCA) based on cosine model；Second is that being based on small angle
The multistage square for scattering (Small Angle X-ray Scattering, SAXS) analyzes (Multi-order Moments
Analysis, MMA).FCA method is the analysis method that the field uses earliest, it is by by background displacement curve and ohject displacement
Curve is expressed as single order cosine function, is then derived according to the physical meaning of three kinds of absorption, phase contrast and dark field contrast informations
The expression formula of three kinds of contrasts parameter of two kinds of displacement curve Fourier ingredients.This method is it can be readily appreciated that physical meaning is clear
The problem of clear, and arithmetic speed is quickly, but phase contrast information is easy to produce phase-wrapping (Phase Wrapping), cause phase
It serves as a contrast image and serious artifact occurs.MMA method is to be proposed since 2012 by researcher, and this method has bypassed three kinds of contrasts
It directly expresses, but it is considered to certain function of the distribution of small-angle scattering caused by object, and have found through FCA method
The three kinds of contrast informations extracted are just exactly in the zeroth order square, normalization first moment and normalization second order of small-angle scattering distribution
Heart square, small-angle scattering distribution therein is the convolution relation extracted based on researcher before: ohject displacement curve can consider
It is the convolution of background displacement curve and object small-angle scattering distribution.Therefore pass through Lucy-Richardson iteration on the basis of MMA
The method of deconvolution can analyze corresponding small-angle scattering distribution from two kinds of displacement curves, and then can obtain corresponding
Multistage square information, that is, wish the three kinds of contrast informations obtained, at the same this method can also calculate higher order time such as three
The information such as rank square, Fourth-order moment can also be extracted although its clear physical meaning not yet and be different from three kinds of information of tradition
Information image has the possibility of further digging utilization.This MMA method (Deconvolution-Based based on deconvolution
MMA, DB-MMA) it is not in phase-wrapping information in FCA, while more possible information can be obtained, therefore in recent years
Also continuous development has been obtained.Since the deconvolution process carried out in DB-MMA needs to expend the longer calculating time, and need
Wanting manual setting the number of iterations, (different the number of iterations have certain influence for the result of deconvolution, without the optimal of same sample
The number of iterations is also not necessarily identical), therefore followed by some improved methods, such as the MMA method based on principal component analysis
MMA method (the Direct of (Principal Component Analysis MMA, PCA-MMA) and a kind of direct Analytical Expression
MMA, D-MMA).PCA-MMA is the method that the one kind proposed according to the principal component feature of two kinds of displacement curves quickly calculates, still
It also is a kind of approximation method.And D-MMA is a kind of multistage square for the Analytical Expression based on the convolution relation in MMA, derived
Calculation formula, and the multistage square for directly calculating two kinds of displacement curves is obtained with the multistage square of small-angle scattering distribution, also
It is three kinds of traditional contrast informations, avoids the deconvolution iterative process of intermediate complexity in this way, arithmetic speed has obtained greatly
Amplitude is promoted.
It has been found that there are still following technical problems up for solving in above-mentioned improved method: D-MMA is main
It is that calculate f (φ) or s (φ) and another kernel function h (φ) (for example be h for phase contrastP(φ)=φ is h for dark fieldD
(φ)=φ2) inner product, corresponding kernel function is in the form of a single, has significant limitation, and freedom degree is limited.
MMA, DB-MMA and D-MMA method are introduced first below.
1, MMA method
The basic assumption of MMA class method is that ohject displacement curve s (φ) can be expressed as background displacement curve f (φ) and small
The convolution of angle scatter distributions g (φ), expression formula are as follows:
2, DB-MMA method
In DB-MMA, pass through the method for Lucy-Richardson iteration first, from collected each detector pixel
On background displacement curve f (φ) and ohject displacement curve s (φ) in deconvolution obtain small-angle scattering distribution, in deconvolution process
Kth time iterative calculation formula are as follows:
Wherein,It indicates mirror symmetry of the f about origin, generally selects g0=s is as initial value.It then can will be traditional
Three kinds of absorption (A) that FCA method extracts, phase contrast (P) and dark field (D) contrast informations are expressed as the zeroth order of small-angle scattering distribution
Square M0(g), first moment is normalizedWith normalization second-order moment around mean
A→M0(g)=∫ g (φ) d φ (3)
Wherein, Mn(g) g (φ) original n rank square is indicated,N rank square after indicating normalization,Expression is returned
N rank central moment after one change.
The advantage of DB-MMA method information extraction is not needing to carry out displacement curve model hypothesis, and the image obtained
Noise level is lower, and the problem of be not in phase-wrapping, can also calculate the richer information of third moment or more,
But limiting factor is that deconvolution iterative process calculates time the number of iterations that is longer, and needing manual setting optimal, with
And iteration may cause the loss etc. of part-structure details.
3, D-MMA method
On the basis of DB-MMA, D-MMA is also from the convolution relation in formula (1), according to the Fourier transformation of convolution
Property is available:
Wherein,WithIt is the Fourier transformation of s (φ), f (φ) and g (φ) respectively, ω is in Fu
Variable in vane space.
Then available according to the Fourier transform property of multistage square:
It is available by formula (6)Zeroth order, single order and second dervative are as follows:
Bringing formula (8)-(10) into formula (7) and carrying out the operations such as abbreviation and normalization can obtain and absorb (A), phase
Serve as a contrast (P) and the corresponding multistage square expression formula of dark field (D) are as follows:
D-MMA method is that avoiding complicated and time-consuming deconvolution changes compared to the maximum advantage of DB-MMA method
For process, and without introducing more model hypothesis, identical multistage square information can be equally obtained.
But the D-MMA method have the drawback that be not particularly suited for when D-MMA method proposes X-ray grating it is comprehensive at
As system, and the application makes maximum value be located at center by carrying out translation transformation to background displacement curve, while to phase
The ohject displacement curvilinear translation same distance answered, this method can be suitable for X-ray grating integrated imaging system, and this Shen
The GD-MMA method that please be proposed is expanded on the basis of D-MMA, (includes background displacement curve and object based on displacement curve
Body displacement curve) integral characteristic that has to calculate f (φ) or s (φ) and another kernel function h (φ) in D-MMA method
(for example be h for phase contrastP(φ)=φ is h for dark fieldD(φ)=φ2) inner product expand to calculate f (φ) or s (φ)
With the inner product form of broad sense kernel function, it is greatly broadened the freedom degree of kernel function selection.
For the purposes, technical schemes and advantages of the disclosure are more clearly understood, below in conjunction with specific embodiment, and reference
The disclosure is further described in attached drawing.
First embodiment
In first exemplary embodiment of the disclosure, a kind of X-ray grating integrated imaging information extraction side is provided
Method.The X-ray grating integrated imaging information extracting method is the D-MMA method of broad sense, referred to as GD-MMA method, the side GD-MMA
Method is expanded on the basis of D-MMA, based on displacement curve (including background displacement curve and ohject displacement curve) tool
Some integral characteristics to calculate f (φ) or s (φ) and another kernel function h (φ) in D-MMA method (such as phase contrast
It is hP(φ)=φ is h for dark fieldD(φ)=φ2) inner product expand to calculate f (φ) or s (φ) and broad sense kernel function
Inner product form, be greatly broadened kernel function selection freedom degree.
Fig. 1 is the X-ray grating integrated imaging information extracting method flow chart according to shown in one embodiment of the disclosure.
Shown in referring to Fig.1, the X-ray grating integrated imaging information extracting method of the disclosure, comprising:
Step S11: background displacement curve and corresponding ohject displacement curve are obtained；
The method of the acquisition background displacement curve and corresponding ohject displacement curve includes one kind or several of following method
Kind: background displacement curve and corresponding ohject displacement curve are obtained based on phase stepping mode, based on perpendicular to grating fringe side
To the mode that successively exposes of the incoherent x-ray source of distribution obtain background displacement curve and corresponding ohject displacement curve, with
And the side successively exposed based on alternating expression or tilting grating with the incoherent x-ray source of distribution for being parallel to grating fringe direction
Formula obtains background displacement curve and corresponding ohject displacement curve.
Step S12: translation transformation is carried out to background displacement curve, maximum value is made to be located at center, while to object position
Move curvilinear translation same distance；
In the present embodiment, maximum value is made to be located at center by carrying out translation transformation to background displacement curve, simultaneously
To ohject displacement curvilinear translation same distance, enable the method to be suitable for X-ray grating integrated imaging system.
Step S13: the background displacement curve after translation transformation is expressed as cosine function model；
According to the fresnel diffraction scheduling theory in wave optics and the Moire deflectometry scheduling theory in geometric optics, only
In the case where considering ± 1 rank diffraction, background displacement curve can regard cosine function model as, and also send out in actual experiment
Now the model coincide very good with real data, therefore in X-ray grating integrated imaging system, by background displacement curve table
It is shown as following form:
Wherein,WithFor real number,Usually positive number, φ are the phase variant of displacement curve,For background displacement
The initial phase of curve.
Step S14: background displacement curve after the ohject displacement curve after translation transformation to be expressed as to translation transformation and small
The convolution of angle scatter distributions；
Based on the fundamental relation of MMA class method, after the ohject displacement curve after translation transformation is expressed as translation transformation
The convolution of background displacement curve and small-angle scattering distribution, concrete form can be with reference formula (1).
Step S15: it is based on above-mentioned convolution relation, obtains small angle in Fourier space according to the Fourier transform property of convolution
Zeroth order square, normalization first moment and the normalization second-order moment around mean of scatter distributions, it is right to respectively correspond absorption, phase contrast and three kinds of dark field
Than degree information, wherein the kernel function in normalization first moment and normalization second-order moment around mean is broad sense kernel function, the broad sense core letter
Number is that the integral characteristic being had based on ohject displacement curve and background displacement curve is obtained, the integral characteristic are as follows: ohject displacement
The inner product of curve or background displacement curve and any one function only includes zeroth order item and single order item in its Fourier space；
In this step S15, the kernel function of use is that had based on displacement curve (ohject displacement curve and background displacement curve)
What some integral characteristics obtained, which is broad sense kernel function, more than as introduced in D-MMA method for phase contrast core letter
Number form formula is hP(φ)=φ is h for dark field kernel function formD(φ)=φ2, the expression-form of the broad sense kernel function are as follows: former
Beginning kernel function and the sum of the arbitrary function comprising second order and the above Fourier space ingredient of second order, have greatly widened kernel function
Selection range and freedom degree additionally provide bigger oneself under the premise of with guaranteeing has cracking calculating speed D-MMA as
It is picked out by spending for kernel function most suitable in practical application.
Introduce the integral characteristic of displacement curve below, which is the characteristic that applicant has found for the first time, right
It is a kind of completely new guidance method for information extraction algorithm, comes especially for the MMA class method for being related to small-angle scattering
Saying has good application value.
Since the trigonometric function of different orders has good orthogonality, accordingly it can be found that background displacement curve f (φ)
Inner product with any one function h (φ) is 0, as long as h (φ) is only made of the trigonometric function of second order or higher order, it may be assumed that
In addition, in conjunction with the convolution relation between two kinds of displacement curves in formula (1), it has also been found that s (φ) and h (φ)
Inner product can also obtain identical result:
A kind of product of displacement curve (comprising ohject displacement curve and background displacement curve) is summed up from analysis above
Intrinsic: the inner product of displacement curve and any one function only includes zeroth order item and single order item in its Fourier space.This
Integral characteristic is a kind of completely new guidance method for information extraction algorithm, especially for being related to small-angle scattering
MMA class method.
Fig. 2 is tri- kinds of method contrast schematic diagrams of DB-MMA, D-MMA and GD-MMA according to shown in one embodiment of the disclosure.
It referring to shown in Fig. 2, and compares in D-MMA method presented hereinbefore, can be found through observation D-MMA method master
If calculating f (φ) or s (φ) and another kernel function h (φ) (for example is h for phase contrastP(φ)=φ is for dark field
hD(φ)=φ2) inner product.Wherein by the integral characteristic application of the displacement curve of above-mentioned discovery, a kind of D- of broad sense is then established
MMA method, that is, GD-MMA method.In GD-MMA method, phase contrast (P) will be corresponded to respectively and dark field (D) information calculates
Kernel function in the process indicates are as follows:
Found out by formula (17) and (18), in the present embodiment, the expression-form of the broad sense kernel function are as follows: original kernel function with
The sum of arbitrary function comprising second order and the above Fourier space ingredient of second order, therefore, GD-MMA method is compared to D-MMA
Bigger freedom degree is provided for method to pick out for kernel function most suitable in practical application, and will not influence
The structural information of effect, while there is cracking calculating speed as D-MMA.
Certainly, in a preferred embodiment, further includes: described in optimum option the step of kernel function.It can from analysis above
To find out, GD-MMA method has many potential optimal methods in practical applications, below for most normal in practical application
The noise suppressed problem seen is illustrated.
In one embodiment, referring for example to shown in Fig. 2, the step of kernel function described in the optimum option in, include: being directed to
Noise suppressed carrys out kernel function described in optimum option, wherein for noise suppressed come the method for kernel function described in optimum option are as follows: will
Original kernel function carries out Fourier expansion, and the kernel function ingredient for choosing zero and first order is optimal kernel function.
Specifically, countless possible kernel function form is contained in formula (15) and (16), but it is bent according to displacement
The integral characteristic of line it is recognised that wherein there was only zero and first order ingredient is effective signal section, and second order or more at
Dividing does not have any contribution for the multistage square information finally calculated, can amplify noise contribution therein in real data instead.
Therefore, in order to reduce the influence of noise as far as possible, by choosing the kernel function ingredient of zero and first order as optimal core letter
Number, to reduce noise contribution.In the step of choosing the kernel function ingredient of zero and first order, it is necessary first to original kernel function:
hP(φ)=φ and hD(φ)=φ2Carry out Fourier expansion:
To obtain the kernel function of optimal GD-MMA are as follows:
In addition, the ingredient of second order and higher order time can also be calculated in parent according to the solution formula of Basel problem
Ratio in function are as follows:
It can be seen that using the GD-MMA algorithm of noise optimization compared to original D-MMA algorithm for signal-to-noise ratio
Promotion is that very significantly, ratio shared by the ingredient of high-order (being greater than 2 ranks) is although smaller in formula (24), but this is
A part in dark field information is not only made an uproar by second moment from formula (13) it can be seen that dark field information is second-order moment around mean
Sound shadow is rung, and also includes the noise contribution of first moment, therefore actual noise optimization ratio is higher.
Second embodiment
In second exemplary embodiment of the disclosure, a kind of X-ray grating integrated imaging system, the system are provided
For executing above- mentioned information extracting method.
Fig. 3 is the X-ray grating integrated imaging system block diagram according to shown in one embodiment of the disclosure.
Referring to shown in Fig. 3, in the present embodiment, which includes:
Data acquisition module 31, for obtaining background displacement curve and corresponding ohject displacement curve；
Translation transformation operation module 32 makes maximum value be located at centre bit for carrying out translation transformation to background displacement curve
It sets, while to ohject displacement curvilinear translation same distance；
Function representation module 33, for the background displacement curve after translation transformation to be expressed as cosine function model；And
For the ohject displacement curve after translation transformation to be expressed as to the background displacement curve after translation transformation and small-angle scattering distribution
Convolution；
Parsing module 34 obtains Fourier's sky according to the Fourier transform property of convolution for being based on above-mentioned convolution relation
Between medium and small angle scatter distributions zeroth order square, normalization first moment and normalization second-order moment around mean, respectively correspond absorption, phase contrast and dark
The three kinds of contrast informations in field, wherein the kernel function in normalization first moment and normalization second-order moment around mean is broad sense kernel function, is somebody's turn to do
Broad sense kernel function is that the integral characteristic being had based on ohject displacement curve and background displacement curve is obtained, the integral characteristic are as follows:
The inner product of ohject displacement curve or background displacement curve and any one function only include zeroth order item in its Fourier space and
Single order item.
Wherein, it includes such as that the data acquisition module, which obtains background displacement curve and the method for corresponding ohject displacement curve,
The one or more of lower method: based on phase stepping mode or can wait be all phase stepping mode obtain background displacement curve and
Corresponding ohject displacement curve, the mode successively exposed based on the incoherent x-ray source of distribution perpendicular to grating fringe direction
Background displacement curve and corresponding ohject displacement curve are obtained, and based on alternating expression or tilting grating and is parallel to gratings strips
The mode that the incoherent x-ray source of distribution in line direction successively exposes obtains background displacement curve and corresponding ohject displacement is bent
Line.
In some embodiments of the present disclosure, the expression-form of the broad sense kernel function are as follows: original kernel function and include two
The sum of the arbitrary function of rank and the above Fourier space ingredient of second order.
In some embodiments of the present disclosure, X-ray grating integrated imaging system, further includes: kernel optimization module is used
The kernel function described in optimum option.
In some embodiments of the present disclosure, the kernel optimization module is for noise suppressed come core described in optimum option
Function chooses the kernel function ingredient of zero and first order as optimal core by the way that original kernel function is carried out Fourier expansion
Function.
In the present embodiment, the system type of X-ray grating integrated imaging system is unlimited, including Talbot type, Talbot-
The various systems such as Lau type, geometric projection type and quarter-phase grating type, the difference of the system of these types of type in composition
It is two pieces of grating type differences therein, Talbot type does not need source grating G0, and first block of grating is phase grating, and second piece
Grating is absorption grating, and Talbot-Lau type is that first block of grating is phase grating, and second block of grating is absorption grating, geometry
Projection type is that two blocks of gratings are absorption grating, and it is phase grating that quarter-phase grating type, which is two pieces,.Furthermore object position
It sets either be placed between light source and first block of grating, is also possible to be placed between two blocks of gratings.Furthermore x-ray source had been both
It can be conventional X-ray source, be also possible to Microfocus X-ray X source or synchrotron radiation X light source.This external detector is either energy product
Parting detector is also possible to photon counting-type detector.
The background of Fig. 4 output for (a) X-ray grating integrated imaging system according to shown in one embodiment of the disclosure and (b)
Displacement curve and ohject displacement curve synoptic diagram.
By taking Talbot-Lau system as an example, in Fig. 4 shown in (a), the X-ray grating integrated imaging system is along optical path side
To mainly including x-ray source, source grating G0, phase grating G1, analysis grating G2 and X-ray detector, corresponding two kinds of positions
It moves in curve synoptic diagram such as Fig. 4 shown in (b).
Use exemplary Talbot-Lau system referring to (a) in Fig. 4, f (φ) and s (φ) respectively indicate background displacement curve with
Ohject displacement curve, wherein in above-mentioned formula (1)-(24), in the Talbot-Lau system of the present embodiment,
Formula (25) indicates the displacement curve lateral shift as caused by the α of refraction angle, and D is the distance between G1 and G2, p2
It is the period for analyzing grating G2.
Fig. 5 is the experimental data acquired on Talbot type system according to shown in one embodiment of the disclosure, is led to respectively
Cross absorption, phase contrast and dark field hum pattern that tetra- kinds of methods of GD-MMA of FCA, DB-MMA, D-MMA and noise optimization extract.By
The structural information that Fig. 5 can be seen that four kinds of methods extract is almost the same, but noise level is different.
Table 1 is the average and standard deviation of part ROI region of three kinds of methods in phase contrast and darkfield image in Fig. 3.This
Outside in order to eliminate background mean value difference, uses Contrast-to-noise ratio (Contrast Noise Ratio, CNR) and make an uproar as evaluation
The index of sound level, CNR is defined as:
Wherein, C1And C2The mean value of respectively two ROI regions, σ0For the standard deviation of background area.
The mean value of two ROI regions and the CNR value of standard deviation and four kinds of methods in phase contrast and dark field plot in 1. Fig. 3 of table
The GD-MMA method of noise optimization is decreased obviously compared to D-MMA method noise level as can be seen from Table 1, in phase
15.1% and 6.86% are improved respectively on lining and darkfield image.
3rd embodiment
In the third exemplary embodiment of the disclosure, a kind of storage medium is provided, can be read by computer,
On be stored with executable instruction, which is subsequently can by computer device and executes any information extracting method for referring to of the disclosure.
In conclusion present disclose provides a kind of X-ray grating integrated imaging information extracting method, system and storages to be situated between
Matter makes maximum value be located at center by carrying out translation transformation to background displacement curve, while to corresponding ohject displacement
Curvilinear translation same distance, this method can be suitable for X-ray grating integrated imaging system, and the GD- that the application proposes
MMA method is expanded on the basis of D-MMA, based on displacement curve (including background displacement curve and ohject displacement curve)
The integral characteristic having to calculate f (φ) or s (φ) and another kernel function h (φ) in D-MMA method (such as phase
Lining is hP(φ)=φ is h for dark fieldD(φ)=φ2) inner product expand to calculate f (φ) or s (φ) and broad sense core letter
Several inner product forms has greatly widened the selection range and freedom degree of kernel function, cracking guaranteeing to have as D-MMA
Bigger freedom degree is additionally provided under the premise of calculating speed to pick out for kernel function most suitable in practical application.
The disclosure can by means of include several different elements hardware and by means of properly programmed computer come
It realizes.The various component embodiments of the disclosure can be implemented in hardware, or to run on one or more processors
Software module is realized, or is implemented in a combination thereof.It will be understood by those of skill in the art that can be used in practice micro-
Processor or digital signal processor (DSP) are some or all in the relevant device according to the embodiment of the present disclosure to realize
The some or all functions of component.The disclosure be also implemented as a part for executing method as described herein or
Whole device or device programs (for example, computer program and computer program product).Such journey for realizing the disclosure
Sequence can store on a computer-readable medium, or may be in the form of one or more signals.Such signal can
To download from internet website, perhaps it is provided on the carrier signal or is provided in any other form.
In addition, unless specifically described or the step of must sequentially occur, there is no restriction in the above institute for the sequence of above-mentioned steps
Column, and can change or rearrange according to required design.And above-described embodiment can be based on the considerations of design and reliability, that
This mix and match is used using or with other embodiments mix and match, i.e., the technical characteristic in different embodiments can be freely combined
Form more embodiments.
Particular embodiments described above has carried out further in detail the purpose of the disclosure, technical scheme and beneficial effects
Describe in detail it is bright, it is all it should be understood that be not limited to the disclosure the foregoing is merely the specific embodiment of the disclosure
Within the spirit and principle of the disclosure, any modification, equivalent substitution, improvement and etc. done should be included in the guarantor of the disclosure
Within the scope of shield.
Claims (10)
1. a kind of X-ray grating integrated imaging information extracting method characterized by comprising
Obtain background displacement curve and corresponding ohject displacement curve；
Carrying out translation transformation to background displacement curve makes maximum value be located at center, while to ohject displacement curvilinear translation phase
Same distance；
Background displacement curve after translation transformation is expressed as cosine function model；
What background displacement curve and small-angle scattering after the ohject displacement curve after translation transformation to be expressed as to translation transformation were distributed
Convolution；
Based on above-mentioned convolution relation, zero of small angle scatter distributions in Fourier space is obtained according to the Fourier transform property of convolution
Rank square, normalization first moment and normalization second-order moment around mean, respectively correspond three kinds of absorption, phase contrast and dark field contrast informations,
In, normalization first moment and the kernel function normalized in second-order moment around mean are broad sense kernel function, which is based on object
What the integral characteristic that body displacement curve and background displacement curve have obtained, the integral characteristic are as follows: ohject displacement curve or background
The inner product of displacement curve and any one function only includes zeroth order item and single order item in its Fourier space.
2. X-ray grating integrated imaging information extracting method according to claim 1, which is characterized in that the broad sense core letter
Several expression-form are as follows: original kernel function and the sum of the arbitrary function comprising second order and the above Fourier space ingredient of second order.
3. X-ray grating integrated imaging information extracting method according to claim 2, which is characterized in that further include: optimization
The step of choosing the kernel function.
4. X-ray grating integrated imaging information extracting method according to claim 3, which is characterized in that the optimization choosing
In the step of taking the kernel function, include: for noise suppressed come kernel function described in optimum option, wherein for noise suppressed come
The method of kernel function described in optimum option are as follows:
Original kernel function is subjected to Fourier expansion, the kernel function ingredient for choosing zero and first order is optimal kernel function.
5. X-ray grating integrated imaging information extracting method according to any one of claim 1 to 4, which is characterized in that
The method of the acquisition background displacement curve and corresponding ohject displacement curve includes the one or more of following method:
Based on phase stepping mode or can to wait be all that the other way that handles of phase stepping mode obtains background displacement curve and phase
The ohject displacement curve answered,
The mode successively exposed based on the incoherent x-ray source of distribution perpendicular to grating fringe direction obtains background displacement curve
With corresponding ohject displacement curve, and
It is successively exposed based on alternating expression or tilting grating with the incoherent x-ray source of distribution for being parallel to grating fringe direction
Mode obtains background displacement curve and corresponding ohject displacement curve.
6. a kind of X-ray grating integrated imaging system, which is characterized in that the system includes:
Data acquisition module, for obtaining background displacement curve and corresponding ohject displacement curve；
Translation transformation operation module makes maximum value be located at center, together for carrying out translation transformation to background displacement curve
When to ohject displacement curvilinear translation same distance；
Function representation module, for the background displacement curve after translation transformation to be expressed as cosine function model；And for inciting somebody to action
Ohject displacement curve after translation transformation is expressed as the convolution of the background displacement curve after translation transformation and small-angle scattering distribution；
Parsing module obtains small in Fourier space for being based on above-mentioned convolution relation according to the Fourier transform property of convolution
The zeroth order squares of angle scatter distributions, normalization first moment and normalization second-order moment around mean respectively correspond absorption, phase contrast and three kinds of dark field
Contrast information, wherein the kernel function in normalization first moment and normalization second-order moment around mean is broad sense kernel function, the broad sense core
Function is that the integral characteristic being had based on ohject displacement curve and background displacement curve is obtained, the integral characteristic are as follows: object position
The inner product of shifting curve or background displacement curve and any one function only includes zeroth order item and single order item in its Fourier space.
7. X-ray grating integrated imaging system according to claim 6, which is characterized in that the table of the broad sense kernel function
Up to form are as follows: original kernel function and the sum of the arbitrary function comprising second order and the above Fourier space ingredient of second order.
8. X-ray grating integrated imaging system according to claim 6, which is characterized in that further include:
Kernel optimization module, for kernel function described in optimum option.
9. X-ray grating integrated imaging system according to claim 8, which is characterized in that the kernel optimization module
Zeroth order is chosen by the way that original kernel function is carried out Fourier expansion for noise suppressed come kernel function described in optimum option
Kernel function ingredient with single order is as optimal kernel function.
10. a kind of storage medium can be read by computer, be stored thereon with executable instruction, which is characterized in that the instruction quilt
The method that computer processor perform claim requires any one of 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910593224.0A CN110197486B (en) | 2019-07-02 | 2019-07-02 | X-ray grating comprehensive imaging information extraction method, system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910593224.0A CN110197486B (en) | 2019-07-02 | 2019-07-02 | X-ray grating comprehensive imaging information extraction method, system and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110197486A true CN110197486A (en) | 2019-09-03 |
| CN110197486B CN110197486B (en) | 2021-04-06 |
Family
ID=67755707
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910593224.0A Active CN110197486B (en) | 2019-07-02 | 2019-07-02 | X-ray grating comprehensive imaging information extraction method, system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110197486B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111643100A (en) * | 2019-11-21 | 2020-09-11 | 清华大学 | Phase contrast imaging system information characterization method and system |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2003040828A2 (en) * | 2001-10-26 | 2003-05-15 | Applied Materials, Inc. | Real-time prediction of and correction of proximity resist heating in raster scan particle beam lithography |
| CN103557808A (en) * | 2013-11-19 | 2014-02-05 | 东南大学 | Defocusing projection grating measurement method based on Sierra Lite dithering algorithm |
| CN104215340A (en) * | 2014-09-02 | 2014-12-17 | 桂林电子科技大学 | Immediate optical interference fringe processing method |
| US20160042533A1 (en) * | 2014-08-06 | 2016-02-11 | Konica Minolta, Inc. | X-ray imaging system and image processing device |
| EP3021110A1 (en) * | 2014-11-11 | 2016-05-18 | Paul Scherrer Institut | System for obtaining quantitative x-ray images using Hilbert transform on imaged fringes |
| CN106153646A (en) * | 2015-04-08 | 2016-11-23 | 清华大学 | X-ray imaging system and method |
| CN107358614A (en) * | 2017-06-14 | 2017-11-17 | 南京邮电大学 | The method of separate picture information from K-edge characteristic X-ray image |
| CN107392786A (en) * | 2017-07-11 | 2017-11-24 | 中国矿业大学 | Mine fiber grating monitoring system missing data compensation method based on SVMs |
| CN109211803A (en) * | 2018-09-17 | 2019-01-15 | 中国科学院生态环境研究中心 | A kind of device that micro- plastics are quickly identified based on micro- multispectral technology |
| CN109785406A (en) * | 2019-03-18 | 2019-05-21 | 清华大学 | The information extracting method and extraction element of grating image system |
-
2019
- 2019-07-02 CN CN201910593224.0A patent/CN110197486B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2003040828A2 (en) * | 2001-10-26 | 2003-05-15 | Applied Materials, Inc. | Real-time prediction of and correction of proximity resist heating in raster scan particle beam lithography |
| CN103557808A (en) * | 2013-11-19 | 2014-02-05 | 东南大学 | Defocusing projection grating measurement method based on Sierra Lite dithering algorithm |
| US20160042533A1 (en) * | 2014-08-06 | 2016-02-11 | Konica Minolta, Inc. | X-ray imaging system and image processing device |
| CN104215340A (en) * | 2014-09-02 | 2014-12-17 | 桂林电子科技大学 | Immediate optical interference fringe processing method |
| EP3021110A1 (en) * | 2014-11-11 | 2016-05-18 | Paul Scherrer Institut | System for obtaining quantitative x-ray images using Hilbert transform on imaged fringes |
| CN106153646A (en) * | 2015-04-08 | 2016-11-23 | 清华大学 | X-ray imaging system and method |
| CN107358614A (en) * | 2017-06-14 | 2017-11-17 | 南京邮电大学 | The method of separate picture information from K-edge characteristic X-ray image |
| CN107392786A (en) * | 2017-07-11 | 2017-11-24 | 中国矿业大学 | Mine fiber grating monitoring system missing data compensation method based on SVMs |
| CN109211803A (en) * | 2018-09-17 | 2019-01-15 | 中国科学院生态环境研究中心 | A kind of device that micro- plastics are quickly identified based on micro- multispectral technology |
| CN109785406A (en) * | 2019-03-18 | 2019-05-21 | 清华大学 | The information extracting method and extraction element of grating image system |
Non-Patent Citations (9)
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111643100A (en) * | 2019-11-21 | 2020-09-11 | 清华大学 | Phase contrast imaging system information characterization method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110197486B (en) | 2021-04-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Alsmirat et al. | Accelerating compute intensive medical imaging segmentation algorithms using hybrid CPU-GPU implementations | |
| Mirone et al. | The PyHST2 hybrid distributed code for high speed tomographic reconstruction with iterative reconstruction and a priori knowledge capabilities | |
| CN103608825A (en) | Method to determine pixon map in iterative image reconstruction | |
| Li et al. | Reconstruction of heterogeneous materials via stochastic optimization of limited-angle X-ray tomographic projections | |
| Zhang et al. | Application of separable parameter space techniques to multi-tracer PET compartment modeling | |
| Yan et al. | A general theory of interference fringes in x‐ray phase grating imaging | |
| CN110197486A (en) | X-ray grating integrated imaging information extracting method, system and storage medium | |
| Clark et al. | MCR toolkit: A GPU‐based toolkit for multi‐channel reconstruction of preclinical and clinical x‐ray CT data | |
| Schlifske et al. | A fast GPU-based approach to branchless distance-driven projection and back-projection in cone beam CT | |
| Petersen et al. | An electron tomography algorithm for reconstructing 3D morphology using surface tangents of projected scattering interfaces | |
| Carmo et al. | Multitasking segmentation of lung and COVID-19 findings in CT scans using modified EfficientDet, UNet and MobileNetV3 models | |
| Koljonen et al. | A mathematical model and iterative inversion for fluorescent optical projection tomography | |
| Corda et al. | Reduced-precision acceleration of radio-astronomical imaging on reconfigurable hardware | |
| Ortega et al. | High performance computing for optical diffraction tomography | |
| van Gogh et al. | INSIDEnet: Interpretable nonexpansive data‐efficient network for denoising in grating interferometry breast CT | |
| Howison et al. | GPU-accelerated denoising of 3D magnetic resonance images | |
| Smith et al. | Local complexity metrics to quantify the effect of anatomical noise on detectability of lung nodules in chest CT imaging | |
| Zhang et al. | FPGA acceleration by asynchronous parallelization for simultaneous image reconstruction and segmentation based on the Mumford-Shah regularization | |
| Vo et al. | Practical implementations of speckle-based phase-retrieval methods in Python and GPU for tomography | |
| Samala et al. | Digital breast tomosynthesis: effects of projection-view distribution on computer-aided detection of microcalcification clusters | |
| Latifis et al. | A MATLAB vectorizing compiler targeting application-specific instruction set processors | |
| Samei et al. | Estimability index for volume quantification of homogeneous spherical lesions in computed tomography | |
| Six et al. | Joint reconstruction of attenuation, refraction and dark field x-ray phase contrasts using split Barzilai-Borwein steps | |
| Zhang et al. | Study of OSEM with different subsets in grating-based X-ray differential phase-contrast imaging | |
| Gifford et al. | An assessment of PET dose reduction with penalized likelihood image reconstruction using a computationally efficient model observer |
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 |