CN105574828A - Image scattering correction method, device and apparatus - Google Patents
Image scattering correction method, device and apparatus Download PDFInfo
- Publication number
- CN105574828A CN105574828A CN201510974005.9A CN201510974005A CN105574828A CN 105574828 A CN105574828 A CN 105574828A CN 201510974005 A CN201510974005 A CN 201510974005A CN 105574828 A CN105574828 A CN 105574828A
- Authority
- CN
- China
- Prior art keywords
- initial projections
- perspective
- view
- target
- mapping relations
- 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
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000012937 correction Methods 0.000 title abstract description 12
- 238000013507 mapping Methods 0.000 claims abstract description 70
- 238000010586 diagram Methods 0.000 claims abstract description 55
- 238000004422 calculation algorithm Methods 0.000 claims description 51
- 238000010801 machine learning Methods 0.000 claims description 49
- 238000007408 cone-beam computed tomography Methods 0.000 claims description 36
- 238000012549 training Methods 0.000 claims description 33
- 239000006185 dispersion Substances 0.000 claims description 21
- 230000008569 process Effects 0.000 claims description 14
- 238000009499 grossing Methods 0.000 claims description 5
- 230000006978 adaptation Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000001727 in vivo Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008520 organization Effects 0.000 description 2
- 238000007637 random forest analysis Methods 0.000 description 2
- 238000005284 basis set Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000002059 diagnostic imaging Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- 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
- G06T2207/20212—Image combination
- G06T2207/20224—Image subtraction
-
- 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
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Apparatus For Radiation Diagnosis (AREA)
Abstract
The present invention discloses an image scattering correction method, a device and an apparatus. The method comprises the steps of acquiring the initial projection drawing of a to-be-checked person, wherein the initial projection drawing is in the form of a two-dimensional projection drawing obtained through scanning the to-be-checked person by means of a medical facility; based on a mapping relationship model between the pre-established projection drawing and a scattering diagram, estimating the target scattering diagram of the initial projection drawing; and correcting the initial projection drawing through the target scattering diagram to obtain a target projection drawing. According to the embodiments of the invention, after the initial projection drawing of the to-be-checked person is acquired through scanning the to-be-checked person by the medical facility, the accurate target scattering diagram of the initial projection drawing is estimated based on the mapping relationship model capable of accurately representing the relationship between the pre-established projection drawing and the scattering diagram. After that, the initial projection drawing is corrected through the target scattering diagram. In this way, the accurate target projection drawing can be obtained, so that the initial projection drawing is effectively corrected.
Description
Technical field
The present invention relates to medical image processing technology field, particularly relate to image dispersion bearing calibration, device and equipment.
Background technology
CBCT (ConeBeamComputedTomography, cone-beam x-ray computer tomoscan) equipment is a kind of medical imaging devices being widely used in medical field.When utilizing CBCT equipment to scan detected person, X ray is launched by point source, X ray is through after examinee, receive by the sensing point on flat panel detector, according to flat panel detector receive X ray photon number can generate the CBCT perspective view of two dimension, finally by CBCT perspective view being carried out to reconstructions generation three-dimensional reconstruction image.But, because the size of examinee's in-vivo tissue is different with density, X ray is when through examinee, may generating portion scattering, this part X ray accurately cannot be received by the sensing point on flat panel detector, to there is error in the photon number therefore for generating the X ray of CBCT perspective view, cause CBCT perspective view inaccurate.
In the related, can classify to examinee's in-vivo tissue in advance, and set up X ray through scattering model during different tissues, thus after the organization type determining the examinee needing scanning, utilize the scattering model corresponding with this organization type to obtain scattering drawing for estimate, thus utilize the CBCT perspective view that scattering drawing for estimate correct scan obtains.But the usual more complicated of the tissue due to different detected person, the mode setting up scattering model in advance still accurately cannot estimate all scattering situations, thus effectively cannot correct CBCT perspective view.
Summary of the invention
The invention provides image dispersion bearing calibration, device and equipment, to solve the existing problem being difficult to carry out CBCT perspective view effectively correction.
According to the first aspect of the embodiment of the present invention, provide a kind of image dispersion bearing calibration, described method comprises:
Gather the initial projections figure of examinee, described initial projections figure is the two-dimension projection obtained by Medical Devices scanning examinee;
According to the mapping relations model of the perspective view set up in advance and scatter diagram, estimate the target scattering figure of described initial projections figure;
By described target scattering figure, described initial projections figure is corrected, obtain target projection figure.
According to the second aspect of the embodiment of the present invention, provide a kind of image dispersion means for correcting, described device comprises:
Collecting unit, for gathering the initial projections figure of examinee, described initial projections figure is the two-dimension projection obtained by Medical Devices scanning examinee;
Estimation unit, for the mapping relations model according to the perspective view set up in advance and scatter diagram, estimates the target scattering figure of described initial projections figure;
Correcting unit, for being corrected described initial projections figure by described target scattering figure, obtains target projection figure.
According to the third aspect of the embodiment of the present invention, a kind of Medical Devices are provided, comprise: control desk and scanning support; Wherein, described control desk comprises processor, and for storing the storer of described processor executable;
Wherein, described processor is configured to:
Gather the initial projections figure of examinee, described initial projections figure is the two-dimension projection obtained by Medical Devices scanning examinee;
According to the mapping relations model of the perspective view set up in advance and scatter diagram, estimate the target scattering figure of described initial projections figure;
By described target scattering figure, described initial projections figure is corrected, obtain target projection figure.
The application embodiment of the present invention, the mapping relations model of perspective view and scatter diagram can be set up in advance, because this mapping relations model can adopt machine learning algorithm to set up, accurately can indicate the relation between perspective view and scatter diagram, therefore by Medical Devices scanning examinee and after collecting the initial projections figure of examinee, the accurate target scattering figure of this initial projections figure can be estimated according to this mapping relations model, by this target scattering figure, initial projections figure is corrected, target projection figure accurately can be obtained, thus the effective correction achieved initial projections figure.
Should be understood that, it is only exemplary and explanatory that above general description and details hereinafter describe, and can not limit the present invention.
Accompanying drawing explanation
Accompanying drawing to be herein merged in instructions and to form the part of this instructions, shows embodiment according to the invention, and is used from instructions one and explains principle of the present invention.
Fig. 1 is an embodiment process flow diagram of image dispersion bearing calibration of the present invention;
Fig. 2 is another embodiment process flow diagram of image dispersion bearing calibration of the present invention;
Fig. 3 is a kind of structural representation of image dispersion means for correcting place of the present invention Medical Devices;
Fig. 4 is an embodiment block diagram of image dispersion means for correcting of the present invention;
Fig. 5 is another embodiment block diagram of image dispersion means for correcting of the present invention.
Embodiment
The term used in the present invention is only for the object describing specific embodiment, and not intended to be limiting the present invention." one ", " described " and " being somebody's turn to do " of the singulative used in the present invention and appended claims is also intended to comprise most form, unless context clearly represents other implications.It is also understood that term "and/or" used herein refer to and comprise one or more project of listing be associated any or all may combine.
Term first, second, third, etc. may be adopted although should be appreciated that to describe various information in the present invention, these information should not be limited to these terms.These terms are only used for the information of same type to be distinguished from each other out.Such as, without departing from the present invention, the first information also can be called as the second information, and similarly, the second information also can be called as the first information.Depend on linguistic context, word as used in this " if " can be construed as into " ... time " or " when ... time " or " in response to determining ".
CT equipment, as a kind of Medical Devices gathering medical image, generally includes scan table and control desk.Dissimilar CT equipment can adopt different imaging techniques to scan examinee, such as, and common CBCT equipment.CBCT equipment is that pencil-beam is thrown according to computing machine restructuring fault image equipment, it is annular DR (DigitalRadiography with lower quantity of X-rays X (usual tube current is at about 10 milliamperes) around examinee's (also can be described as and throw according to body) by x-ray generator, digital throwing is shone), to obtain two-dimentional CBCT perspective view, the three-dimensional reconstruction image of examinee can be obtained after two-dimentional CBCT perspective view is rebuild.Because the institutional framework of examinee is different, may there is scattering through during examinee in X ray, there is artifact in the two-dimentional CBCT perspective view causing CBCT device scan to obtain.Certainly, aforementionedly only describe the situation that may there is artifact in the two-dimension projection of initial acquisition for CBCT equipment, adopt based on the Medical Devices of other imaging techniques, examinee is scanned time, also may obtain the two-dimension projection comprising artifact.
Therefore in embodiment provided by the invention, the mapping relations model of perspective view and scatter diagram can be set up in advance, because this mapping relations model can adopt machine learning algorithm to set up, accurately can indicate the relation between perspective view and scatter diagram, therefore by Medical Devices scanning examinee and after collecting the initial projections figure of examinee, the accurate target scattering figure of this initial projections figure can be estimated according to this mapping relations model, by this target scattering figure, initial projections figure is corrected, target projection figure accurately can be obtained, thus the effective correction achieved initial projections figure.Describe the present invention below in conjunction with specific embodiment.
See Fig. 1, be an embodiment process flow diagram of image dispersion bearing calibration of the present invention, comprise the following steps:
Step 101: the initial projections figure gathering examinee, this initial projections figure are the two-dimension projection obtained by Medical Devices scanning examinee.
In the present embodiment, the two-dimension projection of the unprocessed mistake obtained by Medical Devices scanning examinee is called initial projections figure.Wherein, Medical Devices can be CBCT equipment, and CBCT equipment adopts pencil-beam x-ray to scan examinee, can be obtained the initial projections figure of examinee's all angles, i.e. two-dimentional CBCT perspective view in scanning process by rotating 360 degrees.
Step 102: according to the mapping relations model of the perspective view set up in advance and scatter diagram, estimates the target scattering figure of initial projections figure.
In the present embodiment, machine learning algorithm can be adopted in advance to set up the mapping relations model of perspective view and scatter diagram.
When setting up mapping relations model, first can obtain training data, training data comprises to be enough to support the initial projections figure of predetermined number required for machine learning and the target scattering figure of correspondence, initial projections figure as training data can be undressed two-dimentional CBCT perspective view, and corresponding target scattering figure is then for carrying out the scatter diagram of accurate correction to these initial projections figure; In one example in which, Monte Carlo (MonteCarlo) emulation tool can be adopted to obtain the initial projections figure of predetermined number and the target scattering figure of correspondence.
After have selected certain target machine learning algorithm of employing, can be trained by the training data of this target machine learning algorithm to aforementioned acquisition, namely can using the input vector of the initial projections figure of predetermined number as target machine learning algorithm, using the desired output vector of the target scattering figure of correspondence as target machine learning algorithm, set up the mapping relations model of perspective view and scatter diagram by operational objective machine learning algorithm; Wherein, target machine learning algorithm can comprise: neural network algorithm, random forests algorithm, degree of depth learning algorithm etc., above-mentioned machine learning algorithm is the machine learning algorithm of widespread use in prior art, and the operational process the present embodiment therefore for arbitrary target machine learning algorithm repeats no more.
Based on the mapping relations model of set up perspective view and scatter diagram, in this step, this mapping relations model can be called, the initial projections figure of the examinee collected is inputted this mapping relations model as input vector, by carrying out computing to this mapping relations model, the target scattering figure of the initial projections figure as output vector namely can be obtained.
Step 103: corrected initial projections figure by target scattering figure, obtains target projection figure.
In this step, initial projections figure can be deducted target scattering figure, the pixel value by pixel each in initial projections figure deducts the pixel value of corresponding pixel points in target scattering figure, thus obtains the target projection figure after correcting.
As seen from the above-described embodiment, being scanned examinee by Medical Devices after collecting the initial projections figure of examinee, can according to set up in advance and the mapping relations model that accurately can indicate the relation between perspective view and scatter diagram estimates the accurate target scattering figure of initial projections figure, by this target scattering figure, initial projections figure is corrected, target projection figure accurately can be obtained, thus achieve the effective correction to initial projections figure.
See Fig. 2, be another embodiment process flow diagram of image dispersion bearing calibration of the present invention, this embodiment describes the process setting up mapping relations model and image dispersion correction in detail, comprises the following steps:
Step 201: obtain training data, training data comprises the initial projections figure of predetermined number and the target scattering figure of correspondence.
In the present embodiment, in order to effectively correct the artifact in the initial projections figure collected by Medical Devices, machine learning algorithm can be adopted in advance to set up the mapping relations model of perspective view and scatter diagram.When setting up mapping relations model, first training data can be obtained, training data comprises to be enough to support the initial projections figure of predetermined number required for machine learning and the target scattering figure of correspondence, initial projections figure as training data can be what gathered by CBCT equipment, and undressed two-dimentional CBCT perspective view, corresponding target scattering figure is then for carrying out the scatter diagram of accurate correction to these initial projections figure.
Optionally, initial projections figure can be the overall perspective view of the examinee collected by CBCT equipment, or also can be the partial view intercepted from this overall perspective view, adopt overall perspective view or partial view can need to determine according to actual medical as training data, this embodiment of the present invention is not limited, to be follow-uply described for overall perspective view.
In this step, following either type can be adopted to obtain training data:
Mode one, obtains training data based on emulation tool.
Such as, Monte Carlo (MonteCarlo) emulation tool can be adopted to obtain the initial projections figure of predetermined number and the target scattering figure of correspondence.
Mode two, obtains training data based on the standard MDCT figure preserved.
To use CBCT equipment, first, the initial projections figure of the predetermined number of different examinee can be gathered by CBCT equipment, and reconstruction acquisition three-dimensional reconstruction figure is carried out to initial projections figure, for each examinee, can by the three-dimensional reconstruction figure at its same position and the standard MDCT (MultiDetectorCT preserved, multi-detector x-ray computer tomoscan) figure carries out registration, obtain registration MDCT to scheme, such as, the three-dimensional reconstruction figure of examinee's chest and standard MDCT figure is carried out registration, and the registration MDCT that can obtain chest schemes;
Secondly, orthogonal projection algorithm can be called, scheme according to registration MDCT the middle perspective view calculating examinee's initial projections figure.For each examinee, the initial projections figure of CBCT equipment collection includes the two-dimentional CBCT perspective view under different angles, such as, within the scope of 360 degree, once two-dimentional CBCT perspective view is gathered every 10 degree, therefore registration MDCT image can be calculated the perspective view corresponding with the two-dimentional CBCT perspective view under each angle according to orthogonal projection algorithm, namely can be described as the middle perspective view of this two-dimentional CBCT perspective view.
Finally, calculate the difference of initial projections figure perspective view middle with it, obtain error image, to the smoothing process of error image, obtain scattering drawing for estimate, initial projections figure is deducted scattering drawing for estimate, obtains the target scattering figure corresponding with the initial projections figure of predetermined number.
Step 202: after carrying out machine learning to described training data, sets up the mapping relations model of described perspective view and scatter diagram.
In this step, default machine learning algorithm can be called, using the input vector of the initial projections figure of the predetermined number as training data as machine learning algorithm, using the desired output vector of the target scattering figure of correspondence as machine learning algorithm, by running machine learning algorithm, the mapping relations model of perspective view and scatter diagram namely can be obtained.Wherein, machine learning algorithm can comprise: neural network algorithm, random forests algorithm, degree of depth learning algorithm etc., above-mentioned machine learning algorithm is the machine learning algorithm of widespread use in prior art, and the operational process the present embodiment therefore for arbitrary target machine learning algorithm repeats no more.
In an optional implementation, before machine learning is carried out to training data, first can carry out pre-service respectively to as the initial projections figure of training data and target scattering figure.Pre-service can comprise: adopt wave filter to carry out Denoising disposal to above-mentioned image, such as, frequency domain low-pass filter can be adopted to remove high frequency noise in above-mentioned image; Such as, and/or carrying out down-sampled process to above-mentioned image, is that the above-mentioned image drop sampling of 512*512 is the image of 32*32 by original size.
Step 203: the initial projections figure gathering examinee, this initial projections figure are the two-dimension projection obtained by Medical Devices scanning examinee.
Step 204: according to the mapping relations model of the perspective view set up in advance and scatter diagram, estimate the target scattering figure of this initial projections figure.
In this step, the perspective view of aforementioned foundation and the mapping relations model of mapping graph can be called, the initial projections figure of the examinee collected is inputted this mapping relations model as input vector, by carrying out computing to this mapping relations model, the target scattering figure of the initial projections figure as output vector namely can be obtained.
Step 205: corrected initial projections figure by target scattering figure, obtains target projection figure.
The perspective view of aforementioned foundation and the mapping relations model of mapping graph can be called, the initial projections figure of the examinee collected is inputted this mapping relations model as input vector, by carrying out computing to this mapping relations model, the target scattering figure of the initial projections figure as output vector namely can be obtained.
As seen from the above-described embodiment, this embodiment can set up the mapping relations model of perspective view and scatter diagram in advance, because this mapping relations model can adopt machine learning algorithm to set up, accurately can indicate the relation between perspective view and scatter diagram, therefore by Medical Devices scanning examinee and after collecting the initial projections figure of examinee, the accurate target scattering figure of this initial projections figure can be estimated according to this mapping relations model, by this target scattering figure, initial projections figure is corrected, target projection figure accurately can be obtained, thus the effective correction achieved initial projections figure.
Corresponding with the embodiment of image dispersion bearing calibration of the present invention, present invention also offers the embodiment of image dispersion means for correcting and Medical Devices.
As shown in Figure 3, be the hardware configuration schematic diagram of Medical Devices of the present invention, these Medical Devices carry out example for CBCT equipment.These Medical Devices include:
Control desk 310 and scanning support 320.Control desk comprises processor 311, storer 312, input equipment 313 and display device 314;
Scanning support 320 comprises x-ray generator 321, high pressure bulb 322, detector 323 and collimating apparatus 324.Wherein, image dispersion means for correcting 300 in storer 312 is as the device on a logical meaning, when needs carry out Air correction, from storer 312, the computer program instructions of this device 300 correspondence can be read in internal memory by processor 311 and run.In one example in which, when needs carry out image dispersion timing, processor 311 performs by reading corresponding instruction in storer 312:
Gather the initial projections figure of examinee, described initial projections figure is the two-dimension projection obtained by Medical Devices scanning examinee;
According to the mapping relations model of the perspective view set up in advance and scatter diagram, estimate the target scattering figure of described initial projections figure;
By described target scattering figure, described initial projections figure is corrected, obtain target projection figure.
In another example, processor 311 can also perform by reading corresponding instruction in storer 312:
Obtain training data, described training data comprises the initial projections figure of predetermined number and the target scattering figure of correspondence;
After machine learning is carried out to described training data, set up the mapping relations model of described perspective view and scatter diagram.
In another example, processor 311 performs described acquisition training data by reading corresponding instruction in storer 312, can comprise:
Gather the initial projections figure of predetermined number, and carry out reconstruction acquisition three-dimensional reconstruction figure to described initial projections figure, wherein, described initial projections figure is the two-dimension projection obtained by cone-beam x-ray computer tomoscan CBCT device scan examinee;
Described three-dimensional reconstruction figure and the standard multi-detector x-ray computer tomoscan MDCT figure preserved are carried out registration, obtains registration MDCT and scheme;
Call orthogonal projection algorithm, scheme according to described registration MDCT the middle perspective view calculating described initial projections figure;
By the smoothing process of error image to described initial projections figure and described middle perspective view, obtain scattering drawing for estimate;
Described initial projections figure is deducted described scattering drawing for estimate, obtains the target scattering figure corresponding with the initial projections figure of described predetermined number.
In another example, processor 311 performs described by after carrying out machine learning to described training data by reading corresponding instruction in storer 312, sets up the mapping relations model of described perspective view and scatter diagram, can comprise:
Call default machine learning algorithm, using the input vector of the initial projections figure of described predetermined number as described machine learning algorithm, using the desired output vector of the target scattering figure of described correspondence as described machine learning algorithm;
By running described machine learning algorithm, obtain the mapping relations model of described perspective view and scatter diagram.
In another example, processor 311 performs the mapping relations model of perspective view that described basis sets up in advance and scatter diagram by reading corresponding instruction in storer 312, estimate the target scattering figure of described initial projections figure, can comprise:
Call the mapping relations model of described perspective view and mapping graph;
Described initial projections figure is inputted described mapping relations model as input vector, and obtains the target scattering figure as the described initial projections figure of output vector.
In another example, processor 311 is corrected described initial projections figure by described target scattering figure described in corresponding instruction execution by reading in storer 312, obtains target projection figure, can comprise:
The pixel value of each pixel in described initial projections figure is deducted the pixel value of corresponding pixel points in described target scattering figure, obtains the target projection figure after correcting.
See Fig. 4, an embodiment block diagram for image dispersion means for correcting of the present invention:
This device comprises: collecting unit 410, estimation unit 420 and correcting unit 430.
Wherein, collecting unit 410, for gathering the initial projections figure of examinee, described initial projections figure is the two-dimension projection obtained by Medical Devices scanning examinee;
Estimation unit 420, for the mapping relations model according to the perspective view set up in advance and scatter diagram, estimates the target scattering figure of described initial projections figure;
Correcting unit 430, for being corrected described initial projections figure by described target scattering figure, obtains target projection figure.
See Fig. 5, another embodiment block diagram for image dispersion means for correcting of the present invention:
This device comprises: acquiring unit 510, unit 520, collecting unit 530, estimation unit 540 and correcting unit 550.
Wherein, acquiring unit 510, for obtaining training data, described training data comprises the initial projections figure of predetermined number and the target scattering figure of correspondence;
Unit 520, for by after carrying out machine learning to described training data, sets up the mapping relations model of described perspective view and scatter diagram;
Collecting unit 530, for gathering the initial projections figure of examinee, described initial projections figure is the two-dimension projection obtained by Medical Devices scanning examinee;
Estimation unit 540, for the mapping relations model according to the perspective view set up in advance and scatter diagram, estimates the target scattering figure of described initial projections figure;
Correcting unit 550, for being corrected described initial projections figure by described target scattering figure, obtains target projection figure.
In an optional implementation, described acquiring unit 510 can comprise (not shown in Fig. 5):
Image Acquisition subelement, for gathering the initial projections figure of predetermined number, and carry out reconstruction acquisition three-dimensional reconstruction figure to described initial projections figure, wherein, described initial projections figure is the two-dimension projection obtained by CBCT device scan examinee;
Image registration subelement, for described three-dimensional reconstruction figure and the standard MDCT figure preserved are carried out registration, obtains registration MDCT and schemes;
Image procossing subelement, for calling orthogonal projection algorithm, the middle perspective view calculating described initial projections figure is schemed according to described registration MDCT, by the smoothing process of error image to described initial projections figure and described middle perspective view, obtain scattering drawing for estimate, and described initial projections figure is deducted described scattering drawing for estimate, obtain the target scattering figure corresponding with the initial projections figure of described predetermined number.
In another optional implementation, described unit 520 can comprise (not shown in Fig. 5):
Algorithm calls subelement, for using the input vector of the initial projections figure of described predetermined number as described machine learning algorithm, using the desired output vector of the target scattering figure of described correspondence as described machine learning algorithm;
Machine learning subelement, for by running described machine learning algorithm, obtains the mapping relations model of described perspective view and scatter diagram.
In another optional implementation, described estimation unit 540 can comprise (not shown in Fig. 5):
Model calls subelement, for calling the mapping relations model of described perspective view and mapping graph;
Image estimation subelement, for described initial projections figure is inputted described mapping relations model as input vector, and obtains the target scattering figure as the described initial projections figure of output vector.
In another optional implementation, described correcting unit 550, specifically for the pixel value of each pixel in described initial projections figure is deducted the pixel value of corresponding pixel points in described target scattering figure, can obtain the target projection figure after correcting.
In said apparatus, the implementation procedure of the function and efficacy of unit specifically refers to the implementation procedure of corresponding step in said method, does not repeat them here.
For device embodiment, because it corresponds essentially to embodiment of the method, so relevant part illustrates see the part of embodiment of the method.Device embodiment described above is only schematic, the wherein said unit illustrated as separating component or can may not be and physically separates, parts as unit display can be or may not be physical location, namely can be positioned at a place, or also can be distributed in multiple network element.Some or all of module wherein can be selected according to the actual needs to realize the object of the present invention program.Those of ordinary skill in the art, when not paying creative work, are namely appreciated that and implement.
As seen from the above-described embodiment, this embodiment can set up the mapping relations model of perspective view and scatter diagram in advance, because this mapping relations model can adopt machine learning algorithm to set up, accurately can indicate the relation between perspective view and scatter diagram, therefore by Medical Devices scanning examinee and after collecting the initial projections figure of examinee, the accurate target scattering figure of this initial projections figure can be estimated according to this mapping relations model, by this target scattering figure, initial projections figure is corrected, target projection figure accurately can be obtained, thus the effective correction achieved initial projections figure.
Those skilled in the art, at consideration instructions and after putting into practice invention disclosed herein, will easily expect other embodiment of the present invention.The present invention is intended to contain any modification of the present invention, purposes or adaptations, and these modification, purposes or adaptations are followed general principle of the present invention and comprised the undocumented common practise in the art of the present invention or conventional techniques means.Instructions and embodiment are only regarded as exemplary, and true scope of the present invention and spirit are pointed out by claim below.
Should be understood that, the present invention is not limited to precision architecture described above and illustrated in the accompanying drawings, and can carry out various amendment and change not departing from its scope.Scope of the present invention is only limited by appended claim.
Claims (13)
1. an image dispersion bearing calibration, is characterized in that, described method comprises:
Gather the initial projections figure of examinee, described initial projections figure is the two-dimension projection obtained by Medical Devices scanning examinee;
According to the mapping relations model of the perspective view set up in advance and scatter diagram, estimate the target scattering figure of described initial projections figure;
By described target scattering figure, described initial projections figure is corrected, obtain target projection figure.
2. method according to claim 1, is characterized in that, before the initial projections figure of described collection examinee, also comprises:
Obtain training data, described training data comprises the initial projections figure of predetermined number and the target scattering figure of correspondence;
After machine learning is carried out to described training data, set up the mapping relations model of described perspective view and scatter diagram.
3. method according to claim 2, is characterized in that, described acquisition training data, comprising:
Gather the initial projections figure of predetermined number, and carry out reconstruction acquisition three-dimensional reconstruction figure to described initial projections figure, wherein, described initial projections figure is the two-dimension projection obtained by cone-beam x-ray computer tomoscan CBCT device scan examinee;
Described three-dimensional reconstruction figure and the standard multi-detector x-ray computer tomoscan MDCT figure preserved are carried out registration, obtains registration MDCT and scheme;
Call orthogonal projection algorithm, scheme according to described registration MDCT the middle perspective view calculating described initial projections figure;
By the smoothing process of error image to described initial projections figure and described middle perspective view, obtain scattering drawing for estimate;
Described initial projections figure is deducted described scattering drawing for estimate, obtains the target scattering figure corresponding with the initial projections figure of described predetermined number.
4. method according to claim 2, is characterized in that, described by after carrying out machine learning to described training data, sets up the mapping relations model of described perspective view and scatter diagram, comprising:
Call default machine learning algorithm, using the input vector of the initial projections figure of described predetermined number as described machine learning algorithm, using the desired output vector of the target scattering figure of described correspondence as described machine learning algorithm;
By running described machine learning algorithm, obtain the mapping relations model of described perspective view and scatter diagram.
5. method according to claim 1, is characterized in that, the mapping relations model of the perspective view that described basis is set up in advance and scatter diagram, estimates the target scattering figure of described initial projections figure, comprising:
Call the mapping relations model of described perspective view and mapping graph;
Described initial projections figure is inputted described mapping relations model as input vector, and obtains the target scattering figure as the described initial projections figure of output vector.
6. method according to claim 1, is characterized in that, is describedly corrected described initial projections figure by described target scattering figure, obtains target projection figure, comprising:
The pixel value of each pixel in described initial projections figure is deducted the pixel value of corresponding pixel points in described target scattering figure, obtains the target projection figure after correcting.
7. an image dispersion means for correcting, is characterized in that, described device comprises:
Collecting unit, for gathering the initial projections figure of examinee, described initial projections figure is the two-dimension projection obtained by Medical Devices scanning examinee;
Estimation unit, for the mapping relations model according to the perspective view set up in advance and scatter diagram, estimates the target scattering figure of described initial projections figure;
Correcting unit, for being corrected described initial projections figure by described target scattering figure, obtains target projection figure.
8. device according to claim 7, is characterized in that, described device also comprises:
Acquiring unit, for obtaining training data, described training data comprises the initial projections figure of predetermined number and the target scattering figure of correspondence;
Unit, for by after carrying out machine learning to described training data, sets up the mapping relations model of described perspective view and scatter diagram.
9. device according to claim 8, is characterized in that, described acquiring unit comprises:
Image Acquisition subelement, for gathering the initial projections figure of predetermined number, and carry out reconstruction acquisition three-dimensional reconstruction figure to described initial projections figure, wherein, described initial projections figure is the two-dimension projection obtained by CBCT device scan examinee;
Image registration subelement, for described three-dimensional reconstruction figure and the standard MDCT figure preserved are carried out registration, obtains registration MDCT and schemes;
Image procossing subelement, for calling orthogonal projection algorithm, the middle perspective view calculating described initial projections figure is schemed according to described registration MDCT, by the smoothing process of error image to described initial projections figure and described middle perspective view, obtain scattering drawing for estimate, and described initial projections figure is deducted described scattering drawing for estimate, obtain the target scattering figure corresponding with the initial projections figure of described predetermined number.
10. device according to claim 8, is characterized in that, described unit comprises:
Algorithm calls subelement, for using the input vector of the initial projections figure of described predetermined number as described machine learning algorithm, using the desired output vector of the target scattering figure of described correspondence as described machine learning algorithm;
Machine learning subelement, for by running described machine learning algorithm, obtains the mapping relations model of described perspective view and scatter diagram.
11. devices according to claim 7, is characterized in that, described estimation unit comprises:
Model calls subelement, for calling the mapping relations model of described perspective view and mapping graph;
Image estimation subelement, for described initial projections figure is inputted described mapping relations model as input vector, and obtains the target scattering figure as the described initial projections figure of output vector.
12. devices according to claim 7, is characterized in that,
Described correcting unit, specifically for the pixel value of each pixel in described initial projections figure is deducted the pixel value of corresponding pixel points in described target scattering figure, obtains the target projection figure after correcting.
13. 1 kinds of Medical Devices, is characterized in that, comprising: control desk and scanning support; Wherein, described control desk comprises processor, and for storing the storer of described processor executable;
Wherein, described processor is configured to:
Gather the initial projections figure of examinee, described initial projections figure is the two-dimension projection obtained by Medical Devices scanning examinee;
According to the mapping relations model of the perspective view set up in advance and scatter diagram, estimate the target scattering figure of described initial projections figure;
By described target scattering figure, described initial projections figure is corrected, obtain target projection figure.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510974005.9A CN105574828B (en) | 2015-12-22 | 2015-12-22 | Image dispersion bearing calibration, device and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510974005.9A CN105574828B (en) | 2015-12-22 | 2015-12-22 | Image dispersion bearing calibration, device and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105574828A true CN105574828A (en) | 2016-05-11 |
| CN105574828B CN105574828B (en) | 2019-01-25 |
Family
ID=55884923
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510974005.9A Active CN105574828B (en) | 2015-12-22 | 2015-12-22 | Image dispersion bearing calibration, device and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105574828B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107742536A (en) * | 2017-10-16 | 2018-02-27 | 成都黑杉科技有限公司 | The method and device of information processing |
| EP3326534A1 (en) * | 2016-11-25 | 2018-05-30 | Samsung Electronics Co., Ltd. | X-ray apparatus and method of acquiring medical image thereof |
| CN108389242A (en) * | 2018-04-10 | 2018-08-10 | 上海联影医疗科技有限公司 | The bearing calibration of projected image and device |
| CN110490948A (en) * | 2019-08-12 | 2019-11-22 | 东软医疗系统股份有限公司 | A kind of scatter correction method and device of PET image |
| CN110660111A (en) * | 2019-09-18 | 2020-01-07 | 东软医疗系统股份有限公司 | PET scattering correction and image reconstruction method, device and equipment |
| WO2020082280A1 (en) * | 2018-10-25 | 2020-04-30 | Shanghai United Imaging Healthcare Co., Ltd. | System and method for scatter correction |
| CN111492406A (en) * | 2017-10-11 | 2020-08-04 | 通用电气公司 | Image generation using machine learning |
| JP2020534929A (en) * | 2017-09-28 | 2020-12-03 | コーニンクレッカ フィリップス エヌ ヴェＫｏｎｉｎｋｌｉｊｋｅ Ｐｈｉｌｉｐｓ Ｎ．Ｖ． | Deep learning-based scattering correction |
| CN113643394A (en) * | 2021-07-22 | 2021-11-12 | 上海联影医疗科技股份有限公司 | Scattering correction method, device, computer equipment and storage medium |
| CN113939339A (en) * | 2020-07-27 | 2022-01-14 | 西安大医集团股份有限公司 | Tracking method and device |
| US12008689B2 (en) | 2021-12-03 | 2024-06-11 | Canon Medical Systems Corporation | Devices, systems, and methods for deep-learning kernel-based scatter estimation and correction |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6163589A (en) * | 1998-06-13 | 2000-12-19 | General Electric Company | Monte Carlo scatter correction method for computed tomography of general object geometries |
| CN1635850A (en) * | 2002-02-21 | 2005-07-06 | 罗切斯特大学 | X-ray scatter correction |
| US20080013673A1 (en) * | 2004-06-16 | 2008-01-17 | Ernst-Peter Ruhmschopf | Apparatus and Method for Scatter Correction in Projection Radiography |
| CN101158653A (en) * | 2007-11-16 | 2008-04-09 | 西北工业大学 | Diffuse transmission measuring and correcting method of cone-beam CT system |
| US20090225932A1 (en) * | 2008-02-27 | 2009-09-10 | Lei Zhu | Cone-beam CT imaging scheme |
| WO2010058329A1 (en) * | 2008-11-21 | 2010-05-27 | Philips Intellectual Property & Standards Gmbh | System and method for x-ray scatter correction |
| US20130058450A1 (en) * | 2011-09-01 | 2013-03-07 | Xin Liu | Method and apparatus for adaptive scatter correction |
| CN103987320A (en) * | 2011-12-12 | 2014-08-13 | 株式会社日立医疗器械 | X-ray ct device and method for correcting scattered x-rays |
-
2015
- 2015-12-22 CN CN201510974005.9A patent/CN105574828B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6163589A (en) * | 1998-06-13 | 2000-12-19 | General Electric Company | Monte Carlo scatter correction method for computed tomography of general object geometries |
| CN1635850A (en) * | 2002-02-21 | 2005-07-06 | 罗切斯特大学 | X-ray scatter correction |
| US20080013673A1 (en) * | 2004-06-16 | 2008-01-17 | Ernst-Peter Ruhmschopf | Apparatus and Method for Scatter Correction in Projection Radiography |
| CN101158653A (en) * | 2007-11-16 | 2008-04-09 | 西北工业大学 | Diffuse transmission measuring and correcting method of cone-beam CT system |
| US20090225932A1 (en) * | 2008-02-27 | 2009-09-10 | Lei Zhu | Cone-beam CT imaging scheme |
| WO2010058329A1 (en) * | 2008-11-21 | 2010-05-27 | Philips Intellectual Property & Standards Gmbh | System and method for x-ray scatter correction |
| US20130058450A1 (en) * | 2011-09-01 | 2013-03-07 | Xin Liu | Method and apparatus for adaptive scatter correction |
| CN103987320A (en) * | 2011-12-12 | 2014-08-13 | 株式会社日立医疗器械 | X-ray ct device and method for correcting scattered x-rays |
Non-Patent Citations (3)
| Title |
|---|
| G.EL FAKHRI 等: "A New Scatter Compensation Method for Ga-67 Imaging Using Artificial Neural Networks", 《IEEE TRANSACTIONS ON NUCLEAR SCIENCE》 * |
| TIANYE NIU 等: "Shading correction for on-board cone-beam CT in radiation therapy using planning MDCT images", 《MEDICAL PHYSICS》 * |
| 邵义文 等: "锥形束CT系统的散射校正方法分析", 《中国医学物理学杂志》 * |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11589831B2 (en) | 2016-11-25 | 2023-02-28 | Samsung Electronics Co., Ltd. | X-ray apparatus and method of acquiring medical image thereof |
| KR20180059329A (en) * | 2016-11-25 | 2018-06-04 | 삼성전자주식회사 | X-ray image apparatus and method for obtaining medicalimage thereof |
| US11083421B2 (en) | 2016-11-25 | 2021-08-10 | Samsung Electronics Co., Ltd. | X-ray apparatus and method of acquiring medical image thereof |
| EP3326534A1 (en) * | 2016-11-25 | 2018-05-30 | Samsung Electronics Co., Ltd. | X-ray apparatus and method of acquiring medical image thereof |
| KR102399148B1 (en) * | 2016-11-25 | 2022-05-19 | 삼성전자주식회사 | X-ray image apparatus and method for obtaining medicalimage thereof |
| JP2020534929A (en) * | 2017-09-28 | 2020-12-03 | コーニンクレッカ フィリップス エヌ ヴェＫｏｎｉｎｋｌｉｊｋｅ Ｐｈｉｌｉｐｓ Ｎ．Ｖ． | Deep learning-based scattering correction |
| CN111492406B (en) * | 2017-10-11 | 2023-12-22 | 通用电气公司 | Method for training machine learning algorithm, image processing system and image reconstruction method |
| CN111492406A (en) * | 2017-10-11 | 2020-08-04 | 通用电气公司 | Image generation using machine learning |
| CN107742536A (en) * | 2017-10-16 | 2018-02-27 | 成都黑杉科技有限公司 | The method and device of information processing |
| CN107742536B (en) * | 2017-10-16 | 2021-04-06 | 成都黑杉科技有限公司 | Information processing method and device |
| CN108389242A (en) * | 2018-04-10 | 2018-08-10 | 上海联影医疗科技有限公司 | The bearing calibration of projected image and device |
| WO2020082280A1 (en) * | 2018-10-25 | 2020-04-30 | Shanghai United Imaging Healthcare Co., Ltd. | System and method for scatter correction |
| US11875433B2 (en) | 2018-10-25 | 2024-01-16 | Shanghai United Imaging Healthcare Co., Ltd. | System and method for scatter correction |
| US11087507B2 (en) | 2018-10-25 | 2021-08-10 | Shanghai United Imaging Healthcare. Co., Ltd. | System and method for scatter correction |
| CN110490948B (en) * | 2019-08-12 | 2023-05-16 | 东软医疗系统股份有限公司 | Scattering correction method and device for PET image |
| CN110490948A (en) * | 2019-08-12 | 2019-11-22 | 东软医疗系统股份有限公司 | A kind of scatter correction method and device of PET image |
| CN110660111A (en) * | 2019-09-18 | 2020-01-07 | 东软医疗系统股份有限公司 | PET scattering correction and image reconstruction method, device and equipment |
| CN110660111B (en) * | 2019-09-18 | 2023-05-30 | 东软医疗系统股份有限公司 | PET scattering correction and image reconstruction method, device and equipment |
| CN113939339A (en) * | 2020-07-27 | 2022-01-14 | 西安大医集团股份有限公司 | Tracking method and device |
| CN113939339B (en) * | 2020-07-27 | 2024-03-01 | 西安大医集团股份有限公司 | Tracking method and device |
| CN113643394A (en) * | 2021-07-22 | 2021-11-12 | 上海联影医疗科技股份有限公司 | Scattering correction method, device, computer equipment and storage medium |
| US12008689B2 (en) | 2021-12-03 | 2024-06-11 | Canon Medical Systems Corporation | Devices, systems, and methods for deep-learning kernel-based scatter estimation and correction |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105574828B (en) | 2019-01-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105574828A (en) | Image scattering correction method, device and apparatus | |
| US7912180B2 (en) | Scattered radiation correction method and scattered radiation correction apparatus | |
| RU2529478C2 (en) | Method and apparatus for image formation in wide field of vision and detection and compensation of motion artefacts | |
| Sun et al. | An iterative projection‐based motion estimation and compensation scheme for head x‐ray CT | |
| CN102016923A (en) | Image artifact reduction | |
| US10445904B2 (en) | Method and device for the automatic generation of synthetic projections | |
| CN107085860A (en) | Image rebuilding method and device based on computed tomography imaging system | |
| JP2006025868A (en) | Image processing apparatus, image processing method, and x-ray ct system | |
| KR20170088742A (en) | Workstation, medical imaging apparatus comprising the same and control method for the same | |
| EP3349655B1 (en) | Tomography apparatus and controlling method for the same | |
| US20220319071A1 (en) | Confidence map for neural network based limited angle artifact reduction in cone beam ct | |
| KR20200086919A (en) | Tomographic imaging apparatus and method for tomographic imaging | |
| CN107106106B (en) | Adaptive segmentation for rotational C-arm computed tomography with reduced angular range | |
| US20220160323A1 (en) | Method of calibrating x-ray projection geometry in x-ray cone beam computed tomography | |
| JP2006334319A (en) | X-ray ct apparatus and method of preprocessing thereof, data preparation device and method thereof, and control program | |
| US20220207794A1 (en) | Method of metal artefact reduction in x-ray dental volume tomography | |
| KR20170128975A (en) | Vessel segmentation device and vessel segmentation method thereof | |
| CN101478920A (en) | Local motion compensated reconstruction of stenosis | |
| US20170323461A1 (en) | Method and evaluation device for evaluating projection data of an object being examined | |
| US20210321969A1 (en) | Providing a dynamic mask image | |
| US20220414832A1 (en) | X-ray imaging restoration using deep learning algorithms | |
| JP2024500828A (en) | peripheral perfusion measurement | |
| US11406343B2 (en) | CT scan results | |
| CN102499705B (en) | Method and system for eliminating ring artifact in tomography | |
| Redgate et al. | Using a registration-based motion correction algorithm to correct for respiratory motion during myocardial perfusion imaging |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: 110167 No. 177-1 Innovation Road, Hunnan District, Shenyang City, Liaoning Province Patentee after: DongSoft Medical System Co., Ltd. Address before: Hunnan New Century Road 110179 Shenyang city of Liaoning Province, No. 16 Patentee before: Dongruan Medical Systems Co., Ltd., Shenyang |