CN108898622A - A kind of the representation of athletic method, apparatus and computer readable storage medium of heart - Google Patents

A kind of the representation of athletic method, apparatus and computer readable storage medium of heart Download PDF

Info

Publication number
CN108898622A
CN108898622A CN201810730181.1A CN201810730181A CN108898622A CN 108898622 A CN108898622 A CN 108898622A CN 201810730181 A CN201810730181 A CN 201810730181A CN 108898622 A CN108898622 A CN 108898622A
Authority
CN
China
Prior art keywords
phase
graph structure
heart
vertex
heart graph
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.)
Pending
Application number
CN201810730181.1A
Other languages
Chinese (zh)
Inventor
杨烜
张正瑞
郭伟
裴继红
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen University
Original Assignee
Shenzhen University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenzhen University filed Critical Shenzhen University
Priority to CN201810730181.1A priority Critical patent/CN108898622A/en
Publication of CN108898622A publication Critical patent/CN108898622A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • 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
    • G06T2207/10088—Magnetic resonance imaging [MRI]
    • 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
    • G06T2207/20081—Training; Learning
    • 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
    • G06T2207/20084—Artificial neural networks [ANN]
    • 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
    • G06T2207/30048—Heart; Cardiac

Abstract

The invention discloses the representation of athletic method, apparatus and computer readable storage medium of a kind of heart, and the MR image by extracting first phase, second phase is handled and constructs to obtain first phase heart graph structure A and second phase heart graph structure Bi, convex objective function is matched using two heart graph structure solution figures and obtains the vertex set with corresponding relationship, deformation function is constructed based on the vertex set with corresponding relationship, and using it to BiIt carries out deformation and obtains new second phase heart graph structure Bi+1, calculate BiWith Bi+1Changing value and judge whether changing value is greater than preset threshold, if then enabling i=i+1 and recycling step of the present invention, if otherwise using deformation function characterization heart movement.This method constructs to obtain after deformation function by the way of circulation, deformation also is carried out to original image structure using the deformation function that building obtains, the whether accurate of deformation function is further verified according to deformation changing value, therefore this method can more accurately characterize the movement of heart.

Description

A kind of the representation of athletic method, apparatus and computer readable storage medium of heart
Technical field
The present invention relates to field of computer technology, a kind of representation of athletic method, apparatus more specifically to heart and Computer readable storage medium.
Background technique
The motion change of tracking cardiac anatomy and heart disease tissue is analyzed using cardiac imaging, i.e. analysis of cardiac is transported It is dynamic, it is heart disease diagnosis and the important means for formulating therapeutic scheme, the existing method for registering based on B-spline free deformation, Optical flow method and method based on shape and structure all cannot accurately characterize the movement of heart, therefore how utilize the structure of myocardium The movement that feature more accurately characterizes heart is a urgently to be resolved.
Summary of the invention
The main purpose of the present invention is to provide a kind of representation of athletic methods of heart, it is intended to which solving the prior art can not essence Really the technical issues of characterization heart movement.
To achieve the above object, the present invention provides a kind of representation of athletic method of heart, and this method includes：
Step 1, the first phase magnetic resonance MR image and second phase MR image for extracting heart, using full convolutional Neural net Network divides first phase MR image, second phase MR image respectively, obtains the endocardial contours and epicardial contours of left ventricle, and It is sampled in endocardial contours and epicardial contours, obtains several cardiod diagram vertex；
Step 2, the center for connecting first phase MR picture centre epicardium contours and the cardiod diagram on first phase MR image Vertex obtains first phase heart graph structure A；Connect center and the second phase MR of second phase MR picture centre epicardium contours Cardiod diagram vertex on image obtains second phase heart graph structure Bi；The initial value of i is 1；
Step 3 utilizes first phase heart graph structure A and second phase heart graph structure BiIt solves figure and matches convex target letter Number obtains vertex and the second phase heart graph structure B of first phase heart graph structure AiIn with corresponding relationship vertex set It closes；
Step 4 constructs deformation function based on the vertex set with corresponding relationship, and using deformation function to second Phase heart graph structure BiDeformation is carried out, new second phase heart graph structure B is obtainedi+1；
Step 5 calculates second phase heart graph structure BiWith new second phase heart graph structure Bi+1Changing value, when When changing value is greater than preset threshold, i=i+1 is enabled, returns to step 3；Otherwise, using the movement of deformation function characterization heart.
Optionally, first phase heart graph structure A, second phase heart graph structure Bi be respectively adopted four-tuple P1, E1, G1, H1 }, { P2, E2, G2, H2 } indicate；
For n in first phase heart graph structure A1A dpThe point on the cardiod diagram vertex of dimension is special Collection is closed；
For m in first phase heart graph structure A1A deThe side characteristic set of dimension；
G1And H1The set of the cardiod diagram beginning and end on side is constituted in respectively first phase heart graph structure A,Vertex set is made of endocardial apex and epicardial apex.
For n in first phase heart graph structure A2A dpThe point on the cardiod diagram vertex of dimension is special Collection is closed；
For m in first phase heart graph structure A2A deThe side characteristic set of dimension；
G2And H2Respectively second phase heart graph structure BiThe set of the middle cardiod diagram beginning and end for constituting side,
Optionally, it in step 1, is sampled in endocardial contours and epicardial contours, respectively obtains several the heart Inner membrance vertex and epicardial apex, then figure in step 3 matches convex objective function and includes：
Wherein,Ax-b=0；
X=vec (Kp), y=vec (Y), p2=vec (P2), e2=vec (E2), vec is vectorization operator；
It is long-pending for Hadamard,For Kronecker product, λ, γ are constant；
1m×nIt is the m * n matrix that element is all 1, InIt is the unit matrix of n × n；
n1、n2In respectively first phase heart graph structure A and second phase heart graph structure Bi outside endocardial apex and the heart The number on film vertex；
Optionally, the vertex set with corresponding relationship includes the vertex set U=of the first phase heart graph structure A {ui, i=1 ... K } and the second phase heart graph structure BiVertex set V={ vi, i=1 ..., K }；The point feature Set includes the vertex position on the cardiod diagram vertex, then step 4 includes：
Objective function, which is solved, using iteration threshold contraction algorithm obtains affine deformation coefficient and elastic registration coefficient, target letter Number is：
φ is radial basis function, | | * | |2For Euclidean distance, z=[α1,...αK,β0,β1,β2]T, elastic registration coefficient is α =[α1,...αK]T, affine deformation coefficient is β=[β0,β1,β2]T, Dc=[0,1,1]T, λ1And λ2For constant,R For real number field；
It constructs to obtain deformation function using affine deformation coefficient, elastic registration coefficient and the radial basis function：
Using the deformation function to the second phase heart graph structure BiPoint feature set P2In vertex position into Row mapping obtains new vertex, and connects the center and the new top of epicardial contours described in the second phase MR image Point obtains new second phase heart graph structure Bi+1。
Optionally, in step 5, changing value is the changing value between the point position of mapping front and back.
Further, the present invention also provides a kind of representation of athletic device of heart, which includes processor, memory And communication bus；
Communication bus is for realizing the connection communication between processor and memory；
Processor is for executing one or more program stored in memory, to realize following steps：
Step 1, the first phase magnetic resonance MR image and second phase MR image for extracting heart, using full convolutional Neural net Network divides first phase MR image, second phase MR image respectively, obtains the endocardial contours and epicardial contours of left ventricle, and It is sampled in endocardial contours and epicardial contours, obtains several cardiod diagram vertex；
Step 2, the center for connecting first phase MR picture centre epicardium contours and the cardiod diagram on first phase MR image Vertex obtains first phase heart graph structure A；Connect center and the second phase MR of second phase MR picture centre epicardium contours Cardiod diagram vertex on image obtains second phase heart graph structure Bi；The initial value of i is 1；
Step 3 utilizes first phase heart graph structure A and second phase heart graph structure BiIt solves figure and matches convex target letter Number, obtain the first phase heart graph structure A vertex and the second phase heart graph structure BiIn have corresponding relationship Vertex set；
Step 4 constructs deformation function based on vertex set, and using deformation function to second phase heart graph structure BiInto Row deformation obtains new second phase heart graph structure Bi+1；
Step 5 calculates second phase heart graph structure BiWith new second phase heart graph structure Bi+1Changing value, when When changing value is greater than preset threshold, i=i+1 is enabled, returns to step 3；Otherwise, using the movement of deformation function characterization heart.
Optionally, processor is for executing one or more program stored in memory, to realize：Using four-tuple {P1、E1、G1、H1}、{P2、E2、G2、H2Indicate first phase heart graph structure A, second phase heart graph structure Bi；
For n in first phase heart graph structure A1A dpThe point on the cardiod diagram vertex of dimension is special Collection is closed；
For m in first phase heart graph structure A1A deThe side characteristic set of dimension；
G1And H1The set of the cardiod diagram beginning and end on side is constituted in respectively first phase heart graph structure A,Vertex set is made of endocardial apex and epicardial apex.
For second phase heart graph structure BiMiddle n2A dpThe point on the cardiod diagram vertex of dimension Characteristic set；
For second phase heart graph structure BiMiddle m2A deThe side characteristic set of dimension；
G2And H2Respectively second phase heart graph structure BiThe set of the middle cardiod diagram beginning and end for constituting side,
Optionally, it in step 1, is sampled in endocardial contours and epicardial contours, respectively obtains several the heart Inner membrance vertex and epicardial apex, then figure in step 3 matches convex objective function and includes：
Wherein,Ax-b=0；
X=vec (Kp), y=vec (Y), p2=vec (P2), e2=vec (E2), vec is vectorization operator；
It is long-pending for Hadamard,For Kronecker product, λ, γ are constant；
1m×nIt is the m * n matrix that element is all 1, InIt is the unit matrix of n × n；
n1、n2In respectively first phase heart graph structure A and second phase heart graph structure Bi outside endocardial apex and the heart The number on film vertex；
Optionally, the vertex set with corresponding relationship includes the vertex set U of the first phase heart graph structure A ={ ui, i=1 ... K } and the second phase heart graph structure BiVertex set V={ vi, i=1 ..., K }；The point is special The vertex position including the cardiod diagram vertex is closed in collection；
Processor is for executing one or more program stored in memory, to realize step 4：
Objective function, which is solved, using iteration threshold contraction algorithm obtains affine deformation coefficient and elastic registration coefficient, target letter Number is：
φ is radial basis function, | | * | |2For Euclidean distance, z=[α1,...αK,β0,β1,β2]T, elastic registration coefficient is α =[α1,...αK]T, affine deformation coefficient is β=[β0,β1,β2]T, Dc=[0,1,1]T, λ1And λ2For constant,R For real number field；
It constructs to obtain deformation function using affine deformation coefficient, elastic registration coefficient and the radial basis function：
Using deformation function to second phase heart graph structure BiPoint feature set P2In vertex position map To new vertex, and the center and new vertex of second phase MR picture centre epicardium contours are connected, obtains new second phase Heart graph structure Bi+1
Further, the present invention also provides a kind of computer readable storage medium, computer readable storage medium is contained One or more program, one or more program can be executed by one or more processor, to realize heart as above Representation of athletic method the step of.
Beneficial effect
The present invention provides the representation of athletic method, apparatus and computer readable storage medium of a kind of heart, passes through following step The rapid characterization that can be achieved to heart movement：
Step 1, the first phase magnetic resonance MR image and second phase MR image for extracting heart, using full convolutional Neural net Network divides first phase MR image, second phase MR image respectively, obtains the endocardial contours and epicardial contours of left ventricle, and It is sampled in endocardial contours and epicardial contours, obtains several endocardial apex and epicardial apex；
Step 2, the center for connecting first phase MR picture centre epicardium contours and endocardial apex and epicardial apex, obtain To first phase heart graph structure A；It connects outside the center and endocardial apex and the heart of second phase MR picture centre epicardium contours Film vertex obtains second phase heart graph structure Bi；The initial value of i is 1；
Step 3 utilizes first phase heart graph structure A and second phase heart graph structure BiIt solves figure and matches convex target letter Number obtains vertex and the second phase heart graph structure B of first phase heart graph structure AiIn with corresponding relationship vertex set It closes；
Step 4 constructs deformation function based on the vertex set with corresponding relationship, and using deformation function to second phase Heart graph structure BiDeformation is carried out, new second phase heart graph structure B is obtainedi+1；
Step 5 calculates second phase heart graph structure BiWith new second phase heart graph structure Bi+1Changing value, when When changing value is greater than preset threshold, i=i+1 is enabled, returns to step 3；Otherwise, using the movement of deformation function characterization heart.
After the representation of athletic method of the heart constructs to obtain deformation function by the way of circulation, also obtained using building Deformation function deformation is carried out to last time obtained heart graph structure, calculate the changing value of the heart graph structure before and after deformation into And the whether accurate of deformation function is verified, to obtain to characterize the deformation function of the movement of heart, this method is relative to existing Technology can more accurately characterize the movement of heart.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this Some embodiments of invention for those skilled in the art without creative efforts, can also basis These attached drawings obtain other attached drawings.
Fig. 1 is the basic flow chart of the representation of athletic method of heart provided in an embodiment of the present invention；
Fig. 2 is the heart graph structure that existing Delaunay Triangulation algorithm subdivision obtains；
Fig. 3 is the heart graph structure of construction of the embodiment of the present invention；
Fig. 4 is the structural schematic diagram of the representation of athletic device of heart provided in an embodiment of the present invention.
Specific embodiment
In order to make the invention's purpose, features and advantages of the invention more obvious and easy to understand, below in conjunction with the present invention Attached drawing in embodiment, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described reality Applying example is only a part of the embodiment of the present invention, and not all embodiments.Based on the embodiments of the present invention, those skilled in the art Member's every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
Referring to Fig. 1, Fig. 1 is the basic flow chart of the representation of athletic method of heart provided in this embodiment, this method packet It includes：
S101, first phase MR image and second phase MR image are extracted, and it is split and is sampled respectively, obtained Several cardiod diagram vertex.
Specifically, the S101 step includes：The first phase magnetic resonance MR image and second phase MR image of heart are extracted, Divide first phase MR image, second phase MR image respectively using full convolutional neural networks, obtains first phase MR image The endocardial contours and epicardial contours of left ventricle and the left ventricle inner membrance profile and epicardial contours of second phase MR image, And sampled on first phase MR image, the endocardial contours of second phase MR image and epicardial contours respectively, it obtains Several cardiod diagram vertex on to corresponding phase MR image.
It is to be appreciated that the full convolutional neural networks of the S101 step are to take a large amount of hearts with artificial annotation results Dirty image data is trained full convolutional neural networks, and optional cardiac image data includes Canadian Duolun at present Whatever amount 33 cardiac short axis MR images that virgin hospital image diagnostics division provides, which provides left ventricle in 20 phases Segmentation result, it is possible to provide about more than 5,000 training datas；Another training set is the cardiac image number of University of Toronto According to the data share 45 patient datas, and each data provide the left ventricle segmentation result of end-systole and diastasis, can To provide about more than 500 training datas.MR image is split to obtain a left side by the full convolutional neural networks that training obtains The endocardial contours and epicardial contours of ventricle further increase segmentation precision in other some examples, solve complete Convolutional network divides the case where existing over-segmentation, and raising segmentation precision is then realized using ellipses detection.
The center and cardiod diagram vertex of S102, epicardial contours on the identical MR image of connection, respectively obtain first phase Heart graph structure A and second phase heart graph structure Bi。
The S102 step includes：It connects on center and the first phase MR image of first phase MR picture centre epicardium contours Cardiod diagram vertex, obtain first phase heart graph structure A；Connect the center and the of second phase MR picture centre epicardium contours Cardiod diagram vertex on two phase MR image obtains second phase heart graph structure Bi；The initial value of i is 1；
It is shown in Figure 2, be the embodiment of the present invention in method formed heart graph structure, be different from it is shown in Fig. 3, The heart graph structure obtained by Delaunay Triangulation algorithm subdivision, the heart graph structure description obtained by the method for the invention The nearly round structure of left ventricle, connection side has anisotropic directionality, and its length has similitude, and each edge is all With specific geometric interpretation, the similarity measurement on side has specific Auxiliary Significance to figure similitude, is well suited for describing left ventricle Anatomical structure.
In other some examples, first phase heart graph structure A, second phase heart graph structure BiIt can be respectively adopted Four-tuple { P1、E1、G1、H1}、{P2、E2、G2、H2Indicate；
For n in first phase heart graph structure A1A dpThe point on the cardiod diagram vertex of dimension is special Collection is closed；
For m in first phase heart graph structure A1A deThe side characteristic set of dimension；
G1And H1The set of the cardiod diagram beginning and end on side is constituted in respectively first phase heart graph structure A,Vertex set is made of endocardial apex and epicardial apex.
For second phase heart graph structure BiMiddle n2A dpThe point on the cardiod diagram vertex of dimension Characteristic set；
For second phase heart graph structure BiMiddle m2A deThe side characteristic set of dimension；
G2And H2Respectively second phase heart graph structure BiThe set of the middle cardiod diagram beginning and end for constituting side,
It is to be appreciated that point feature set is the set of the feature vector composition on all cardiod diagram vertex herein；Bian Te The set that as all sides are constituted is closed in collection, and side herein is that the center for connecting epicardial contours and cardiod diagram vertex are constituted Side；G1And H1It is two matrixes, respectively indicates each two endvertex of side, G on first phase graph structure2And H2It is two matrixes, point It Biao Shi not each two endvertex of side on second phase graph structure.
S103, first phase heart graph structure A and second phase heart graph structure B are utilizediIt solves figure and matches convex target letter Number, obtains the vertex set with corresponding relationship.
Specifically, S103 is including the use of first phase heart graph structure A and second phase heart graph structure BiSolve figure With convex objective function, vertex and the second phase heart graph structure B of first phase heart graph structure A is obtainediIn there is corresponding close The vertex set of system.
It is to be appreciated that in other examples, several cardiod diagram vertex that step 1 obtains include endocardial apex And epicardial apex, the figure in step 3 match convex objective function and include：
Wherein,Ax-b=0；
X=vec (Kp), y=vec (Y), p2=vec (P2), e2=vec (E2), vec is vectorization operator；
It is long-pending for Hadamard,For Kronecker product, λ, γ are constant；
1m×nIt is the m * n matrix that element is all 1, InIt is the unit matrix of n × n；
n1、n2Respectively first phase heart graph structure A and second phase heart graph structure BiOutside middle endocardial apex and the heart The number on film vertex；
It is to be appreciated that using submatrix KpAnd KqRespectively indicate first phase heart graph structure A and the second phase heart Dirty graph structure BiVertex correspondence degree and side degree of correspondence, first phase heart graph structure A and second phase heart graph structure Bi Point feature set be respectively P1And P2, corresponding relationship matrix K between pointpIf letter can be used under conditions of meeting permutation matrix Number | | P2-P1Kp||2The departure degree between point set is measured, similarly, can equally use the line set E of two figures1And E2With And side corresponding relationship matrix KqMeasure the departure degree between side | | E2-E1Kq||2, wherein
S104, deformation function is constructed based on the vertex set with corresponding relationship, and using deformation function to second phase Heart graph structure BiDeformation is carried out, new second phase heart graph structure B is obtainedi+1。
In other examples, the vertex set with corresponding relationship includes the vertex set U of first phase heart graph structure A ={ ui, i=1 ... K } and second phase heart graph structure BiVertex set V={ vi, i=1 ..., K }, vertex set herein For the subset of point feature set, while it is to be appreciated that point feature set further includes the vertex position on cardiod diagram vertex, this When, deformation function is constructed based on vertex set, and deformation is carried out to second phase heart graph structure Bi using deformation function, obtained The step of new second phase heart graph structure Bi+1 includes：
S1041, affine deformation coefficient and elastic registration coefficient are obtained using iteration threshold contraction algorithm solution objective function, Objective function is：
φ is radial basis function, | | * | |2For Euclidean distance, z=[α1,...αK,β0,β1,β2]T, elastic registration coefficient is α =[α1,...αK]T, affine deformation coefficient is β=[β0,β1,β2]T, Dc=[0,1,1]T, λ1And λ2For constant,R For real number field；
S1042, it constructs to obtain deformation function using affine deformation coefficient, elastic registration coefficient and radial basis function：
S1043, using deformation function to second phase heart graph structure BiPoint feature set P2In vertex position carry out Mapping obtains new vertex, and connects the center and new vertex of second phase MR picture centre epicardium contours, obtains new the Two phase heart graph structure Bi+1。
It is to be appreciated that R is real number field, R2Indicate two-dimensional surface；U indicates that the point on two-dimensional surface, u1 indicate the Point on one-dimensional plane, u2 indicate the point on the second dimensional plane.
S105, second phase heart graph structure B is calculatediWith new second phase heart graph structure Bi+1Changing value, and sentence Whether disconnected changing value is greater than preset threshold.
Specifically, then enabling i=i+1 when changing value is greater than preset threshold, returns and execute S103；If it is not, then executing S106：Using the movement of deformation function characterization heart.
In other example, the changing value in S106 is second phase heart graph structure BiWith the new second phase heart Dirty graph structure Bi+1The changing value of the position on cardiod diagram vertex.
After the representation of athletic method of heart provided in this embodiment constructs to obtain deformation function by the way of circulation, also Deformation is carried out to original image structure using the deformation function that building obtains, is according to what deformation changing value further verified deformation function It is no accurate, therefore this method can accurately characterize the movement of heart compared with the existing technology.Meanwhile this method not only can solve The dynamic partition of cardiac image also can be obtained by deformation model as a result, with accessory heart disease for cardiac motion model estimation problem The clinical applications such as disease diagnosis, cardiac function measurement, analysis.
It is shown in Figure 4 the present invention also provides a kind of representation of athletic device of heart comprising processor 41, storage Device 42 and communication bus 43, wherein：
Communication bus 43 is for realizing the connection communication between processor 41 and memory 42；
Processor 41 for executing the program stored in memory 42, with realize above-mentioned heart representation of athletic method it is each Step.
The present embodiment additionally provides a kind of computer readable storage medium, which is characterized in that computer readable storage medium It is stored with one or more program, one or more program can be executed by one or more processor, as above to realize The step of representation of athletic method of the heart.
It should be noted that for the various method embodiments described above, describing for simplicity, therefore, it is stated as a series of Combination of actions, but those skilled in the art should understand that, the present invention is not limited by the sequence of acts described because According to the present invention, certain steps can use other sequences or carry out simultaneously.Secondly, those skilled in the art should also know It knows, the embodiments described in the specification are all preferred embodiments, and related actions and modules might not all be this hair Necessary to bright.
In the above-described embodiments, it all emphasizes particularly on different fields to the description of each embodiment, there is no the portion being described in detail in some embodiment Point, it may refer to the associated description of other embodiments, meanwhile, the serial number of the above embodiments of the invention is only for description, does not represent The superiority and inferiority of embodiment, those skilled in the art under the inspiration of the present invention, are not departing from present inventive concept and right is wanted It asks under protected ambit, can also make many forms, all of these belong to the protection of the present invention.

Claims (10)

1. a kind of representation of athletic method of heart, which is characterized in that including：
Step 1, the first phase magnetic resonance MR image and second phase MR image for extracting heart, using full convolutional neural networks point Do not divide the first phase MR image, second phase MR image, obtains the endocardial contours and epicardial contours of left ventricle, and It is sampled in the endocardial contours and epicardial contours, obtains several cardiod diagram vertex；
On the center and the first phase MR image of epicardial contours described in step 2, the connection first phase MR image The cardiod diagram vertex, obtain first phase heart graph structure A；Connect external membrane of heart wheel described in the second phase MR image The cardiod diagram vertex on wide center and the second phase MR image, obtains second phase heart graph structure Bi；The i Initial value be 1；
Step 3 utilizes the first phase heart graph structure A and the second phase heart graph structure BiIt solves figure and matches convex mesh Scalar functions, obtain the first phase heart graph structure A vertex and the second phase heart graph structure BiIn have correspond to The vertex set of relationship；
Step 4 constructs deformation function based on the vertex set with corresponding relationship, and using the deformation function to described Second phase heart graph structure BiDeformation is carried out, new second phase heart graph structure B is obtainedi+1；
Step 5 calculates the second phase heart graph structure BiWith the new second phase heart graph structure Bi+1Changing value, When the changing value is greater than preset threshold, i=i+1 is enabled, returns to step 3；Otherwise, institute is characterized using the deformation function State the movement of heart.
2. the representation of athletic method of heart as described in claim 1, which is characterized in that the first phase heart graph structure A, Second phase heart graph structure BiFour-tuple { P is respectively adopted1、E1、G1、H1}、{P2、E2、G2、H2Indicate；
It is describedFor n in the first phase heart graph structure A1A dpThe cardiod diagram vertex of dimension Point feature set；
It is describedFor m in the first phase heart graph structure A1A deThe side characteristic set of dimension；
The G1And H1The set of the cardiod diagram beginning and end on side, institute are constituted in the respectively described first phase heart graph structure A It states
It is describedFor the second phase heart graph structure BiMiddle n2A dpThe cardiod diagram vertex of dimension Point feature set；
It is describedFor the second phase heart graph structure BiMiddle m2A deThe side characteristic set of dimension；
The G2And H2The respectively described second phase heart graph structure BiThe set of the middle cardiod diagram beginning and end for constituting side, It is described
3. the representation of athletic method of heart as claimed in claim 2, which is characterized in that the cardiod diagram vertex in the step 1 Including endocardial apex and epicardial apex, then the convex objective function of figure matching in the step 3 includes：
Wherein,Ax-b=0；
X=vec (the Kp), y=vec (Y), p2=vec (P2), e2=vec (E2), the vec is vectorization operator；
It is describedIt is described for Hadamard productFor Kronecker product, described λ, γ are constant；
Described 1m×nIt is the m * n matrix that element is all 1, the InIt is the unit matrix of n × n；
The n1、n2The respectively described first phase heart graph structure A and second phase heart graph structure BiMiddle internal membrane of heart top The number of point and epicardial apex；
It is describedIt is described
4. the representation of athletic method of heart as described in any one of claims 1-3, which is characterized in that described that there is corresponding relationship Vertex set include the first phase heart graph structure A vertex set U={ ui, i=1 ... K } and the second phase Heart graph structure BiVertex set V={ vi, i=1 ..., K }；The point feature set includes the vertex on the cardiod diagram vertex Position；
The step 4 includes：
Objective function, which is solved, using iteration threshold contraction algorithm obtains affine deformation coefficient and elastic registration coefficient, the target letter Number is：
It is described
The φ is radial basis function, described | | * | |2For Euclidean distance, the z=[α1,...αK,β0,β1,β2]T, the elasticity Transformation coefficient is α=[α1,...αK]T, the affine deformation coefficient is β=[β0,β1,β2]T, the Dc=[0,1,1]T, described λ1And λ2It is described for constantThe R is real number field；
It constructs to obtain deformation function using the affine deformation coefficient, the elastic registration coefficient and the radial basis function：
Using the deformation function to the second phase heart graph structure BiPoint feature set P2In vertex position reflected It penetrates to obtain new vertex, and connects the center and the new vertex of epicardial contours described in the second phase MR image, Obtain new second phase heart graph structure Bi+1。
5. the representation of athletic method of heart as claimed in claim 4, which is characterized in that in the step 5, the changing value is The second phase heart graph structure BiWith the new second phase heart graph structure Bi+1The variation of the position on cardiod diagram vertex Value.
6. a kind of representation of athletic device of heart, which is characterized in that described device includes processor, memory and communication bus；
The communication bus is for realizing the connection communication between processor and memory；
The processor is for executing one or more program stored in memory, to realize following steps：
Step 1, the first phase magnetic resonance MR image and second phase MR image for extracting heart, using full convolutional neural networks point Do not divide the first phase MR image, second phase MR image, obtains the endocardial contours and epicardial contours of left ventricle, and It is sampled in the endocardial contours and epicardial contours, obtains several cardiod diagram vertex；
On the center and the first phase MR image of epicardial contours described in step 2, the connection first phase MR image The cardiod diagram vertex, obtain first phase heart graph structure A；Connect external membrane of heart wheel described in the second phase MR image The cardiod diagram vertex on wide center and the second phase MR image, obtains second phase heart graph structure Bi；The i Initial value be 1；
Step 3 utilizes the first phase heart graph structure A and the second phase heart graph structure BiIt solves figure and matches convex mesh Scalar functions, obtain the first phase heart graph structure A vertex and the second phase heart graph structure BiIn have correspond to The vertex set of relationship；
Step 4 constructs deformation function based on the vertex set with corresponding relationship, and using the deformation function to described Second phase heart graph structure BiDeformation is carried out, new second phase heart graph structure B is obtainedi+1；
Step 5 calculates the second phase heart graph structure BiWith the new second phase heart graph structure Bi+1Changing value, When the changing value is greater than preset threshold, i=i+1 is enabled, returns to step 3；Otherwise, institute is characterized using the deformation function State the movement of heart.
7. the representation of athletic device of heart as claimed in claim 6, which is characterized in that the processor is for executing memory One or more program of middle storage, to realize：Using four-tuple { P1、E1、G1、H1}、{P2、E2、G2、H2Indicate described the One phase heart graph structure A, second phase heart graph structure Bi；
It is describedFor n in the first phase heart graph structure A1A dpThe cardiod diagram vertex of dimension Point feature set；
It is describedFor m in the first phase heart graph structure A1A deThe side characteristic set of dimension；
The G1And H1The set of the cardiod diagram beginning and end on side, institute are constituted in the respectively described first phase heart graph structure A It states
It is describedFor the second phase heart graph structure BiMiddle n2A dpThe cardiod diagram vertex of dimension Point feature set；
It is describedFor the second phase heart graph structure BiMiddle m2A deThe side characteristic set of dimension；
The G2And H2The respectively described second phase heart graph structure BiThe set of the middle cardiod diagram beginning and end for constituting side, It is described
8. the representation of athletic device of heart as claimed in claim 7, which is characterized in that the cardiod diagram vertex in the step 1 Including endocardial apex and epicardial apex, then the convex objective function of figure matching in the step 3 includes：
Wherein,Ax-b=0；
X=vec (the Kp), y=vec (Y), p2=vec (P2), e2=vec (E2), the vec is vectorization operator；
It is describedIt is described for Hadamard productFor Kronecker product, described λ, γ are constant；
Described 1m×nIt is the m * n matrix that element is all 1, the InIt is the unit matrix of n × n；
The n1、n2Internal membrane of heart top in the respectively described first phase heart graph structure A and second phase heart graph structure Bi The number of point and epicardial apex；
It is describedIt is described
9. such as the representation of athletic device of the described in any item hearts of claim 6-8, which is characterized in that described that there is corresponding relationship Vertex set include the first phase heart graph structure A vertex set U={ ui, i=1 ... K } and the second phase Heart graph structure BiVertex set V={ vi, i=1 ..., K }；The point feature set includes the vertex on the cardiod diagram vertex Position；
The processor is for executing one or more program stored in memory, to realize the step 4：
Objective function, which is solved, using iteration threshold contraction algorithm obtains affine deformation coefficient and elastic registration coefficient, the target letter Number is：
It is described
The φ is radial basis function, described | | * | |2For Euclidean distance, the z=[α1,...αK,β0,β1,β2]T, the elasticity Transformation coefficient is α=[α1,...αK]T, the affine deformation coefficient is β=[β0,β1,β2]T, the Dc=[0,1,1]T, described λ1And λ2It is described for constantThe R is real number field；
It constructs to obtain deformation function using the affine deformation coefficient, the elastic registration coefficient and the radial basis function：
Using the deformation function to the second phase heart graph structure BiPoint feature set P2In vertex position reflected It penetrates to obtain new vertex, and connects the center and the new vertex of epicardial contours described in the second phase MR image, Obtain new second phase heart graph structure Bi+1。
10. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage have one or Multiple programs, one or more of programs can be executed by one or more processor, to realize such as claim 1 to 5 Any one of described in heart representation of athletic method the step of.
CN201810730181.1A 2018-07-05 2018-07-05 A kind of the representation of athletic method, apparatus and computer readable storage medium of heart Pending CN108898622A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810730181.1A CN108898622A (en) 2018-07-05 2018-07-05 A kind of the representation of athletic method, apparatus and computer readable storage medium of heart

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810730181.1A CN108898622A (en) 2018-07-05 2018-07-05 A kind of the representation of athletic method, apparatus and computer readable storage medium of heart

Publications (1)

Publication Number Publication Date
CN108898622A true CN108898622A (en) 2018-11-27

Family

ID=64347729

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810730181.1A Pending CN108898622A (en) 2018-07-05 2018-07-05 A kind of the representation of athletic method, apparatus and computer readable storage medium of heart

Country Status (1)

Country Link
CN (1) CN108898622A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110136111A (en) * 2019-05-14 2019-08-16 深圳大学 A kind of heart movement estimation method, system and terminal device
CN110148150A (en) * 2019-06-20 2019-08-20 深圳大学 It is dense to connect asymmetric hierarchical network training method and heart movement field estimation method
CN110400298A (en) * 2019-07-23 2019-11-01 中山大学 Detection method, device, equipment and the medium of heart clinical indices
CN111784732A (en) * 2020-06-28 2020-10-16 深圳大学 Method and system for training cardiac motion field estimation model and cardiac motion field estimation

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102890824A (en) * 2011-07-19 2013-01-23 株式会社东芝 Method and device for tracking contour of motion object and method and device for analyzing myocardial motion
CN103761745A (en) * 2013-07-31 2014-04-30 深圳大学 Estimation method and system for lung motion model
CN104933716A (en) * 2015-06-16 2015-09-23 山东大学（威海） Non-rigid registration method applied to medical image
US20170140530A1 (en) * 2015-11-17 2017-05-18 Shanghai United Imaging Healthcare Co., Ltd. Systems and methods for image processing in magnetic resonance imaging
CN108042154A (en) * 2017-12-08 2018-05-18 浙江中医药大学 Two dimensional echocardiogram formation center flesh shape, movement and deformation analysis method
CN108230342A (en) * 2017-12-29 2018-06-29 电子科技大学 A kind of left and right ventricles level-set segmentation methods based on cardiac anatomy

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102890824A (en) * 2011-07-19 2013-01-23 株式会社东芝 Method and device for tracking contour of motion object and method and device for analyzing myocardial motion
CN103761745A (en) * 2013-07-31 2014-04-30 深圳大学 Estimation method and system for lung motion model
CN104933716A (en) * 2015-06-16 2015-09-23 山东大学（威海） Non-rigid registration method applied to medical image
US20170140530A1 (en) * 2015-11-17 2017-05-18 Shanghai United Imaging Healthcare Co., Ltd. Systems and methods for image processing in magnetic resonance imaging
CN108042154A (en) * 2017-12-08 2018-05-18 浙江中医药大学 Two dimensional echocardiogram formation center flesh shape, movement and deformation analysis method
CN108230342A (en) * 2017-12-29 2018-06-29 电子科技大学 A kind of left and right ventricles level-set segmentation methods based on cardiac anatomy

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
JONATHAN LONG等: "Fully Convolutional Networks for Semantic Segmentation", 《2015 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)》 *
WEI GUO等: "Left ventricle motion estimation for cardiac cine MRI using graph matching", 《2017 IEEE INTERNATIONAL CONFERENCE ON BIOINFORMATICS AND BIOMEDICINE (BIBM)》 *
XUAN YANG等: "Robust landmark-based image registration using l1 and l2 norm regularizations", 《2015 IEEE INTERNATIONAL CONFERENCE ON BIOINFORMATICS AND BIOMEDICINE (BIBM)》 *
刁晨: "心电信号质量评估与心率变异性分析方法研究", 《中国优秀博士学位论文全文数据库 医药卫生科技辑（月刊）》 *

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110136111A (en) * 2019-05-14 2019-08-16 深圳大学 A kind of heart movement estimation method, system and terminal device
CN110136111B (en) * 2019-05-14 2021-08-10 深圳大学 Heart motion estimation method, system and terminal equipment
CN110148150A (en) * 2019-06-20 2019-08-20 深圳大学 It is dense to connect asymmetric hierarchical network training method and heart movement field estimation method
CN110148150B (en) * 2019-06-20 2021-07-02 深圳大学 Dense connection asymmetric hierarchical network training method and cardiac motion field estimation method
CN110400298A (en) * 2019-07-23 2019-11-01 中山大学 Detection method, device, equipment and the medium of heart clinical indices
CN110400298B (en) * 2019-07-23 2023-10-31 中山大学 Method, device, equipment and medium for detecting heart clinical index
CN111784732A (en) * 2020-06-28 2020-10-16 深圳大学 Method and system for training cardiac motion field estimation model and cardiac motion field estimation
CN111784732B (en) * 2020-06-28 2023-07-28 深圳大学 Method and system for training heart motion field estimation model and heart motion field estimation

Similar Documents

Publication Publication Date Title
KR101855369B1 (en) Systems and methods for estimating physiological heart measurements from medical images and clinical data
CN108898622A (en) A kind of the representation of athletic method, apparatus and computer readable storage medium of heart
US10296809B2 (en) Method and system for fast patient-specific cardiac electrophysiology simulations for therapy planning and guidance
US11664125B2 (en) System and method for deep learning based cardiac electrophysiology model personalization
CN106535741B (en) System and method for characterizing electrical properties of the heart from medical images and body surface potentials
Sermesant et al. An electromechanical model of the heart for image analysis and simulation
CN109993730B (en) 3D/2D blood vessel registration method and device
Faghih Roohi et al. 4D statistical shape modeling of the left ventricle in cardiac MR images
US9514280B2 (en) Method and apparatus for creating model of patient specified target organ based on blood vessel structure
CN105264533A (en) Method and system for interactive computation of cardiac electromechanics
RU2669680C2 (en) View classification-based model initialisation
CN113112486B (en) Tumor motion estimation method and device, terminal equipment and storage medium
US9092666B2 (en) Method and apparatus for estimating organ deformation model and medical image system
US9196049B2 (en) Method and system for regression-based 4D mitral valve segmentation from 2D+t magnetic resonance imaging slices
Wong et al. Velocity-based cardiac contractility personalization from images using derivative-free optimization
Meister et al. Graph convolutional regression of cardiac depolarization from sparse endocardial maps
Johnston Accuracy of electrocardiographic imaging using the method of fundamental solutions
CN116130090A (en) Ejection fraction measuring method and device, electronic device, and storage medium
US11663727B2 (en) Systems and methods for non-invasive cardiac assessment
WO2000048509A1 (en) Methods of processing tagged mri data indicative of tissue motion including 4-d lv tissue tracking
Goyal et al. MRI image based patient specific computational model reconstruction of the left ventricle cavity and myocardium
EP3614338B1 (en) Post-mapping automatic identification of pulmonary veins
Gil et al. Manifold parametrization of the left ventricle for a statistical modelling of its complete anatomy
Piuze et al. Moving frames for heart fiber reconstruction
CN116077093A (en) Early myocardial infarction detection device, method and medium based on echocardiography

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
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20181127