CN113223061B - Elastic registration method and device for medical image - Google Patents
Elastic registration method and device for medical image Download PDFInfo
- Publication number
- CN113223061B CN113223061B CN202110600433.0A CN202110600433A CN113223061B CN 113223061 B CN113223061 B CN 113223061B CN 202110600433 A CN202110600433 A CN 202110600433A CN 113223061 B CN113223061 B CN 113223061B
- Authority
- CN
- China
- Prior art keywords
- deformation field
- energy functional
- epsilon
- total energy
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000004800 variational method Methods 0.000 claims abstract description 17
- 238000009499 grossing Methods 0.000 claims abstract description 8
- 238000007667 floating Methods 0.000 claims description 14
- 239000011159 matrix material Substances 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 description 9
- 230000009466 transformation Effects 0.000 description 6
- 238000012545 processing Methods 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 238000002591 computed tomography Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013421 nuclear magnetic resonance imaging Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000002600 positron emission tomography Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
-
- G06T5/70—
-
- 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
Abstract
The invention provides a medical image elastic registration method, which comprises the following steps: energy functional ε in Demons algorithm 1 On the basis of (U), increasing the smoothness constraint term epsilon of the deformation field 2 (U) and increasing the symmetric Kullback-Leibler distance ε 3 (U) constructing a total energy functional epsilon (U); wherein the total energy functional is a multi-dimensional parameter functional; and obtaining a group of parameters capable of minimizing the total energy functional through a variational method to obtain an optimal deformation field. Based on the energy functional, the partial derivative square sum of the deformation field is added and used as a smooth constraint term for smoothing the deformation field; the symmetrical Kullback-Leibler distance is increased to maintain an image topological structure, non-rigid registration is converted into an energy functional minimization problem, and a set of parameters capable of enabling the energy functional to be minimized is searched through a variational method to obtain an optimal deformation field.
Description
Technical Field
The invention relates to the technical field of medical image processing, in particular to a medical image elastic registration method and device.
Background
The medical image acquisition modes comprise computer tomography, nuclear magnetic resonance imaging, positron emission tomography and the like, information differences exist among different mode images, and the different mode images are fused through image processing, so that the clinical diagnosis and treatment level can be improved.
Image registration is largely divided into rigid and non-rigid image registration as a means of medical image processing. In the prior art, non-rigid registration of medical images is typically processed using a Demons algorithm that treats a reference image and a floating image as two images in a sequence of continuous motion images, and calculates motion vectors from the floating image to the reference image. The traditional Demons algorithm uses a free deformation model, and the space transformation with high freedom degree can not meet the smoothing requirement; the linear regularization method of the Gaussian filter provided according to the regularization theory is only suitable for treating the problem of small deformation; in addition, there is a differential homoembryo degons algorithm, which combines a degons algorithm with a prune group structure framework and a prune group optimization algorithm of a differential homoembryo transformation space, and the differential homoembryo degons algorithm has the advantages that the topological structure after image deformation can be kept unchanged, but the algorithm can not guarantee the differential homoembryo characteristics on a discrete domain, and is also influenced by various factors, such as approximation, difference and the like, during actual calculation, so that the differential homoembryo characteristics of the space transformation are further influenced.
Therefore, there is a need to propose a new non-rigid registration method for medical images.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to provide a medical image elastic registration method and a medical image elastic registration device.
The technical scheme of the invention is summarized as follows:
in one aspect, the present invention provides a medical image elastic registration method, comprising:
energy functional ε in Demons algorithm 1 On the basis of (U), increasing the smoothness constraint term epsilon of the deformation field 2 (U) and increasing the symmetric Kullback-Leibler distance ε 3 (U) constructing a total energy functional epsilon (U); wherein the total energy functional is a multi-dimensional parameter functional;
and obtaining a group of parameters capable of minimizing the total energy functional through a variational method to obtain an optimal deformation field.
Further, the energy function ε of the Demons algorithm 1 (U) is:
wherein sigma i Is the noise level, sigma, of the image x Is the iteration step length;
smoothness constraint term epsilon of deformation field 2 (U) is:
symmetric Kullback-Leibler distance ε 3 (U) is:
the total energy function is obtained as:
wherein 0 is<λ 1 ,λ 2 <1, a step of; a (X) is a reference image, B (X, U) is a floating image, R (X, U) =b (X, U) -a (X), representing the difference between the reference image and the floating image; omega shape 3 Representing the deformation region.
Further, DU is Jacobian determinant |Du i,j,k Matrix of | du| is defined as:
wherein,
the minimization formula is obtained:
minimisation formula (2) approximates the Jacobian determinant distribution of forward and reverse transforms to a constant map.
Further, obtaining a set of parameters capable of minimizing the total energy functional to obtain an optimal deformation field through a variance method comprises:
using a variational method to calculate epsilon (U) to obtain an Euler-Lagrange equation taking a deformation field U as a variable, and rewriting;
and solving the iteration type deformation field U by using a partial differential equation.
Further, epsilon (U) is calculated by using a variational method to obtain an Euler-Lagrange equation taking a deformation field U as a variable, and the Euler-Lagrange equation is rewritten, and the method comprises the following steps:
and (3) calculating epsilon (U) by using a variational method to obtain an Euler-Lagrange equation taking the deformation field U as a variable:
wherein the method comprises the steps ofIs the Laplacian, i.e.)>Since the reference image a (X) is not a function of the deformation field U, there are:
so that the formula (3) can be rewritten as:
wherein formula (4) uses Dirichlet boundary conditions
Further, an iterative method for solving the deformation field U using partial differential equations includes:
solving (4) by finite difference method to obtain
Solving by Newton's method, the following can be obtained:
wherein,
because there isSubstituting into (6) to obtain:
discarding the higher order partial derivatives can result in:
substituted into (5) there are
Further, the iterative method for solving the deformation field U by using the partial differential equation further comprises:
optimizing the iteration type of the deformation field, and obtaining the optimized iteration type as follows:
wherein the method comprises the steps of
Further, the step of optimizing includes:
when the gray scale of the floating image is flat, there isAt the same time lambda 1 ,σ i The iteration (7) is unstable, so the component R is added in the denominator 2 (X, U), and the rewritable formula (7) is
When the two images are gradually aligned, the gradients of the two images are gradually close, i.eThus formula (8) can be written as:
wherein,
further, lambda 1 Is related to the smoothing and registration accuracy of the deformation field, lambda 1 The value of (2) is between 0.1 and 0.3.
Correspondingly, the invention also provides a medical image elastic registration device, which comprises:
a construction module for energy function epsilon of Demons algorithm 1 On the basis of (U), increasing the smoothness constraint term epsilon of the deformation field 2 (U) and increasing the symmetric Kullback-Leibler distance ε 3 (U) constructing a total energy functional epsilon (U); wherein the total energy functional is a multi-dimensional parameter functional;
and the acquisition module is used for obtaining a group of parameters which can minimize the total energy functional through a variation method to obtain an optimal deformation field.
Compared with the prior art, the invention has the beneficial effects that: based on the energy functional, the partial derivative square sum of the deformation field is added and used as a smooth constraint term for smoothing the deformation field; the symmetrical Kullback-Leibler distance is increased to maintain an image topological structure, non-rigid registration is converted into an energy functional minimization problem, and a set of parameters capable of enabling the energy functional to be minimized is searched through a variational method to obtain an optimal deformation field.
The foregoing description is only an overview of the present invention, and is intended to provide a better understanding of the present invention, as it is embodied in the following description, with reference to the preferred embodiments of the present invention and the accompanying drawings. Specific embodiments of the present invention are given in detail by the following examples and the accompanying drawings.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiments of the invention and together with the description serve to explain the invention and do not constitute a limitation on the invention. In the drawings:
FIG. 1 is a flow chart of a medical image elastic registration method of the present invention;
FIG. 2 is another flow chart of a medical image elastic registration method of the present invention;
fig. 3 is a block schematic diagram of a medical image elastic registration device of the present invention.
Reference numerals: 100. constructing a module; 200. and an acquisition module.
Detailed Description
The foregoing and other objects, features, aspects and advantages of the present invention will become more apparent to those skilled in the art from the following detailed description, which, taken in conjunction with the annexed drawings, discloses a device for practicing the invention. In the drawings, the shape and size may be exaggerated for clarity, and the same reference numerals will be used throughout the drawings to designate the same or similar components. In the following description, terms such as center, thickness, height, length, front, back, rear, left, right, top, bottom, upper, lower, etc. are based on the orientation or positional relationship shown in the drawings. In particular, "height" corresponds to the top-to-bottom dimension, "width" corresponds to the left-to-right dimension, and "depth" corresponds to the front-to-back dimension. These relative terms are for convenience of description and are not generally intended to require a particular orientation. Terms (e.g., "connected" and "attached") referring to an attachment, coupling, etc., refer to a relationship wherein these structures are directly or indirectly secured or attached to one another through intervening structures, as well as both movable or rigid attachments or relationships, unless expressly described otherwise.
The present invention will be further described with reference to the accompanying drawings and detailed description, wherein it is to be understood that, on the premise of no conflict, new embodiments may be formed by any combination of the embodiments or technical features described below. It will be understood that terms, such as "having," "including," and "comprising," as used herein, do not preclude the presence or addition of one or more other elements or groups thereof.
Image registration technology is taken as the basis of an image processing process, and research of an image registration algorithm is more important in image registration research.
As shown in fig. 1-3, a medical image elastic registration method of the present invention includes:
s1, energy functional epsilon in Demons algorithm 1 On the basis of (U), increasing the smoothness constraint term epsilon of the deformation field 2 (U) and increasing the symmetric Kullback-Leibler distance ε 3 (U) constructing a total energy functional epsilon (U); wherein the total energy functional is a multi-dimensional parameter functional;
s2, obtaining a group of parameters capable of minimizing the total energy functional through a variation method to obtain an optimal deformation field.
In particular, the essence of the non-rigid registration of three-dimensional medical images is to construct a three-dimensional deformation field U such that
U:B(X,U)→A(X)
Wherein u= { U i,j,k The deformation field, x= { X i,j,k And (3) is a voxel set, i, j, k are indexes along (X, y, z) coordinate axes, A (X) is a reference image, and B (X, U) is a floating image. To construct a suitable deformation field, a deformation vector for each voxel in the deformation field needs to be calculated. As the registration process proceeds, the voxel coordinates also move.
X'(X,t)=X+U(X,t)
The optimal deformation field U for non-rigid registration is to minimize the deformation energy of the deformation field, i.e
Energy functional epsilon of traditional Demons algorithm 1 (U) is:
wherein sigma i Is the noise level, sigma, of the image x Is the iteration step length; since the conventional Demons algorithm cannot generate smoother deformation field when large deformation occurs, the smoothness constraint term of the deformation field is added on the basis of the original energy functional.
Smoothness constraint term epsilon of deformation field 2 (U) is:
because the energy function of the traditional Demons algorithm cannot maintain the topological structure of a large deformation image, a symmetrical Kullback-Leibler distance sKL (-) is added to the energy function.
Symmetric Kullback-Leibler distance ε 3 (U) is:
the total energy function is obtained as:
wherein 0 is<λ 1 ,λ 2 <1, a step of; a (X) is a reference image, B (X, U) is a floating image, R (X, U) =b (X, U) -a (X), representing the difference between the reference image and the floating image; omega shape 3 Representing the deformation region.
ε 2 (U) is mainly used for smoothing deformation field, when epsilon 1 (U) is much greater than lambda 1 ε 2 (U), i.e. the reference image and the floating image differ greatly, ε (U) depends mainly on ε 1 (U) this can result inThe larger deformation field causes the floating image to generate larger deformation; epsilon after the images are gradually registered 1 When (U) becomes smaller, ε (U) is mainly dependent on λ 1 ε 2 (U) minimizing the energy functional, i.e. constraining the amplitude of variation of the deformation field in the (x, y, z) direction, by means of a set of parameters such that the functional produces a smooth deformation field; and when the integral is at the integral boundaryOr epsilon (U) can smooth U when partial data is incomplete. Epsilon 3 (U) use the symmetric Kullback-Leibler distance as an auxiliary regularization term for the energy functional.
Preferably, DU is Jacobian determinant |Du i,j,k Matrix of | du| is defined as:
wherein,
the minimization formula is obtained:
minimisation formula (2) approximates the Jacobian determinant distribution of forward and reverse transforms to a constant map.
Epsilon (U) needs to be solved using a variational method, since epsilon (U) involves minimizing the nonlinear functional (1) and epsilon (U) is a functional of a multidimensional parameter.
Preferably, S2 obtains an optimal deformation field by a set of parameters that minimize the total energy functional by a variational method, including:
s21, calculating epsilon (U) by using a variational method to obtain an Euler-Lagrange equation taking a deformation field U as a variable, and rewriting;
s22, solving the iteration type deformation field U by using a partial differential equation.
Specifically, S21 calculates epsilon (U) using a variational method to obtain an euler-lagrangian equation with a deformation field U as a variable, and rewrites the equation, including:
and (3) calculating epsilon (U) by using a variational method to obtain an Euler-Lagrange equation taking the deformation field U as a variable:
wherein,is the Laplacian, i.e.)>Since the reference image a (X) is not a function of the deformation field U, there are:
therefore, the formula (3) can be rewritten as:
wherein formula (4) uses Dirichlet boundary conditions
Non-rigid registration is often decomposed with pyramids in multiple scales, and registration of images of different scales can use different lambda' s 1 . Generally, for coarse registration of images, λ is used to generate a smoother deformation field 1 The value can be slightly larger, so that the deformation field can be well smoothed; conversely, lambda when registering image details 1 The value should be slightly smaller, which is beneficial to improving the registration accuracy. Lambda (lambda) 1 Generally takes a value of between 0.1 and 0.3, if lambda 1 Too large, the registration accuracy may be degraded. To ensure the matching ofAccurate, set lambda 1 =0.1. When lambda is 1 The value is smaller, the deformation field is not smooth enough, and the topology maintenance characteristic of the image is correspondingly deteriorated; if lambda is 2 Too large a value can also affect registration accuracy. Therefore, in order to improve the topology maintenance characteristic of the image without decreasing the registration accuracy, the value λ is preferably selected 2 =0.5。
Specifically, S22 solves the iteration of the deformation field U using partial differential equations, including:
solving (4) by finite difference method to obtain
Solving by Newton's method, the following can be obtained:
wherein,
because there isSubstituting into (6) to obtain:
discarding the higher order partial derivatives can result in:
substituted into (5) there are
Preferably, S22 solves the iteration of the deformation field U using partial differential equations, and then further comprises:
s23, optimizing the iteration type of the deformation field, and obtaining the optimized iteration type as follows:
wherein the method comprises the steps of
Specifically, the step of optimizing in S23 includes:
when the gray scale of the floating image is flat, there isIf at the same time lambda 1 ,σ i Is also small, making the iteration (7) unstable, so the component R is added in the denominator 2 (X, U), and the rewritable formula (7) is
When the two images are gradually aligned, the gradients of the two images are gradually close, i.eThus formula (8) can be written as:
wherein,
the invention realizes accurate registration by continuously optimizing the deformation field, lambda 1 The value of (2) is related to the smoothing of the deformation field and the registration accuracy. Tool withIn the case of coarse registration of images, λ is used to generate a smoother deformation field 1 The value can be slightly larger, so that the deformation field can be well smoothed; conversely, lambda when registering image details 1 The value should be slightly smaller, which is beneficial to improving the registration accuracy. I.e. lambda 1 The value is positively correlated with the smoothness of the deformation field, and lambda 1 The value is inversely related to the registration accuracy.
The topology preserving variation algorithm provided by the invention is added with the partial derivative of the deformation field and the symmetric Kullback-Leibler distance on the basis of the energy functional of the traditional Demons algorithm, and is respectively used for representing the smooth constraint item and the topology preserving constraint item of the deformation field. The non-rigid registration problem of the medical image is solved by using a variational principle, and a physical constraint model or priori knowledge is avoided. As the smoothness constraint item is added in the energy functional, the deformation field generated by the topology preserving variation algorithm has a regularization effect similar to that of the Demons algorithm, and the generated deformation field is smoother; the symmetric Kullback-Leibler distance is added in the energy functional, so that the Jacobian matrix forward transformation and reverse transformation of the deformation field are close to constant mapping, and the topological structure of the image is maintained.
In the invention, a differential synblast Demons algorithm is adopted as a comparison algorithm in a result verification stage, and experimental results show that the algorithm registration accuracy of the invention is similar to that of the differential synblast Demons algorithm no matter a single-mode image or a multi-mode image, but the deformation field after registration of the invention is obviously smoother than that of the differential synblast Demons algorithm, and the topology retention characteristic is better than that of the differential synblast Demons algorithm. Because the topology preserving variation algorithm provided by the invention adopts vector addition approximation of spatial transformation in iteration, when the image deformation is very large, the registration accuracy of the topology preserving variation algorithm is lower than that of the differential homoembryo Demons algorithm.
Correspondingly, the invention also provides a medical image elastic registration device, which comprises:
a building block 100 for energy function epsilon in the Demons algorithm 1 On the basis of (U), increasing the smoothness constraint term epsilon of the deformation field 2 (U) and increasing the symmetric Kullback-Leibler distance ε 3 (U) construction of the Total energy functionalEpsilon (U); wherein the total energy functional is a multi-dimensional parameter functional;
the obtaining module 200 is configured to obtain, by a variance method, a set of parameters that can minimize the total energy functional to obtain an optimal deformation field.
Based on the energy functional, the partial derivative square sum of the deformation field is added and used as a smooth constraint term for smoothing the deformation field; the symmetrical Kullback-Leibler distance is increased to maintain an image topological structure, non-rigid registration is converted into an energy functional minimization problem, and a set of parameters capable of enabling the energy functional to be minimized is searched through a variational method to obtain an optimal deformation field.
Furthermore, the device and method embodiments in the device embodiments are based on the same inventive concept.
The embodiment of the invention also provides a computer storage medium, which comprises a memory and a processor, wherein at least one instruction and at least one section of program are stored in the memory, and the at least one instruction and the at least one section of program are loaded and executed by the processor to realize the medical image elastic registration method provided by the embodiment of the method.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. And the foregoing description has been directed to specific embodiments of this specification. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for the apparatus and electronic device embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and references to the parts of the description of the method embodiments are only required.
The foregoing description has fully disclosed specific embodiments of this invention. It should be noted that any modifications to the specific embodiments of the invention may be made by those skilled in the art without departing from the scope of the invention as defined in the appended claims. Accordingly, the scope of the claims of the present invention is not limited to the foregoing detailed description.
Claims (6)
1. A method of elastic registration of medical images, comprising:
energy functional ε in Demons algorithm 1 On the basis of (U), increasing the smoothness constraint term epsilon of the deformation field 2 (U) and increasing the symmetric Kullback-Leibler distance ε 3 (U) constructing a total energy functional epsilon (U); wherein the total energy functional is a multi-dimensional parameter functional;
obtaining a group of parameters which can minimize the total energy functional through a variational method to obtain an optimal deformation field;
the energy functional ε 1 (U) of the Demons algorithm is:
wherein sigma i Is the noise level, sigma, of the image x Is the iteration step length;
the smoothness constraint term ε 2 (U) of the deformation field is:
the symmetric Kullback-Leibler distance ε 3 (U) is:
the total energy functional is obtained as follows:
wherein 0 is<λ1,λ2<1;U={u i,j,k The deformation field, x= { X i,j,k -voxel set, i, j, k are indices along (X, y, z) coordinate axes, a (X) is a reference image, B (X, U) is a floating image, R (X, U) =b (X, U) -a (X), representing the difference between the reference image and the floating image; omega shape 3 Representing a deformation region;
DU is Jacobian determinant |Du i,j,k Matrix of | du| is defined as:
wherein,
obtaining a set of parameters capable of minimizing the total energy functional to obtain an optimal deformation field through a variation method, wherein the method comprises the following steps:
using a variational method to calculate epsilon (U) to obtain an Euler-Lagrange equation taking a deformation field U as a variable, and rewriting;
solving an iteration type deformation field U by using a partial differential equation;
using a variational method to calculate epsilon (U) to obtain an Euler-Lagrange equation taking a deformation field U as a variable, and rewriting the Euler-Lagrange equation, wherein the method comprises the following steps:
and (3) calculating epsilon (U) by using a variational method to obtain an Euler-Lagrange equation taking the deformation field U as a variable:
wherein the method comprises the steps ofIs the Laplacian, i.e.)>Since the reference image a (X) is not a function of the deformation field U, there are:
so that the formula (3) can be rewritten as:
wherein formula (4) uses Dirichlet boundary conditions
An iterative method for solving a deformation field U using partial differential equations, comprising:
solving (4) by finite difference method to obtain
Solving by Newton's method, the following can be obtained:
wherein,
because there isSubstituting into (6) to obtain:
discarding the higher order partial derivatives can result in:
substituted into (5) there are
2. A medical image elastic registration method according to claim 1, wherein,
the minimization formula is obtained:
minimisation formula (2) approximates the Jacobian determinant distribution of forward and reverse transforms to a constant map.
3. The medical image elastic registration method of claim 1, wherein the iterative solving of the deformation field U using partial differential equations is followed by:
optimizing the iteration type of the deformation field, and obtaining the optimized iteration type as follows:
wherein the method comprises the steps of
4. A medical image elastic registration method according to claim 3, wherein the step of optimizing comprises:
when the gray scale of the floating image is flat, there isAt the same time lambda 1 ,σ i The iteration (7) is unstable, so the component R is added in the denominator 2 (X, U), and the rewritable formula (7) is
When the two images are gradually aligned, the gradients of the two images are gradually close, i.eThus formula (8) can be written as:
wherein,
5. a medical image elastic registration method according to any of claims 2-4, characterized in that λ 1 Is related to the smoothing and registration accuracy of the deformation field, lambda 1 The value of (2) is between 0.1 and 0.3.
6. A medical image elastic registration device implementing the method of claim 1, comprising:
a construction module for energy function epsilon of Demons algorithm 1 On the basis of (U), increasing the smoothness constraint term epsilon of the deformation field 2 (U) and addition of symmetrical Kullback-LeiblDistance epsilon of er 3 (U) constructing a total energy functional epsilon (U); wherein the total energy functional is a multi-dimensional parameter functional;
and the acquisition module is used for obtaining a group of parameters which can minimize the total energy functional through a variation method to obtain an optimal deformation field.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110600433.0A CN113223061B (en) | 2021-05-31 | 2021-05-31 | Elastic registration method and device for medical image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110600433.0A CN113223061B (en) | 2021-05-31 | 2021-05-31 | Elastic registration method and device for medical image |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113223061A CN113223061A (en) | 2021-08-06 |
CN113223061B true CN113223061B (en) | 2024-04-05 |
Family
ID=77082019
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110600433.0A Active CN113223061B (en) | 2021-05-31 | 2021-05-31 | Elastic registration method and device for medical image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113223061B (en) |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107194956A (en) * | 2017-05-16 | 2017-09-22 | 山东大学 | A kind of Non-rigid Medical Image Registration method |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7813592B2 (en) * | 2006-08-09 | 2010-10-12 | Siemens Medical Solutions Usa, Inc. | System and method for non-rigid multi-modal registration on the GPU |
US8265356B2 (en) * | 2008-01-30 | 2012-09-11 | Computerized Medical Systems, Inc. | Method and apparatus for efficient automated re-contouring of four-dimensional medical imagery using surface displacement fields |
EP2369551B1 (en) * | 2010-03-25 | 2019-10-30 | Emory University | Imaging system and method |
US9218643B2 (en) * | 2011-05-12 | 2015-12-22 | The Johns Hopkins University | Method and system for registering images |
-
2021
- 2021-05-31 CN CN202110600433.0A patent/CN113223061B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107194956A (en) * | 2017-05-16 | 2017-09-22 | 山东大学 | A kind of Non-rigid Medical Image Registration method |
Non-Patent Citations (2)
Title |
---|
"具有大形变特征的颅脑CT图像的非刚性配准";徐峰 等;《中国生物医学工程学报》;第29卷(第2期);第172-177页 * |
"肺部非刚性医学图像配准算法的研究";南玲巧;《中国优秀硕士学位论文全文数据库信息科技辑》(第03期);第21-22页 * |
Also Published As
Publication number | Publication date |
---|---|
CN113223061A (en) | 2021-08-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Cao et al. | Large deformation diffeomorphic metric mapping of vector fields | |
US8265356B2 (en) | Method and apparatus for efficient automated re-contouring of four-dimensional medical imagery using surface displacement fields | |
Song et al. | Lung CT image registration using diffeomorphic transformation models | |
Liang et al. | Nonrigid iterative closest points for registration of 3D biomedical surfaces | |
Mair et al. | Estimation of images and nonrigid deformations in gated emission CT | |
US9508140B2 (en) | Quantifying curvature of biological structures from imaging data | |
Walimbe et al. | Automatic elastic image registration by interpolation of 3D rotations and translations from discrete rigid-body transformations | |
Lau et al. | Non-rigid image registration using a median-filtered coarse-to-fine displacement field and a symmetric correlation ratio | |
Ségonne | Active contours under topology control—genus preserving level sets | |
CN112798654A (en) | Rapid gradient method and self-adaptive Jacobian matrix reconstruction method for electrical impedance tomography | |
Sun et al. | Simultaneous multiresolution strategies for nonrigid image registration | |
Wickramasinghe et al. | Deep active surface models | |
CN108428245B (en) | Slip image registration method based on self-adaptive regular term | |
CN108717705A (en) | Differomorphism method for registering images based on static vector field | |
Che et al. | AMNet: Adaptive multi-level network for deformable registration of 3D brain MR images | |
CN113223061B (en) | Elastic registration method and device for medical image | |
Heinrich et al. | Discontinuity preserving regularisation for variational optical-flow registration using the modified Lp norm | |
Azencott et al. | Diffeomorphic matching and dynamic deformable surfaces in 3D medical imaging | |
Salomon et al. | A massively parallel approach to deformable matching of 3D medical images via stochastic differential equations | |
Ashburner et al. | High-dimensional image warping | |
Wang et al. | Multi-affine registration using local polynomial expansion | |
Zhu et al. | Modified fast marching and level set method for medical image segmentation | |
CN113256693A (en) | Multi-view registration method based on K-means and normal distribution transformation | |
Droske | On variational problems and gradient flows in image processing | |
CN113724320A (en) | Shape prior-based kidney image segmentation method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |