+

CN111445479A - Method and device for segmenting interest region in medical image - Google Patents

Method and device for segmenting interest region in medical image Download PDF

Info

Publication number
CN111445479A
CN111445479A CN202010202834.6A CN202010202834A CN111445479A CN 111445479 A CN111445479 A CN 111445479A CN 202010202834 A CN202010202834 A CN 202010202834A CN 111445479 A CN111445479 A CN 111445479A
Authority
CN
China
Prior art keywords
interest
region
mesh
grid
vertex
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
Application number
CN202010202834.6A
Other languages
Chinese (zh)
Other versions
CN111445479B (en
Inventor
张丛嵘
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.)
Neusoft Medical Systems Co Ltd
Original Assignee
Neusoft Medical Systems Co Ltd
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 Neusoft Medical Systems Co Ltd filed Critical Neusoft Medical Systems Co Ltd
Priority to CN202010202834.6A priority Critical patent/CN111445479B/en
Publication of CN111445479A publication Critical patent/CN111445479A/en
Application granted granted Critical
Publication of CN111445479B publication Critical patent/CN111445479B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
    • G06T17/20—Finite element generation, e.g. wire-frame surface description, tesselation
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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 OR CALCULATING; 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
    • G06T2207/30056—Liver; Hepatic
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G06T2207/30096—Tumor; Lesion
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2210/00—Indexing scheme for image generation or computer graphics
    • G06T2210/41—Medical

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computer Graphics (AREA)
  • Geometry (AREA)
  • Software Systems (AREA)
  • Apparatus For Radiation Diagnosis (AREA)

Abstract

The present specification provides a method and an apparatus for segmenting a region of interest in a medical image, an electronic device, and a storage medium, wherein the method includes: establishing a mesh model aiming at an interested area of the medical image, wherein the mesh model takes a space coordinate in the interested area as a mesh vertex; under the condition that the grid vertex is not positioned on the edge of the region of interest, calculating the similarity between the pixel characteristic where the grid vertex coordinate is positioned and the pixel characteristic of the space coordinate away from the grid vertex by the preset moving vector, moving the grid vertex according to the actual moving vector, wherein the distance of the actual moving vector is positively correlated with the similarity, and the direction of the actual moving vector is the same as the direction of the preset moving vector; and under the condition that the grid vertex is positioned on the edge of the region of interest, segmenting the region of interest in the medical image according to the grid model after the grid vertex is moved. Therefore, on the basis of ensuring the accuracy, the calculation amount is greatly reduced, and the efficiency is improved.

Description

Method and device for segmenting interest region in medical image
Technical Field
The present disclosure relates to the field of medical technology, and in particular, to a method and an apparatus for segmenting a region of interest in a medical image, an electronic device, and a storage medium.
Background
Before clinical departments, preoperative planning of medical images is required to segment the areas of interest to medical personnel in medical images. Clinically, many medical personnel still manually delineate a region of interest (ROI) to give an example of performing radiotherapy on a liver tumor, and a radiotherapy doctor needs to manually delineate the liver tumor layer by layer in a CT image. The enormous amount of data makes this a very time and labor intensive task, while inter-observer variability is introduced by the human as a criterion for delineation. It is therefore highly desirable to provide a fast and efficient region of interest delineation method.
Disclosure of Invention
In order to overcome the problems in the related art, the present specification provides a method and an apparatus for segmenting a region of interest in a medical image, an electronic device, and a storage medium.
According to a first aspect of embodiments herein, there is provided a segmentation method of a region of interest in a medical image, the segmentation method comprising:
establishing a mesh model for a region of interest of a medical image, the mesh model taking spatial coordinates in the region of interest as mesh vertices;
under the condition that the grid vertex is not located on the edge of the region of interest, calculating the similarity between the pixel feature where the grid vertex coordinate is located and the pixel feature of the space coordinate away from the grid vertex by the preset moving vector, and moving the grid vertex according to an actual moving vector, wherein the distance of the actual moving vector is positively correlated with the similarity, and the direction of the actual moving vector is the same as the direction of the preset moving vector;
and under the condition that the grid vertex is positioned on the edge of the region of interest, segmenting the region of interest in the medical image according to the grid model after the grid vertex is moved.
Optionally, the segmentation method further includes:
determining a step size proportion of the grid vertex according to the similarity, wherein the step size proportion is positively correlated with the similarity;
and determining the actual motion vector according to the step length proportion and the preset motion vector.
Optionally, before moving the mesh vertex according to the actual motion vector, the method further includes:
based on the step length proportion of the adjacent grid vertex of the grid vertex, smoothing the step length proportion of the grid vertex;
determining the actual motion vector according to the step length proportion and the preset motion vector, wherein the step length proportion comprises the following steps:
and determining the actual motion vector according to the step length proportion subjected to the smoothing processing and the preset motion vector.
Optionally, after moving the mesh vertex according to the actual motion vector, the method further includes:
judging whether the distance between two adjacent grid vertexes in the grid model moving through the grid vertexes is larger than a distance threshold value or not;
if the distance between two adjacent grid vertexes is not greater than the distance threshold, selecting at least one space coordinate between connecting lines of the two adjacent grid vertexes as a newly added grid vertex of the grid model, and enabling the distance between the two adjacent grid vertexes in the grid model to be not greater than the distance threshold.
Optionally, the segmentation method further includes:
determining that the mesh vertex is located on an edge of the region of interest if the similarity is less than a similarity threshold;
and/or determining that the grid vertex is located on the edge of the region of interest under the condition that the change rate of the similarity is smaller than a change rate threshold, wherein the change rate of the similarity is the change rate of the similarity of the grid vertex in the current iteration movement and the similarity of the grid vertex in the previous iteration movement;
and/or determining that the mesh vertex is located on the edge of the region of interest in case of receiving a stop instruction of the mesh vertex.
Optionally, after segmenting the region of interest in the medical image according to the mesh model after the mesh vertices are moved, the method further includes:
and performing image reconstruction processing on the image data corresponding to the segmented region of interest and displaying the image data.
Optionally, the pixel features comprise at least one of the following features: gray scale, texture, adjacency relation, overlap relation;
and/or, if the medical image is a three-dimensional image, the grid model is a three-dimensional model; if the medical image is a two-dimensional image, the grid model is a two-dimensional curve model;
and/or the direction of the actual motion vector is the direction in which the central point of the mesh model points to the mesh vertex.
According to a second aspect of embodiments herein, there is provided a segmentation apparatus for a region of interest in a medical image, the segmentation apparatus comprising: the device comprises an establishing module, a calculating module and a dividing module;
the building module is configured to build a mesh model for a region of interest of a medical image, the mesh model having spatial coordinates in the region of interest as mesh vertices;
under the condition that the grid vertex is not located on the edge of the region of interest, the computing module is configured to compute the similarity between the pixel feature where the grid vertex coordinate is located and a space coordinate away from the grid vertex by a preset motion vector, and move the grid vertex according to an actual motion vector, wherein the distance of the actual motion vector is positively correlated with the similarity, and the direction of the actual motion vector is the same as the direction of the preset motion vector;
in the case where the mesh vertices are located on the edge of the region of interest, the segmentation module is configured to:
and segmenting the region of interest in the medical image according to the mesh model after the mesh vertex movement.
Optionally, the computing module is further configured to:
determining a step size proportion of the grid vertex according to the similarity, wherein the step size proportion is positively correlated with the similarity;
and determining the actual motion vector according to the step length proportion and the preset motion vector.
Optionally, the segmenting device further comprises: a smoothing module;
the smoothing module is configured to:
and based on the step length proportion of the adjacent mesh vertex of the mesh vertex, smoothing the step length proportion of the mesh vertex.
Optionally, the segmenting device further comprises: a module is newly added;
the add-on module is configured to:
judging whether the distance between two adjacent grid vertexes in the grid model after the grid vertexes are moved is larger than a distance threshold value or not;
if the distance between two adjacent grid vertexes is not greater than the distance threshold, selecting at least one space coordinate between connecting lines of the two adjacent grid vertexes as a newly added grid vertex of the grid model, and enabling the distance between the two adjacent grid vertexes in the grid model to be not greater than the distance threshold.
Optionally, the segmenting device further comprises: a judgment module;
the determination module is configured to:
determining that the mesh vertex is located on an edge of the region of interest if the similarity is less than a similarity threshold;
and/or determining that the grid vertex is located on the edge of the region of interest under the condition that the change rate of the similarity is smaller than a change rate threshold, wherein the change rate of the similarity is the change rate of the similarity of the grid vertex in the current iteration movement and the similarity of the grid vertex in the previous iteration movement;
and/or, in the event that a pause instruction is received, determining that the mesh vertices are located on the edge of the region of interest.
According to a third aspect of embodiments of the present specification, there is provided an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the method for segmenting a region of interest in a medical image according to any one of the above items when executing the computer program.
According to a third aspect of embodiments herein, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method for segmenting a region of interest in a medical image as set forth in any of the above.
The technical scheme provided by the embodiment of the specification can have the following beneficial effects:
in the embodiment of the specification, with the help of the mesh model with the common characteristic, the region of interest is segmented by moving the mesh vertexes of the mesh model for multiple times until almost all the mesh vertexes of the model are located at the edge of the region of interest, and for each movement of the mesh vertexes, the mesh vertexes are not moved by fixed preset moving vectors, but the moving vectors are dynamically adjusted according to the size of the similarity, the actual moving step length of the mesh vertexes is determined, and the positions of the mesh vertexes are updated, so that the reduction of accuracy caused by the oversize preset moving vector setting and the oversize calculated amount caused by the undersize preset moving vector setting are avoided, the calculated amount is greatly reduced on the basis of ensuring the accuracy, and the efficiency is improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the specification.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present specification and together with the description, serve to explain the principles of the specification.
Fig. 1 is a flow chart illustrating a method for segmenting a region of interest in a medical image according to an exemplary embodiment.
Fig. 2a is a schematic illustration of an axial plane direction of a three-dimensional medical image shown in the present specification according to an exemplary embodiment.
FIG. 2b is a schematic view in the sagittal plane direction of a three-dimensional medical image shown in accordance with an exemplary embodiment of the present description.
FIG. 2c is a schematic diagram of a structure of a mesh model shown in the present specification according to an exemplary embodiment.
FIG. 2d is a schematic diagram illustrating a neighborhood of pixel points in accordance with an exemplary embodiment of the present description.
FIG. 3 is a flow chart illustrating another method of segmentation of a region of interest in a medical image according to an exemplary embodiment of the present description.
FIG. 4a is a first diagram illustrating a portion of mesh vertices of a mesh model in accordance with an exemplary embodiment.
FIG. 4b is a second diagram illustrating a portion of mesh vertices of a mesh model in accordance with an exemplary embodiment.
Fig. 5 is a block diagram of a segmentation apparatus for a region of interest in a medical image according to an exemplary embodiment.
Fig. 6 is a schematic structural diagram of an electronic device shown in the present specification according to an exemplary embodiment.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present specification. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the specification, as detailed in the appended claims.
The terminology used in the description herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the description. As used in this specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.
It should be understood that although the terms first, second, third, etc. may be used herein to describe various information, these information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, the first information may also be referred to as second information, and similarly, the second information may also be referred to as first information, without departing from the scope of the present specification. The word "if" as used herein may be interpreted as "at … …" or "when … …" or "in response to a determination", depending on the context.
Nowadays, tumor diseases are more and more advanced, and for the treatment of tumors, no matter a radiotherapy method or an operative resection method is adopted, preoperative planning is required, that is, in a medical image containing tumors, a region where the tumors are located is identified or segmented, and then image reconstruction processing is performed on the region. Clinically, many medical personnel still manually delineate the region of the tumor, for example, to perform radiotherapy on the liver tumor, and a radiotherapy doctor needs to manually delineate the liver tumor layer by layer in a CT image. The huge amount of data makes this a very time and labor intensive task. It is therefore highly desirable to provide a fast and efficient method for delineating a tumor region.
Liver lesion segmentation has been a challenging task due to the variability of lesion shape, size and density, for example, liver lesions, which have low contrast with surrounding liver tissue, weak boundaries, and non-uniform gray levels. The current object segmentation methods, such as region growing, threshold, clustering, level set, graph segmentation, machine learning, etc., have the problem that the accuracy is low, so that the segmentation result cannot reach the actual requirement; or the problem of large calculation amount exists; machine learning needs a large amount of sample data support, richer sample data have better effect, and otherwise, expectation is difficult to achieve.
Based on the above situation, in order to solve the problems of large calculation amount, low accuracy and the like when segmenting a region of interest such as a tumor region, an embodiment of the present invention provides a method for segmenting a region of interest in a medical image, in which, by means of a mesh model having a common feature, mesh vertices of the mesh model are moved for multiple times until almost all mesh vertices of the model are located at the edge of the region of interest, and the method can rapidly and accurately segment the region of interest in the medical image with a small calculation amount.
The following provides a detailed description of examples of the present specification.
As shown in fig. 1, fig. 1 is a flowchart illustrating a method for segmenting a region of interest in a medical image according to an exemplary embodiment, including the following steps:
step 101, establishing a mesh model for a region of interest of a medical image.
Wherein, the medical image can be a three-dimensional CT image acquired by scanning a scanned object by adopting a CT (computed tomography) device; the medical image can also be a three-dimensional MR image acquired by scanning a scanned object by an MR (magnetic resonance) device; the medical image can also be a three-dimensional PET image acquired by scanning a scanned object by using a PET (positron emission tomography) device; the medical image may also be a two-dimensional slice image obtained by resampling the three-dimensional image.
The region of interest (ROI) is a region of interest of a medical staff in a medical image, which is a region to be identified or segmented in the medical image, and may be a certain tissue or organ, or a lesion region in the medical image.
It can be understood that when the medical image is a three-dimensional image, a three-dimensional mesh model is established; and when the medical image is a two-dimensional image, establishing a two-dimensional curve model. The mesh model can be built using, but is not limited to, a marching cube. When the model is built, a user can automatically designate a coordinate P in the region of interest0(x0,y0,z0) And the maximum reference radius Rmax or the model area of the model is set as the central point of the mesh model so as to divide the size and the shape of the module. See fig. 2a and 2b for the purpose of creating a three-dimensional medical image with P0(x0,y0,z0) For example, a spherical mesh model with a center point and a radius of Rmax, FIG. 2a is a schematic view of a three-dimensional medical image in an axial plane direction, FIG. 2b is a schematic view of a three-dimensional medical image in a sagittal plane direction, and a cross line intersection point in the schematic view is located at a center point P of the mesh model0(x0,y0,z0) The radius of the sphere is the maximum reference radius Rmax.
The mesh model can be a triangular mesh model which is composed of triangular patches, and the vertexes of the triangular patches (triangles) are the mesh vertexes of the mesh model; the mesh model may also be composed of a plurality of other polygonal patches.
The mesh model established in step 101 is not limited to the spherical shape shown in fig. 2a and 2b, and may be a cube, a rectangular parallelepiped, or the like. The maximum reference radius Rmax (size) of the mesh model can be set according to actual requirements, the Rmax is set to be more appropriate, and mesh vertexes of the mesh model are closer to the edge S of the region of interest, so that the moving times of the mesh model can be reduced, the calculation amount is reduced, and the algorithm efficiency is improved. The center point of the mesh model is selected as the center position of the region of interest by the user, which makes the segmentation result more approximate to the real situation.
After the mesh model is built, the following steps are required to be performed to move the mesh vertices of the mesh model along the edge position of the region of interest until most or all of the mesh vertices of the mesh model are located at the edge position of the region of interest.
Since the movement of the mesh vertices can be performed independently, a specific movement process of the mesh vertices will be described below by taking one mesh vertex as an example.
And 102, calculating the similarity between the space coordinate where the grid vertex is located and the pixel characteristic of the space coordinate away from the grid vertex by a preset moving vector.
The preset movement vector is used for determining the direction and the distance of the planned movement of the grid vertex in the movement. The distance and the vector direction of the preset moving vector can be set according to actual requirements.
If a mesh vertex in the mesh model is moved, the spatial coordinates that are away from the mesh vertex by a preset movement vector are expressed as follows:
Figure BDA0002419955100000081
wherein,
Figure BDA0002419955100000091
representing the spatial coordinate position of the k-th mesh vertex of the mesh model after i times of movement;
Figure BDA0002419955100000092
representation and mesh vertices
Figure BDA0002419955100000093
A spatial coordinate position apart from a preset motion vector;
Figure BDA0002419955100000094
representing a preset motion vector.
In step 102, the mesh vertices are computed
Figure BDA0002419955100000095
And the space coordinate
Figure BDA0002419955100000096
The similarity of the pixel characteristics of (2).
The calculation of the similarity is explained below using a mesh model shown in fig. 2c as an example:
in FIG. 2c, each dot represents a spatial coordinate, and the current network model has a spatial coordinate P0As a central point, by spatial coordinates
Figure BDA0002419955100000097
Are the mesh vertices. If the motion vector is preset to
Figure BDA0002419955100000098
Is the vector direction, L is the distance, and
Figure BDA0002419955100000099
space seat with preset movement vector distanceIs marked as
Figure BDA00024199551000000910
Need to calculate
Figure BDA00024199551000000911
Pixel characteristics of and
Figure BDA00024199551000000912
the similarity of the pixel characteristics of (2).
Wherein the pixel characteristics include at least one of the following: gray scale, texture, adjacency, overlap.
Taking the gray scale as the pixel feature to calculate the similarity as an example, the calculation formula of the similarity may be expressed as follows, but is not limited to:
Figure BDA00024199551000000913
wherein Diff represents similarity; SeedMeanValue represents the spatial coordinates
Figure BDA00024199551000000914
The gray scale of (2). Spatial coordinates
Figure BDA00024199551000000915
May be the spatial coordinates
Figure BDA00024199551000000916
The gray scale of itself can also be the space coordinate
Figure BDA00024199551000000917
The gray level mean value of the neighborhood pixels.
The neighborhood pixels of the spatial coordinate are all pixels within a preset range by taking the spatial coordinate as a center. Referring to FIG. 2d, if the spatial coordinates are selected
Figure BDA00024199551000000918
3 x 3 neighborhood (predetermined range), i.e. in spatial coordinates
Figure BDA00024199551000000919
As a center, each edge parallel to the xyz axis has a region of 3 pixels, except for the spatial coordinates
Figure BDA00024199551000000920
The other 26 pixel points are
Figure BDA00024199551000000921
The neighborhood pixels.
If the similarity of the two space coordinates is determined by adopting the plurality of pixel characteristics, the similarity can be calculated after the weighted summation of the plurality of pixel characteristics; or the similarity of each pixel feature can be calculated first, and then the results of the multiple similarities are subjected to weighted summation. By adopting a plurality of pixel characteristics, the problems of weak boundary, uneven gray level and the like of a tumor area can be solved, and the accuracy of similarity calculation is improved.
Step 103, moving the mesh vertex according to the actual motion vector.
Wherein the actual motion vector is determined by the similarity, and the actual motion vector is positively correlated with the similarity.
In step 102, the larger the calculated similarity Diff value is, the larger the mesh vertex is
Figure BDA0002419955100000101
And the space coordinate
Figure BDA0002419955100000102
The more similar the pixel characteristics of (2), the mesh vertices
Figure BDA0002419955100000103
And the space coordinate
Figure BDA0002419955100000104
More likely to belong to the same region or tissue organ, spatial coordinates
Figure BDA0002419955100000105
The edge of the region of interest is not approached, and the actual motion vector can be set to be larger, so as to make the grid vertex find the edge position as soon as possible, for example, on the basis of the preset motion vector, the motion step length is increased to make the actual motion vector larger than the preset motion vector, and the grid vertex in fig. 2c is used as the reference
Figure BDA0002419955100000106
For example, mesh vertices
Figure BDA0002419955100000107
Move to Pupdate1。
The smaller the value of the similarity Diff, the mesh vertices are represented
Figure BDA0002419955100000108
And the space coordinate
Figure BDA0002419955100000109
The more dissimilar the pixel characteristics of (A) are, the more dissimilar the mesh vertices
Figure BDA00024199551000001010
And the space coordinate
Figure BDA00024199551000001011
The less likely it is that they belong to the same region or tissue organ, the less the mesh vertices
Figure BDA00024199551000001012
The closer to the edge of the region of interest, the smaller the actual motion vector is to be set, for example, based on the preset motion vector, the smaller the motion step is, the actual motion vector is smaller than the preset motion vector, or the grid vertex in fig. 2c is
Figure BDA00024199551000001013
For example, mesh vertices
Figure BDA00024199551000001014
Move to Pupdate2。
In the embodiment, for each movement of the mesh vertex, the mesh vertex is not moved by a fixed preset moving vector, but the actual moving vector is dynamically adjusted according to the size of the similarity, the actual moving step length of the mesh vertex is determined, and the position of the mesh vertex is updated, so that the reduction of accuracy caused by the overlarge setting of the preset moving vector is avoided, and the overlarge calculation caused by the overlarge setting of the preset moving vector is avoided.
Next, a detailed description will be given of a specific implementation process of determining an actual motion vector based on the similarity.
First, a step size ratio is determined according to the similarity calculated in step 102, the step size ratio is positively correlated with the similarity, and the calculation formula can be represented as follows, but is not limited to:
Figure BDA00024199551000001015
wherein,
Figure BDA00024199551000001016
representing mesh vertices
Figure BDA00024199551000001017
Step size ratio of (fstd) denotes a mesh vertex
Figure BDA00024199551000001018
The standard deviation of the pixel characteristics of the neighborhood pixels.
Then, the actual motion vector is determined according to the step length proportion and the preset motion vector
Figure BDA00024199551000001019
The calculation formula may be expressed, but is not limited to, as follows:
Figure BDA00024199551000001020
in step 103, the mesh vertices are moved according to the actual motion vectorsWith mesh vertices at the original spatial coordinates
Figure BDA0002419955100000111
Move to the space coordinate PupdateAt the position of the air compressor, the air compressor is started,
Figure BDA0002419955100000112
the process of performing steps 102 and 103 may be independent of each other for each mesh vertex, however independent of each other does not mean that the movement of the mesh vertices must be done in sequence and that the movement of all mesh vertices may be done in parallel.
In the process of parallel movement of the mesh vertices, the radius of the network model increases as all the mesh vertices of the network model approach to the edge of the region of interest, the triangular patch (or other polygonal patches) also increases, and the segmentation accuracy decreases. Based on this, in another embodiment, after step 103, it may further be determined whether the distance between two adjacent mesh vertices in the current mesh model is greater than the distance threshold, and if the distance between two adjacent mesh vertices is greater than the distance threshold, at least one spatial coordinate is selected between the connection lines of the two adjacent mesh vertices as a newly added mesh vertex of the mesh model, so that the distance between the two adjacent mesh vertices in the mesh model is not greater than the distance threshold.
And 104, judging whether the grid vertex is positioned on the edge of the region of interest.
In step 104, if the judgment is yes, the grid vertex is positioned on the edge, and the movement can be stopped; if not, the vertex of the mesh is not located on the edge, the step 102 is returned, and the moving process of the vertex of the mesh is continued.
And (5) executing steps 102 to 104 for all the mesh vertices in the mesh model, and if most mesh vertices of the mesh model are on the edge, which indicates that the mesh model can more accurately represent the boundary information of the region of interest, executing step 105.
In one embodiment, whether the number of grid vertices located on the edge of the region of interest is greater than the number threshold may be determined according to whether the pause instruction is received. With reference to the above description, in the process of segmenting the region of interest, the mesh vertices on the mesh model move along the edge direction of the mesh model, the whole process of the movement of the mesh vertices may be displayed to the user through the imaging device, and when the user observes that all or most of the mesh vertices of the mesh model are located on the edge of the region of interest, a stop instruction may be sent through the imaging device to stop the movement of the mesh vertices.
In another embodiment, whether the mesh vertex is located on the edge of the region of interest may be determined according to the similarity. For example, if the current mesh vertex is located at the spatial coordinate
Figure BDA0002419955100000113
And the space coordinate
Figure BDA0002419955100000114
The similarity of the pixel characteristics is less than the similarity threshold value, and the grid vertex is shown
Figure BDA0002419955100000121
And the space coordinate
Figure BDA0002419955100000122
Very dissimilar, mesh vertices
Figure BDA0002419955100000123
Already on or very close to the edge of the region of interest, the movement of the grid vertices may be stopped.
In another embodiment, whether the mesh vertex is located on the edge of the region of interest may be determined according to the similarity change rate. In segmenting the region of interest, the mesh vertices are typically moved through a number of iterations, e.g., one mesh vertex from spatial coordinates PaMove to spatial coordinates PbFrom the spatial coordinates PbMove to spatial coordinates Pc，PaAnd PbHas a similarity of A, PbAnd PcIs likeDegree is B, if the change rate of the similarity of A and B is small, the A and B are very close, Pa、PbAnd PcAre all in the region of interest; if the change rate of the similarity between A and B is larger, the difference between A and B is larger, and P isa、PbAnd PcIf 1 or 2 spatial coordinates are not in the region of interest, the movement of the grid vertices is stopped.
The above-described implementations of determining whether a mesh vertex is located on an edge of a region of interest may also be used in combination.
The movement of each mesh vertex can be performed independently, that is, each mesh vertex moves to the edge of the region of interest in turn, and when all the mesh vertices stop moving, it is indicated that all the mesh vertices of the network model are located on the edge of the region of interest.
The movement of each mesh vertex may also be performed synchronously, and for each iteration movement, after all the mesh vertices are moved once, the next iteration movement is performed. It can be understood that the region of interest is generally irregular in shape, and during the movement of the mesh vertices, some mesh vertices may reach the edge first, and some mesh vertices may not reach the edge, and in the following iterative movement, the mesh vertices reaching the edge may not move, and only the mesh vertices that do not reach the edge are moved until all or most of the mesh vertices are located on the edge of the region of interest.
And 105, segmenting the region of interest in the medical image according to the mesh model after the mesh vertex movement.
The mesh model used in step 105, that is, the model in which most mesh vertices of the mesh model are located on the edge, may express the boundary information of the region of interest, and may further determine all image data in the region of interest, display all image data in the region of interest after image reconstruction processing, and may provide a reference for surgical treatment.
FIG. 3 is a flow chart illustrating another method for segmenting a region of interest in a medical image according to an exemplary embodiment in which all mesh vertices are moved synchronously, the method comprising the steps of:
step 301, a mesh model is established for a region of interest of the medical image.
The specific implementation process of step 301 is similar to step 101, and is not described herein again.
And 302, determining the space coordinates of all grid vertexes of the current grid model as the start of the current round of movement iteration.
Step 303, for each mesh vertex, calculating the similarity between the pixel feature of the mesh vertex and the pixel feature of the space coordinate away from the mesh vertex by the preset motion vector.
The preset movement vector is used for determining the planned movement direction and distance of each grid vertex in the movement iteration. The distance of the preset movement vector may be the same for each mesh vertex, but the vector direction is different, and each mesh vertex may adopt a direction in which the center point of the mesh model points to the mesh vertex as the vector direction.
The specific implementation process of calculating the similarity of each mesh vertex may refer to step 102, and is not described herein again.
And step 304, for each mesh vertex, moving the mesh vertex according to the corresponding actual motion vector.
The actual motion vector of each grid vertex is determined by the similarity, and the actual motion vector is positively correlated with the similarity.
For each mesh vertex, the specific implementation process of determining the step size ratio according to the similarity calculated in step 303 and calculating the actual motion vector according to the step size ratio and the preset motion vector may refer to step 103, and details are not described here.
In another embodiment, before moving the mesh vertex, the step size ratio may be smoothed according to the neighboring mesh vertices of the mesh vertex; see FIG. 4a for a partial mesh vertex of the mesh model, illustrated as mesh vertex PaFor example, mesh vertex PaAdjacent mesh vertex of is Pb1～Pb6In the iteration, each adjacent mesh vertex can be calculated to obtain a stepThe long scale may be, but is not limited to, smoothed for each step scale by the following formula:
V′＝ω0*Va+ω1*Vb1+ω2*Vb2+ω3*Vb3+ω4*Vb4+ω5*Vb5+ω6*Vb6；
wherein V' represents the mesh vertex P after the smoothing processaStep length ratio of (1); vaRepresenting mesh vertices P determined from similarityaStep length ratio of (1); vb1～Vb6Respectively representing adjacent mesh vertices Pb1～Pb6Step length ratio determined according to the similarity; omega0～ω6Each represents the weight for a 7 step scale. VaWeight ω of (d)0Several other weights, ω, may be set1～ω6Can be based on the neighboring mesh vertices and mesh vertices PaThe distance between the two is set, the corresponding weight which is far away is smaller, and the corresponding weight which is near is larger.
In step 304, the actual motion vector is determined according to the step size ratio after the smoothing process, and the mesh vertex is moved according to the actual motion vector. The mesh vertexes are moved by using the actual motion vectors subjected to the smoothing processing, so that the moving step length of all the mesh vertexes is not too large or too small in the iteration, the image noise interference can be reduced, and the influence of larger noise on the accuracy of the segmentation result is avoided.
The more all the mesh vertices of the network model approach to the edge of the region of interest, the larger the radius of the network model, the larger the triangular patch (or other polygonal patches) is, and the lower the segmentation accuracy is. Based on this, in another embodiment, after step 304, it may be further determined whether the distance between two adjacent mesh vertices in the current mesh model is greater than the distance threshold, and if the distance between two adjacent mesh vertices is greater than the distance threshold, at least one spatial coordinate is selected between the connection lines of the two adjacent mesh vertices as a newly added mesh vertex of the mesh model, so that the distance between the two adjacent mesh vertices in the mesh model is not greater than the distance threshold.
Referring to fig. 4b, the spatial coordinates P1-P6 in the drawing are partial mesh vertices of the current mesh model after movement, and as can be seen from the drawing, the distance between the spatial coordinates P3 and P6 is relatively long, and if the distance between P3 and P6 is greater than a distance threshold, a spatial coordinate may be selected from the succession of the spatial coordinates P3 and P6, for example, the spatial coordinate P8 located at the midpoint of the connecting line is selected as an additional mesh vertex of the mesh model, and participates in the next iteration movement process. Therefore, the resolution of the grid model is improved, and the accuracy of the region of interest segmentation is further improved.
Step 305, determining whether the number of grid vertices located on the edge of the region of interest is greater than a number threshold.
In step 305, if the judgment result is yes, the fact that most grid vertexes of the network model are located on the edge is described, and the current grid model can accurately represent the boundary information of the region of interest, then step 306 is executed; if not, it is stated that the mesh model still fails to accurately represent the boundary information of the region of interest after the current iteration movement, the process returns to step 302, and the next iteration movement of the mesh vertex is performed.
In one embodiment, whether the number of grid vertices located on the edge of the region of interest is greater than the number threshold may be determined according to whether the pause instruction is received. With reference to the above description, in the process of segmenting the region of interest, the mesh vertices on the mesh model move along the edge direction of the mesh model, the whole process of the movement of the mesh vertices may be displayed to the user through the imaging device, and when the user observes that all or most of the mesh vertices of the mesh model are located on the edge of the region of interest, a stop instruction may be sent through the imaging device to stop the movement of the mesh vertices.
In another embodiment, whether the mesh vertex is located on the edge of the region of interest may be determined according to the similarity. For example, if the current mesh vertex is located at the spatial coordinate
Figure BDA0002419955100000151
And the space coordinate
Figure BDA0002419955100000156
The similarity of the pixel characteristics is less than the similarity threshold value, and the grid vertex is shown
Figure BDA0002419955100000154
And the space coordinate
Figure BDA0002419955100000153
Very dissimilar, mesh vertices
Figure BDA0002419955100000155
Already on or very close to the edge of the region of interest, the movement of the grid vertices is stopped.
In another embodiment, whether the mesh vertex is located on the edge of the region of interest may be determined according to the similarity change rate. In segmenting the region of interest, the mesh vertices are typically moved through a number of iterations, e.g., one mesh vertex from spatial coordinates PaMove to spatial coordinates PbFrom the spatial coordinates PbMove to spatial coordinates Pc，PaAnd PbHas a similarity of A, PbAnd PcIf the change rate of the similarity between A and B is small, the A and B are very close, and P isa、PbAnd PcAre all in the region of interest; if the change rate of the similarity between A and B is larger, the difference between A and B is larger, and P isa、PbAnd PcIf 1 or 2 spatial coordinates are not in the region of interest, the movement of the grid vertices is stopped.
The above-described implementations of determining whether a mesh vertex is located on an edge of a region of interest may also be used in combination.
And step 306, segmenting the region of interest in the medical image according to the mesh model of the iterative movement of the current round.
The grid model can express the boundary information of the region of interest, further determine all image data in the region of interest, display all image data in the region of interest after image reconstruction processing, and provide reference for surgical treatment.
Corresponding to the embodiments of the method, the present specification also provides embodiments of the apparatus and the terminal applied thereto.
Fig. 5 is a block diagram of a segmentation apparatus for a region of interest in a medical image, according to an exemplary embodiment, the segmentation apparatus includes: a building module 51, a calculation module 52 and a segmentation module 53.
The building module is configured to build a mesh model for a region of interest of a medical image, the mesh model having spatial coordinates in the region of interest as mesh vertices;
under the condition that the grid vertex is not located on the edge of the region of interest, the computing module is configured to compute the similarity between the pixel feature where the grid vertex coordinate is located and a space coordinate away from the grid vertex by a preset motion vector, and move the grid vertex according to an actual motion vector, wherein the distance of the actual motion vector is positively correlated with the similarity, and the direction of the actual motion vector is the same as the direction of the preset motion vector;
in the case where the mesh vertices are located on the edge of the region of interest, the segmentation module is configured to:
and segmenting the region of interest in the medical image according to the mesh model after the mesh vertex movement.
Optionally, the computing module is further configured to:
determining a step size proportion of the grid vertex according to the similarity, wherein the step size proportion is positively correlated with the similarity;
and determining the actual motion vector according to the step length proportion and the preset motion vector.
Optionally, the segmenting device further comprises: a smoothing module;
the smoothing module is configured to:
and based on the step length proportion of the adjacent mesh vertex of the mesh vertex, smoothing the step length proportion of the mesh vertex.
Optionally, the segmenting device further comprises: a module is newly added;
the add-on module is configured to:
judging whether the distance between two adjacent grid vertexes in the grid model after the grid vertexes are moved is larger than a distance threshold value or not;
if the distance between two adjacent grid vertexes is not greater than the distance threshold, selecting at least one space coordinate between connecting lines of the two adjacent grid vertexes as a newly added grid vertex of the grid model, and enabling the distance between the two adjacent grid vertexes in the grid model to be not greater than the distance threshold.
Optionally, the segmenting device further comprises: a judgment module;
the determination module is configured to:
determining that the mesh vertex is located on an edge of the region of interest if the similarity is less than a similarity threshold;
and/or determining that the grid vertex is located on the edge of the region of interest under the condition that the change rate of the similarity is smaller than a change rate threshold, wherein the change rate of the similarity is the change rate of the similarity of the grid vertex in the current iteration movement and the similarity of the grid vertex in the previous iteration movement;
and/or, in the event that a pause instruction is received, determining that the mesh vertices are located on the edge of the region of interest.
For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, wherein the modules described as separate parts may or may not be physically separate, and the parts displayed as modules may or may not be physical modules, may be located in one place, or may be distributed on a plurality of network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution in the specification. One of ordinary skill in the art can understand and implement it without inventive effort.
The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may 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 may also be possible or may be advantageous.
Fig. 6 is a schematic diagram of an electronic device according to an exemplary embodiment of the present invention, showing a block diagram of an exemplary electronic device 60 suitable for use in implementing any of the embodiments of the present invention. The electronic device 60 shown in fig. 6 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiment of the present invention.
As shown in fig. 6, the electronic device 60 may be embodied in the form of a general purpose computing device, which may be, for example, a server device. The components of the electronic device 60 may include, but are not limited to: the at least one processor 61, the at least one memory 62, and a bus 63 connecting the various system components (including the memory 62 and the processor 61).
The bus 63 includes a data bus, an address bus, and a control bus.
The memory 62 may include volatile memory, such as Random Access Memory (RAM)621 and/or cache memory 622, and may further include Read Only Memory (ROM) 623.
The memory 62 may also include a program tool 625 (or utility tool) having a set (at least one) of program modules 624, such program modules 624 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may comprise an implementation of a network environment.
The processor 61 executes various functional applications and data processing, such as the methods provided by any of the above embodiments, by running a computer program stored in the memory 62.
The electronic device 60 may also communicate with one or more external devices 64 (e.g., keyboard, pointing device, etc.) such communication may be through AN input/output (I/O) interface 65. also, the model-generated electronic device 60 may communicate with one or more networks (e.g., a local area network (L AN), a Wide Area Network (WAN) and/or a public network, such as the Internet) through a network adapter 66. As shown, the network adapter 66 communicates with other modules of the model-generated electronic device 60 through a bus 63. it should be understood that, although not shown, other hardware and/or software modules may be used in conjunction with the model-generated electronic device 60, including, but not limited to, microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, data backup storage systems, and the like.
It should be noted that although in the above detailed description several units/modules or sub-units/modules of the electronic device are mentioned, such a division is merely exemplary and not mandatory. Indeed, the features and functionality of two or more of the units/modules described above may be embodied in one unit/module according to embodiments of the invention. Conversely, the features and functions of one unit/module described above may be further divided into embodiments by a plurality of units/modules.
The present specification also provides a computer readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps of the method according to any one of the above embodiments.
Other embodiments of the present description will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This specification is intended to cover any variations, uses, or adaptations of the specification following, in general, the principles of the specification and including such departures from the present disclosure as come within known or customary practice within the art to which the specification pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the specification being indicated by the following claims.
It will be understood that the present description is not limited to the precise arrangements described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present description is limited only by the appended claims.
The above description is only a preferred embodiment of the present disclosure, and should not be taken as limiting the present disclosure, and any modifications, equivalents, improvements, etc. made within the spirit and principle of the present disclosure should be included in the scope of the present disclosure.

Claims (14)

1. A method of segmenting a region of interest in a medical image, the method comprising:
establishing a mesh model for a region of interest of a medical image, the mesh model taking spatial coordinates in the region of interest as mesh vertices;
under the condition that the grid vertex is not located on the edge of the region of interest, calculating the similarity between the pixel feature where the grid vertex coordinate is located and the pixel feature of the space coordinate away from the grid vertex by the preset moving vector, and moving the grid vertex according to an actual moving vector, wherein the distance of the actual moving vector is positively correlated with the similarity, and the direction of the actual moving vector is the same as the direction of the preset moving vector;
and under the condition that the grid vertex is positioned on the edge of the region of interest, segmenting the region of interest in the medical image according to the grid model after the grid vertex is moved.
2. The method for segmenting a region of interest in a medical image as set forth in claim 1, further comprising:
determining a step size proportion of the grid vertex according to the similarity, wherein the step size proportion is positively correlated with the similarity;
and determining the actual motion vector according to the step length proportion and the preset motion vector.
3. The method for segmenting a region of interest in a medical image as set forth in claim 2, wherein before moving the mesh vertices according to the actual motion vectors, further comprising:
based on the step length proportion of the adjacent grid vertex of the grid vertex, smoothing the step length proportion of the grid vertex;
determining the actual motion vector according to the step length proportion and the preset motion vector, wherein the step length proportion comprises the following steps:
and determining the actual motion vector according to the step length proportion subjected to the smoothing processing and the preset motion vector.
4. The method for segmenting a region of interest in a medical image as set forth in claim 1, further including, after moving the mesh vertices according to actual motion vectors:
judging whether the distance between two adjacent grid vertexes in the grid model moving through the grid vertexes is larger than a distance threshold value or not;
if the distance between two adjacent grid vertexes is not greater than the distance threshold, selecting at least one space coordinate between connecting lines of the two adjacent grid vertexes as a newly added grid vertex of the grid model, and enabling the distance between the two adjacent grid vertexes in the grid model to be not greater than the distance threshold.
5. The method for segmenting a region of interest in a medical image as set forth in claim 1, further comprising:
determining that the mesh vertex is located on an edge of the region of interest if the similarity is less than a similarity threshold;
and/or determining that the grid vertex is located on the edge of the region of interest under the condition that the change rate of the similarity is smaller than a change rate threshold, wherein the change rate of the similarity is the change rate of the similarity of the grid vertex in the current iteration movement and the similarity of the grid vertex in the previous iteration movement;
and/or determining that the mesh vertex is located on the edge of the region of interest in case of receiving a stop instruction of the mesh vertex.
6. The method for segmenting a region of interest in a medical image according to claim 1, wherein after segmenting the region of interest in the medical image according to the mesh model after moving through the mesh vertices, the method further comprises:
and performing image reconstruction processing on the image data corresponding to the segmented region of interest and displaying the image data.
7. A method of segmentation of a region of interest in a medical image as claimed in any one of the claims 1 to 6,
the pixel features include at least one of the following features: gray scale, texture, adjacency relation, overlap relation;
and/or, if the medical image is a three-dimensional image, the grid model is a three-dimensional model; if the medical image is a two-dimensional image, the grid model is a two-dimensional curve model;
and/or the direction of the actual motion vector is the direction in which the central point of the mesh model points to the mesh vertex.
8. A segmentation apparatus for a region of interest in a medical image, the segmentation apparatus comprising: the device comprises an establishing module, a calculating module and a dividing module;
the establishing module is configured to establish a mesh model for a region of interest of a medical image, the mesh model taking spatial coordinates in the region of interest as mesh vertices;
under the condition that the grid vertex is not located on the edge of the region of interest, the computing module is configured to compute the similarity between the pixel feature where the grid vertex coordinate is located and the pixel feature of the space coordinate away from the grid vertex by a preset motion vector, and move the grid vertex according to an actual motion vector, wherein the distance of the actual motion vector is positively correlated with the similarity, and the direction of the actual motion vector is the same as the direction of the preset motion vector;
in a case where the mesh vertices are located on the edges of the region of interest, the segmentation module is configured to segment the region of interest in the medical image according to the mesh model moved by the mesh vertices.
9. The apparatus for segmenting a region of interest in a medical image according to claim 8, wherein said computing module is further configured to:
determining a step size proportion of the grid vertex according to the similarity, wherein the step size proportion is positively correlated with the similarity;
and determining the actual motion vector according to the step length proportion and the preset motion vector.
10. The apparatus for segmenting a region of interest in a medical image as set forth in claim 9, further comprising: a smoothing module;
the smoothing module is configured to:
and based on the step length proportion of the adjacent mesh vertex of the mesh vertex, smoothing the step length proportion of the mesh vertex.
11. The apparatus for segmenting a region of interest in a medical image as set forth in claim 8, further comprising: a module is newly added;
the add-on module is configured to:
judging whether the distance between two adjacent grid vertexes in the grid model after the grid vertexes are moved is larger than a distance threshold value or not;
if the distance between two adjacent grid vertexes is not greater than the distance threshold, selecting at least one space coordinate between connecting lines of the two adjacent grid vertexes as a newly added grid vertex of the grid model, and enabling the distance between the two adjacent grid vertexes in the grid model to be not greater than the distance threshold.
12. The apparatus for segmenting a region of interest in a medical image as set forth in claim 8, further comprising: a judgment module;
the determination module is configured to:
determining that the mesh vertex is located on an edge of the region of interest if the similarity is less than a similarity threshold;
and/or determining that the grid vertex is located on the edge of the region of interest under the condition that the change rate of the similarity is smaller than a change rate threshold, wherein the change rate of the similarity is the change rate of the similarity of the grid vertex in the current iteration movement and the similarity of the grid vertex in the previous iteration movement;
and/or, in the event that a pause instruction is received, determining that the mesh vertices are located on the edge of the region of interest.
13. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method for segmenting a region of interest in a medical image according to any one of claims 1 to 7 when executing the computer program.
14. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method of segmentation of a region of interest in a medical image of any one of claims 1 to 7.
CN202010202834.6A 2020-03-20 2020-03-20 Method and device for segmenting region of interest in medical image Active CN111445479B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010202834.6A CN111445479B (en) 2020-03-20 2020-03-20 Method and device for segmenting region of interest in medical image

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010202834.6A CN111445479B (en) 2020-03-20 2020-03-20 Method and device for segmenting region of interest in medical image

Publications (2)

Publication Number Publication Date
CN111445479A true CN111445479A (en) 2020-07-24
CN111445479B CN111445479B (en) 2023-08-29

Family

ID=71629454

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010202834.6A Active CN111445479B (en) 2020-03-20 2020-03-20 Method and device for segmenting region of interest in medical image

Country Status (1)

Country Link
CN (1) CN111445479B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113393462A (en) * 2021-08-18 2021-09-14 浙江杜比医疗科技有限公司 Focal region segmentation method and device, electronic equipment and storage medium
CN115690556A (en) * 2022-11-08 2023-02-03 河北北方学院附属第一医院 Image recognition method and system based on multi-modal iconography characteristics

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102467667A (en) * 2010-11-11 2012-05-23 江苏大学 Medical image classification method
CN105678747A (en) * 2015-12-30 2016-06-15 青岛兰信医学科技有限公司 Tooth mesh model automatic segmentation method based on principal curvature
CN106875487A (en) * 2017-03-01 2017-06-20 中国石油大学(华东) A kind of geology hexahedral mesh smoothing method based on neighborhood active force

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102467667A (en) * 2010-11-11 2012-05-23 江苏大学 Medical image classification method
CN105678747A (en) * 2015-12-30 2016-06-15 青岛兰信医学科技有限公司 Tooth mesh model automatic segmentation method based on principal curvature
CN106875487A (en) * 2017-03-01 2017-06-20 中国石油大学(华东) A kind of geology hexahedral mesh smoothing method based on neighborhood active force

Non-Patent Citations (9)

* Cited by examiner, † Cited by third party
Title
党建武;杨旭;王阳萍;: "基于子区域相似度的医学图像分割算法" *
党建武等: "基于子区域相似度的医学图像分割算法", 《计算机应用》 *
李光明等: "基于距离均衡化的网格平滑算法", 《计算机辅助设计与图形学学报》 *
李智琦等: "保持特征的三角网格曲面去噪技术综述与展望", 《计算机辅助设计与图形学学报》 *
薛凌云等: "医学图像感兴趣区域的提取", 《杭州电子科技大学学报》 *
谢从华 等: ""基于动态步长的医学图像聚类分割研究"" *
谢从华 等: "基于动态步长的医学图像聚类分割研究" *
郑汉翔;王美清;: "基于区域的活动轮廓图像分割模型的变步长优化算法" *
陈文山: "基于三角网格模型的心脏超声图像三维表面重建", 《福建电脑》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113393462A (en) * 2021-08-18 2021-09-14 浙江杜比医疗科技有限公司 Focal region segmentation method and device, electronic equipment and storage medium
CN115690556A (en) * 2022-11-08 2023-02-03 河北北方学院附属第一医院 Image recognition method and system based on multi-modal iconography characteristics

Also Published As

Publication number Publication date
CN111445479B (en) 2023-08-29

Similar Documents

Publication Publication Date Title
Lamecker et al. Segmentation of the liver using a 3D statistical shape model
JP4152648B2 (en) Method for segmenting a 3D image contained in an object
US9659364B2 (en) Probabilistic refinement of model-based segmentation
Little et al. Deformations incorporating rigid structures
US8218905B2 (en) Method, system and software product for providing efficient registration of 3D image data
US9715754B2 (en) In-plane and interactive surface mesh adaptation
US8135189B2 (en) System and method for organ segmentation using surface patch classification in 2D and 3D images
US20100189320A1 (en) Method of Segmenting Anatomic Entities in 3D Digital Medical Images
CN110251231A (en) The method and device that ultrasonic three-dimensional is rebuild
EP2238573A2 (en) Automatic 3-d segmentation of the short-axis late-enhancement cardiac mri
EP1171033A1 (en) Method and system of measuring characteristics of an organ
WO2003090173A2 (en) Segmentation of 3d medical structures using robust ray propagation
US20090202150A1 (en) Variable resolution model based image segmentation
US8018456B2 (en) Smooth gray-level based surface interpolation for an isotropic data sets
van Assen et al. Cardiac LV segmentation using a 3D active shape model driven by fuzzy inference
CN108885797B (en) Imaging system and method
WO2010023618A1 (en) Segmentation of the long-axis late-enhancement cardiac mri
CN107516314B (en) Medical image hyper-voxel segmentation method and device
CN111445479B (en) Method and device for segmenting region of interest in medical image
US10922853B2 (en) Reformatting while taking the anatomy of an object to be examined into consideration
EP2016557A1 (en) A method, a system, a computer program product and a user interface for segmenting image sets
KR102689375B1 (en) Skeleton estimate apparatus using multiple x-ray views and method thereof
EP2279489A2 (en) Mesh collision avoidance
Leonardi et al. Multiple reconstruction and dynamic modeling of 3D digital objects using a morphing approach: Application to kidney animation and tumor tracking
CN101719274A (en) Three-dimensional texture analyzing method of medicinal image data

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
点击 这是indexloc提供的php浏览器服务，不要输入任何密码和下载