CN107481252A - Dividing method, device, medium and the electronic equipment of medical image - Google Patents
Dividing method, device, medium and the electronic equipment of medical image Download PDFInfo
- Publication number
- CN107481252A CN107481252A CN201710733452.4A CN201710733452A CN107481252A CN 107481252 A CN107481252 A CN 107481252A CN 201710733452 A CN201710733452 A CN 201710733452A CN 107481252 A CN107481252 A CN 107481252A
- Authority
- CN
- China
- Prior art keywords
- image
- pixel
- profile curve
- medical image
- sub
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10081—Computed x-ray tomography [CT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
- G06T2207/10088—Magnetic resonance imaging [MRI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10132—Ultrasound image
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention provides a kind of dividing method of medical image, device, medium and electronic equipment.The dividing method of the medical image includes:Obtain medical image sequences to be split;First image in the medical image sequences is handled based on algorithm of region growing, obtains the initial profile curve of the cut zone for identifying first image;Based on the initial profile curve, first image is handled according to level set function, obtains the final profile curve of first image;For every image in the medical image sequences, the initial profile curve using the final profile curve of previous image as next image, and every image is handled based on level set function, to obtain the final profile curve of every image;Final profile curve based on every image is split to every image.The technical scheme of the embodiment of the present invention can improve the efficiency and accuracy of image segmentation.
Description
Technical field
The present invention relates to technical field of medical image processing, in particular to a kind of dividing method of medical image, dress
Put, medium and electronic equipment.
Background technology
Medical Image Segmentation Techniques are the key technologies in Medical Image Processing and analysis.Medical image segmentation is a root
Carry out the process about structure (or area-of-interest) in separate picture according to the similitude in region and interregional difference.In early days
Image segmentation be entirely manually to complete, complete artificial dividing method is medical expert in hundreds of sectioning image
The manual delineation work of enterprising row bound, the three-dimensional structure of focus and its surrounding tissue is conceived according to the manual delineation result on border
And spatial relationship, and in this, as the basis for formulating treatment plan.
With computer tomography (Computed Tomography, abbreviation CT), Magnetic resonance imaging (Magnetic
Resonance Image, abbreviation MRI) and the medical imaging technology such as ultrasonic imaging (Ultrasound Image) medical diagnosis,
The extensive use of the links such as preoperative planning, postoperative detection, it has been difficult to meet manually to split this method to waste time and energy
High demand of the people to medical image segmentation.Therefore, study the automatic division method of practicality and cumbersome people's work point is finally replaced
It is always the target that people pursue to cut the semi-automatic segmentation very strong with subjectivity.
At present, mainly there are algorithm of region growing and level set algorithm etc. for the algorithm of image segmentation.Wherein, region growing
Algorithm is from some sub-pixel point, according to certain growth criterion, is gradually added neighborhood pixels, adds when without neighborhood pixels
Fashionable, region growing terminates.The advantages of algorithm of region growing is that principle is simple, implements and is relatively easy to, but region growing
Algorithm can not automatically process the three-dimensional structure of complexity, such as at vascular bifurcation, and the method is more sensitive to noise, noise and ash
Degree inhomogeneities may produce cavity or over-segmentation.
The basic thought of level set algorithm is the set that closed contour is expressed as to higher-dimension curved surface equivalent point, a series of interior
In the presence of power and external force, the evolutionary process of profile is obtained by evolution level set function and the zero level collection that tracks it.It is horizontal
Set algorithm possesses great topological changeability, can tackle the three-dimensional structure of complexity, but level set algorithm is to first closure
Profile is more sensitive, and inappropriate initial profile may result in over-segmentation, closed curve and not restrain or reveal.
Because medical image is often shown as between low contrast, tissue characteristics changeabilities and different soft tissues or soft tissue
The obscurity boundary between focus, and shape and structure and the complexity of fine structure (blood vessel) distribution, therefore area is used alone
Domain growth algorithm and level set algorithm are difficult to complete the automatic segmentation to medical image.
It should be noted that information is only used for strengthening the reason of the background to the present invention disclosed in above-mentioned background section
Solution, therefore can include not forming the information to prior art known to persons of ordinary skill in the art.
The content of the invention
It is an object of the invention to provide a kind of dividing method of medical image, device, medium and electronic equipment, Jin Erzhi
It is few to overcome one or more problem caused by the limitation of correlation technique and defect to a certain extent.
Other characteristics and advantage of the present invention will be apparent from by following detailed description, or partially by the present invention
Practice and acquistion.
First aspect according to embodiments of the present invention, there is provided a kind of dividing method of medical image, including:Obtain to be split
Medical image sequences;First image in the medical image sequences is handled based on algorithm of region growing, obtained
For the initial profile curve for the cut zone for identifying first image;Based on the initial profile curve, according to level
Set function is handled first image, obtains the final profile curve of first image;For the medical science
Every image in image sequence, the initial profile curve using the final profile curve of previous image as next image,
And every image is handled based on level set function, to obtain the final profile curve of every image;It is based on
The final profile curve of every image is split to every image.
In some embodiments of the invention, based on aforementioned schemes, based on algorithm of region growing to the medical image sequence
First image in row is handled, including:Obtain the seed that user is set in the target area of first image
Pixel;Storehouse is established, and the sub-pixel is stored in the storehouse;Sub-pixel is taken out from the storehouse successively, its
In, for any sub-pixel taken out from the storehouse, if any neighborhood territory pixel of any sub-pixel meets area
Domain grows criterion, then is stored in any neighborhood territory pixel as sub-pixel in the storehouse;It is space-time in the storehouse, root
The region formed according to all sub-pixels determines the initial profile curve.
In some embodiments of the invention, based on aforementioned schemes, in addition to:According to the gray value of the sub-pixel and
The gray value of the neighborhood territory pixel of the sub-pixel, calculate the threshold parameter of the region growing criterion.
In some embodiments of the invention, based on aforementioned schemes, according to the gray value of the sub-pixel and the kind
The gray value of the neighborhood territory pixel of sub-pixel, the threshold parameter of the region growing criterion is calculated, including:According to the sub-pixel
Gray value and the sub-pixel neighborhood territory pixel gray value, calculate the gray value of the sub-pixel and the seed picture
Variance between the gray value of the neighborhood territory pixel of element;According to the variance, the threshold parameter of the region growing criterion is determined.
In some embodiments of the invention, based on aforementioned schemes, in addition to:Calculate the gray scale of any sub-pixel
Variable quantity between the gray value of the neighborhood territory pixel of value and any sub-pixel;If in the neighborhood of any sub-pixel
The gray value of any pixel and the gray value of any sub-pixel between variable quantity be less than the region growing criterion
Threshold parameter, it is determined that any pixel meets the region growing criterion.
In some embodiments of the invention, it is based on aforementioned schemes, the neighborhood territory pixel:Four neighborhood territory pixels or eight neighborhood
Pixel.
In some embodiments of the invention, based on aforementioned schemes, based on the level set function and any image just
Beginning contour curve, any image is handled, including:The initial profile curve is set in any image;
Based on the initial profile curve and the evolution parameter set, the level set function is developed;In the level set letter
When iterations when number is developed reaches pre-determined number, zero level collection is extracted from the level set function, to be used as institute
State the final profile curve of any image.
In some embodiments of the invention, based on aforementioned schemes, based on the initial profile curve and the evolution set
Parameter, the level set function is developed, including:Based on the initial profile curve, energy functional model is built;Pass through
Level set function replaces the initial profile curve in the energy functional model;Evolution parameter based on setting, to the energy
Equation is solved corresponding to functional model, to develop to the level set function.
In some embodiments of the invention, it is as follows based on aforementioned schemes, the energy functional model:
E=α ECV+β·EL+ER
Wherein, E represents the energy functional model;ECVRepresent the energy function of CV models;ELRepresent the energy of local entity
Function;ERRepresent the energy function of regularization term;α and β represents control parameter.
In some embodiments of the invention, based on aforementioned schemes, the energy of the CV models is calculated by equation below
Function:
ECV(C, c1, c2)=λ1∫outside(C)|I(x)-c1|2dx+λ2∫inside(C)|I(x)-c2|2dx+v|C|
Wherein, C represents the initial profile curve;Outside (C) and inside (C) represent the initial profile respectively
Curve C exterior domain and inner region;c1And c2Represent constant;λ1、λ2Control parameter is represented with v;I (x) represents any image
In pixel x gray value.
In some embodiments of the invention, based on aforementioned schemes, the energy of the local entity is calculated by equation below
Function:
EL=∫Ωεx(C, f1, f2)dx
Wherein, εx(C, f1, f2)=λ1∫inside(C)gσ(x-y)(I(υ)-f1(x))2dy+λ2∫outside(C)gσ(x-y)(I
(υ)-f2(x))2dy;λ1And λ2Represent control parameter;f1And f (x)2(x) it is initial profile curve C both sides topography gray scale
Preferable approximation;gσRepresent gaussian kernel function;I (υ) represents the gray value of the pixel υ in any image.
In some embodiments of the invention, based on aforementioned schemes, the gaussian kernel function is calculated by below equation:
Wherein, u represents constant;σ represents scale parameter;N represents iterations.
In some embodiments of the invention, based on aforementioned schemes, the energy of the regularization term is calculated by equation below
Function:
ER(φ)=μ L (φ=0)+P (φ)
Wherein, φ represents level set function;μ represents control parameter;L (φ=0) represents length penalty term,H (z) and δ (z) are represented respectively
Heaviside functions and Dirac functions;P (φ) represents energy penalty term,
In some embodiments of the invention, based on aforementioned schemes, in addition to:Based on in the medical image sequences
The segmentation result of every image, carry out the three-dimensional modeling of medical image.
Second aspect according to embodiments of the present invention, there is provided a kind of segmenting device of medical image, including:Obtain single
Member, for obtaining medical image sequences to be split;First processing units, for based on algorithm of region growing to the medical science figure
As first image in sequence is handled, the initial profile for obtaining cut zone for identifying first image is bent
Line;Second processing unit, for based on the initial profile curve, according to level set function to first image at
Reason, the final profile curve of first image, and every image for being directed in the medical image sequences are obtained, will
Initial profile curve of the final profile curve of previous image as next image, and based on level set function to described every
Open image to be handled, to obtain the final profile curve of every image;Cutting unit, for based on every image
Final profile curve every image is split.
The third aspect according to embodiments of the present invention, there is provided a kind of computer-readable medium, be stored thereon with computer
Program, the segmentation side of the medical image as described in first aspect in above-mentioned embodiment is realized when described program is executed by processor
Method.
Fourth aspect according to embodiments of the present invention, there is provided a kind of electronic equipment, including:One or more processors;
Storage device, for storing one or more programs, when one or more of programs are held by one or more of processors
During row so that one or more of processors realize the segmentation side of the medical image as described in first aspect in above-mentioned embodiment
Method.
In the technical scheme that some embodiments of the present invention are provided, by based on algorithm of region growing to medical image
First image in sequence is handled, and to obtain the initial profile curve of first image, is then based on level set function
Final profile curve is obtained to first image procossing with obtained initial profile curve, enabling first calculated based on region growing
Method gets the initial profile curve of first image, and then on the basis of obtained initial profile curve, can improve water
The Evolution Rates and accuracy of flat set function, avoid in use level set function because initial profile curve selection is improper
The problem of causing over-segmentation, closed curve not to restrain or reveal.Due to the adjacent image in medical image sequences have it is extremely strong
Correlation, therefore by the way that the final profile curve of previous image in medical image sequences to be used as to the initial wheel of next image
Wide curve, the processing time of algorithm can not only be shortened, and can ensure that level set algorithm has more accurate result.
In the technical scheme that some embodiments of the present invention are provided, sub-pixel is handled by establishing storehouse, is made
Obtaining can in an orderly manner discharge and store sub-pixel, improve the operation efficiency of algorithm of region growing, and can eliminate region
Repetitive operation of the growth algorithm to some or some sub-pixels.
In the technical scheme that some embodiments of the present invention are provided, pass through the gray value and seed according to sub-pixel
The gray value of the neighborhood territory pixel of pixel carrys out the threshold parameter of zoning growth criterion, enabling considers the local message of image
Carry out the threshold parameter of zoning growth criterion, eliminate the subjectivity that artificial interference is brought, improve algorithm of region growing
Treatment effeciency and accuracy.
In the technical scheme that some embodiments of the present invention are provided, by adding local entity in energy functional model
Energy function, reduce susceptibility of the level set function to the uneven characteristic of gradation of image.By in energy functional model
Adding the energy function of regularization term so that level set function can keep smooth in evolutionary process, and avoid reinitializing,
The operation efficiency of level set algorithm is accelerated, reduces processing time.
It should be appreciated that the general description and following detailed description of the above are only exemplary and explanatory, not
Can the limitation present invention.
Brief description of the drawings
Accompanying drawing herein is merged in specification and forms the part of this specification, shows the implementation for meeting the present invention
Example, and for explaining principle of the invention together with specification.It should be evident that drawings in the following description are only the present invention
Some embodiments, for those of ordinary skill in the art, on the premise of not paying creative work, can also basis
These accompanying drawings obtain other accompanying drawings.In the accompanying drawings:
Fig. 1 diagrammatically illustrates the flow chart of the dividing method of medical image according to an embodiment of the invention;
Fig. 2 diagrammatically illustrates first based on algorithm of region growing to image sequence according to an embodiment of the invention
The flow chart that image is handled;
Fig. 3 diagrammatically illustrates the flow according to an embodiment of the invention handled based on level set function image
Figure;
Fig. 4 is diagrammatically illustrated according to an embodiment of the invention carries out figure based on algorithm of region growing and level set algorithm
As the flow chart of segmentation;
Fig. 5 shows the initial profile for the vascular root that application region growth algorithm according to an embodiment of the invention obtains
The effect diagram of line;
Fig. 6 show it is according to an embodiment of the invention randomly selected from processed image obtained contour line
Image schematic diagram;
Fig. 7 shows the topological changeability schematic diagram of level set function according to an embodiment of the invention;
Fig. 8 shows the arteria carotis that the three dimensional reconstruction according to an embodiment of the invention based on objective contour obtains
Threedimensional model schematic diagram;
Fig. 9 diagrammatically illustrates the block diagram of the segmenting device of medical image according to an embodiment of the invention;
Figure 10 diagrammatically illustrates the structure of the computer system suitable for being used for the electronic equipment for realizing the embodiment of the present invention
Schematic diagram.
Embodiment
Example embodiment is described more fully with referring now to accompanying drawing.However, example embodiment can be with a variety of shapes
Formula is implemented, and is not understood as limited to example set forth herein;On the contrary, these embodiments are provided so that the present invention will more
Fully and completely, and by the design of example embodiment comprehensively it is communicated to those skilled in the art.
In addition, described feature, structure or characteristic can be incorporated in one or more implementations in any suitable manner
In example.In the following description, there is provided many details fully understand so as to provide to embodiments of the invention.However,
It will be appreciated by persons skilled in the art that technical scheme can be put into practice without one or more in specific detail,
Or other methods, constituent element, device, step etc. can be used.In other cases, side known in being not shown in detail or describe
Method, device, realization are operated to avoid fuzzy each aspect of the present invention.
Block diagram shown in accompanying drawing is only functional entity, not necessarily must be corresponding with physically separate entity.
I.e., it is possible to realize these functional entitys using software form, or realized in one or more hardware modules or integrated circuit
These functional entitys, or these functional entitys are realized in heterogeneous networks and/or processor device and/or microcontroller device.
Flow chart shown in accompanying drawing is merely illustrative, it is not necessary to including all contents and operation/step,
It is not required to perform by described order.For example, some operation/steps can also decompose, and some operation/steps can close
And or partly merging, therefore the order actually performed is possible to be changed according to actual conditions.
Fig. 1 diagrammatically illustrates the flow chart of the dividing method of medical image according to an embodiment of the invention.
Reference picture 1, the dividing method of medical image according to an embodiment of the invention, including:
Step S10, obtain medical image sequences to be split.
It should be noted that medical image sequences to be split can be various types of images, for example can be based on
The medical image of the acquisitions such as Computed tomography, nmr imaging technique and ultrasonic imaging technique.
Step S12, first image in the medical image sequences is handled based on algorithm of region growing, obtained
For the initial profile curve for the cut zone for identifying first image.
In step s 12, by being handled based on algorithm of region growing first image in medical image sequences,
Obtain initial profile curve so that subsequently when use level set function is developed, the evolution for the set function that can improve the standard
Speed and accuracy, avoiding causes over-segmentation because initial profile curve selection is improper in use level set function, closes
Close the problem of curve is not restrained or revealed.
In an exemplary embodiment of the present invention, step S12 is specifically included:User is obtained in the target area of first image
The sub-pixel set in domain;Storehouse is established, and the sub-pixel is stored in the storehouse;Taken successively from the storehouse
Go out sub-pixel, wherein, for any sub-pixel taken out from the storehouse, if any neighbour of any sub-pixel
Domain pixel meets region growing criterion, then is stored in any neighborhood territory pixel as sub-pixel in the storehouse;Described
Storehouse is space-time, and the initial profile curve is determined according to the region that all sub-pixels are formed.
In this embodiment, sub-pixel is handled by establishing storehouse, enabling discharge in an orderly manner and store seed
Pixel, the operation efficiency of algorithm of region growing is improved, and algorithm of region growing can be eliminated to some or some seed pictures
The repetitive operation of element.
Further, on the basis of the technical scheme of above-described embodiment, in addition to:According to the gray scale of the sub-pixel
The gray value of the neighborhood territory pixel of value and the sub-pixel, calculate the threshold parameter of the region growing criterion.
In an exemplary embodiment of the present invention, specifically, above-mentioned gray value according to the sub-pixel and described
The gray value of the neighborhood territory pixel of sub-pixel, the threshold parameter of the region growing criterion is calculated, including:According to the seed picture
The gray value of the gray value of element and the neighborhood territory pixel of the sub-pixel, calculate the gray value of the sub-pixel and the seed
Variance between the gray value of the neighborhood territory pixel of pixel;According to the variance, the threshold parameter of the region growing criterion is determined.
In this embodiment, counted by the gray value according to sub-pixel and the gray value of the neighborhood territory pixel of sub-pixel
Calculate the threshold parameter of region growing criterion, enabling consider that the local message of image carrys out the threshold value ginseng of zoning growth criterion
Number, improve the accuracy of algorithm of region growing.
In some embodiments of the invention, based on aforementioned schemes, in addition to:Calculate the gray scale of any sub-pixel
Variable quantity between the gray value of the neighborhood territory pixel of value and any sub-pixel;If in the neighborhood of any sub-pixel
The gray value of any pixel and the gray value of any sub-pixel between variable quantity be less than the region growing criterion
Threshold parameter, it is determined that any pixel meets the region growing criterion.
It should be noted that in an embodiment of the present invention, above-mentioned neighborhood territory pixel both can be four neighborhood territory pixels, also may be used
To be eight neighborhood pixel.
With continued reference to Fig. 1, the dividing method of shown medical image also includes:
Step S14, based on the initial profile curve, first image is handled according to level set function,
Obtain the final profile curve of first image.
Step S16, for every image in the medical image sequences, the final profile curve of previous image is made
For the initial profile curve of next image, and every image is handled based on level set function, it is described to obtain
The final profile curve of every image.
Wherein, it is directed in step S14 and step S16 according to level set function the mistake that is handled image
Journey, in an exemplary embodiment of the present invention, the initial profile curve based on level set function and any image, to described any
Image, which carries out processing, to be included:The initial profile curve is set in any image;Based on the initial profile curve and
The evolution parameter of setting, the level set function is developed;Iterations when the level set function is developed
When reaching pre-determined number, zero level collection is extracted from the level set function, it is bent using the final profile as any image
Line.
In some embodiments of the invention, based on aforementioned schemes, based on the initial profile curve and the evolution set
Parameter, the level set function is developed, including:Based on the initial profile curve, energy functional model is built;Pass through
Level set function replaces the initial profile curve in the energy functional model;Evolution parameter based on setting, to the energy
Equation is solved corresponding to functional model, to develop to the level set function.
In some embodiments of the invention, it is as follows based on aforementioned schemes, the energy functional model:
E=α ECV+β·EL+ER
Wherein, E represents the energy functional model;ECVRepresent the energy function of CV models;ELRepresent the energy of local entity
Function;ERRepresent the energy function of regularization term;α and β represents control parameter.
In an embodiment of the present invention, by adding the energy function of local entity in energy functional model, water is reduced
Susceptibility of the flat set function to the uneven characteristic of gradation of image.And by adding the energy letter of regularization term in energy functional model
Number so that level set function can keep smooth in evolutionary process, and avoid reinitializing, and accelerate level set algorithm
Operation efficiency, reduce processing time.
In some embodiments of the invention, based on aforementioned schemes, the energy of the CV models is calculated by equation below
Function:
ECV(C, c1, c2)=λ1∫outside(C)|I(x)-c1|2dx+λ2∫inside(C)|I(x)-c2|2dx+v|C|
Wherein, C represents the initial profile curve;Outside (C) and inside (C) represent the initial profile respectively
Curve C exterior domain and inner region;c1And c2Represent constant;λ1、λ2Control parameter is represented with v;I (x) represents any image
In pixel x gray value.
In some embodiments of the invention, based on aforementioned schemes, the energy of the local entity is calculated by equation below
Function:
EL=∫Ωεx(C, f1, f2)dx
Wherein, εx(C, f1, f2)=λ1∫inside(C)gσ(x-y)(I(υ)-f1(x))2dy+λ2∫outside(C)gσ(x-y)(I
(υ)-f2(x))2dy;λ1And λ2Represent control parameter;f1And f (x)2(x) it is initial profile curve C both sides topography gray scale
Preferable approximation;gσRepresent gaussian kernel function;I (υ) represents the gray value of the pixel υ in any image.
In some embodiments of the invention, based on aforementioned schemes, the gaussian kernel function is calculated by below equation:
Wherein, u represents constant;σ represents scale parameter;N represents iterations.
In some embodiments of the invention, based on aforementioned schemes, the energy of the regularization term is calculated by equation below
Function:
ER(φ)=μ L (φ=0)+P (φ)
Wherein, φ represents level set function;μ represents control parameter;L (φ=0) represents length penalty term,H (z) and δ (z) difference tables
Show Heaviside functions and Dirac functions;P (φ) represents energy penalty term,
With continued reference to Fig. 1, the dividing method of shown medical image also includes:
Step S18, the final profile curve based on every image are split to every image.
In an embodiment of the present invention, after splitting to medical image, it is also based on to medical image sequences
In every image segmentation result, carry out the three-dimensional modeling of medical image.
In summary, in order to overcome the shortcomings of that single image dividing method is applied in medical image segmentation and reply medical science
The complexity of image, embodiments of the invention propose the image for being combined algorithm of region growing and level set function and split calculation
Method, realize the automatic segmentation to the blood vessel in medical image, bone and portion of tissue organ.
Wherein, the algorithm of region growing proposed in the embodiment of the present invention is by area-of-interest (Region of
Interest the similitude of gray value) merges similar gray scale as region growing criterion according to the threshold parameter of setting
The pixel of value.Because the threshold parameter of traditional algorithm of region growing is usually manually according to the intensity value ranges of area-of-interest
Set, there is stronger subjectivity, in easy region cavity or the problem of over-segmentation, therefore embodiments of the invention
Propose threshold parameter be by the variance between the gray value of sub-pixel and its eight neighborhood (or four neighborhoods) grey scale pixel value Lai
Establish, eliminate artificial interference, improve the accuracy of algorithm.Meanwhile algorithm of region growing is a kind of serial algorithm, works as target
When region is very big, splitting speed is slower, therefore the principle for introducing storehouse in an embodiment of the present invention carrys out design section growth
Algorithm, release in order and storage sub-pixel are realized, the treatment effeciency of algorithm had both been improved, and had also eliminated repetition and be incorporated to seed
The mistake generation of pixel.
In addition, traditional Level Set Method is assuming that gradation of image is statistically to maintain the basis of certain uniformity
On, global energy fitting is carried out to the interior exterior domain of objective contour curve, this method is obviously uneven to gray scale and content is answered
Miscellaneous medical image is unsuitable, it is easy to is occurring the problem of extensive leakage during level set movements.Therefore exist
Following improve is proposed in the embodiment of the present invention to Level Set Method:
1st, local message energy fit term is added, for solving the problems, such as that gradation of image is uneven;
2nd, regularization term is added, maintains slickness of the level set function in evolutionary process, and avoid reinitializing.
In addition, the contour curve that embodiments of the invention using area growth algorithm obtains is as the initial of level set function
Change curve, reduce susceptibility of the level set function to initialization curve, improve the Evolution Rates of level set function and accurate
Degree.
Specifically, in an embodiment of the present invention, figure is carried out based on improved algorithm of region growing and level set algorithm
As the process of segmentation mainly includes the following steps that:
Step S1:Read original sequence;
Step S2:Using area growth algorithm is handled first image of image sequence, obtains initial profile line;
Step S3:The initial profile line that use level set function and step S2 are obtained is handled first image, is obtained
To the final profile line of first figure;
Step S4:Initial profile line using the final profile line of a upper image as next figure, and use level collection
Next image of function pair is handled;
Step S5:Repeat step S4, until whole image sequence all has been processed into;
Step S6:Image segmentation is carried out based on the result to whole image sequence.Can after image, which is split, to be completed
To carry out three-dimensional modeling, the threedimensional model of destination organization is obtained.
For above-mentioned steps S2, specifically, in one embodiment of the invention, reference picture 2, following step is specifically included
Suddenly:
Step S201, read first image in image sequence.
Step S202, obtain one or more sub-pixels that user is set in target area.
Step S203, the threshold parameter T of zoning growth algorithm.
Specifically, threshold value ginseng can be calculated according to the gray value of sub-pixel and the variance of the gray value of its eight neighborhood pixel
Number T, as shown in Equation 1:
In equation 1, x represents the average of the gray value of the eight neighborhood pixel of sub-pixel, xiRepresent the eight of sub-pixel
The gray value of ith pixel in neighborhood territory pixel.
It should be noted that in other embodiments of the invention, can also according to the gray value of sub-pixel and its four
The variance of the gray value of neighborhood territory pixel calculates threshold parameter T.
Step S204, establishes storehouse.
Step S205, sub-pixel is pressed into storehouse.
Step S206, storehouse eject a sub-pixel.
Step S207, judges whether the eight neighborhood pixel of sub-pixel meets growing strategy, if any eight neighborhood pixel expires
Sufficient growing strategy, then the pixel is incorporated to target area, and is set to sub-pixel press-in storehouse, return to step S205.Wherein, grow
Rule is defined by formula 2, as follows:
In formula 2, I represents original image, I1The threshold binary image corresponding with original image is represented, N is threshold binary image
Gray value, the gray value in initial profile line is N, and the gray value outside contour line is-N.User need to be according to different target areas
Domain sets different N value, and in general, between may be configured as 100-200 according to tissue regions difference, too small N values can be led
Cause level set function leakage.
If above-mentioned formula 2 represent implication it is to be understood that the gray value of any eight neighborhood pixel of sub-pixel with should
Variable quantity between the gray value of sub-pixel is less than threshold parameter T, it is determined that the eight neighborhood pixel meets Rule of Region-growth.
Step S208, judge whether storehouse is empty, if storehouse is sky, is stopped growing;If storehouse is not sky, return
Step S206.
Before step S3 is introduced, the improvement content of the level set function of the embodiment of the present invention is illustrated:
The evolutionary process of level set function is the minimum process of energy functional, and the energy used in the embodiment of the present invention is general
Letter is the improvement to traditional CV models, and the formula of traditional CV models is as shown in Equation 3:
ECV(C, c1, c2)=
λ1∫outside(C)|I(x)-c1|2dx+λ2∫inside(C)|I(x)-c2|2Dx+v | C | formula 3
In equation 3, outside (C) and inside (C) represents contour line C exterior domain and inner region respectively.c1And c2
It is two constant approximation outside (C) and inside (C) area grayscale value.Ideally it is used for minimizing energy functional
Constant c1And c2It is outside (C) and inside (C) area grayscale average value respectively, but when the gray scale of image is uneven
When, c1And c2Raw image data can be deviateed, and do not include any local gray level information, therefore traditional CV models are difficult processing
The uneven medical image of gray scale.In an embodiment of the present invention, the improvement to traditional CV models is as follows:
1st, local entity E is addedL, by global and local information fusion together so that cutting procedure not by gray scale not
It is uniform to influence;
2nd, regularization term E is addedR, the smooth of zero level collection is kept, avoids occurring smaller and isolated area in segmentation result.
Therefore the energy functional model in the embodiment of the present invention is as shown in Equation 4:
E=α ECV+β·EL+ERFormula 4
The embodiment of the present invention propose improved Level Set Models why can the uneven image of successful division gray scale, close
Key is that the local entity E introduced using local statistic informationL.The thought of local entity is that the weighting of pixel value in local window is put down
The energy value of the point is approached, therefore obtains the energy function of single-point as shown in Equation 5:
εx(C, f1, f2)=λ1∫inside(C)gσ(x-y)(I(υ)-f1(x))2dy+λ2∫outside(C)gσ(x-y)(I(υ)-f2
(x))2Dy formula 5
Due to energy εxBe defined on each dotted line x in image, therefore integrated on whole space Ω, then
To total energy function EL, as shown in Equation 6:
EL=∫Ωεx(C, f1, f2) dx formula 6
In formula 5 and formula 6, λ1And λ2Represent control parameter;f1And f (x)2(x) it is the Local map of contour line C both sides
As the preferable approximation of gray scale;gσRepresent gaussian kernel function;I (υ) represents the gray value of the pixel υ in medical image.
Function gσLocalization property vital effect is played in local energy item, its formula is as shown in Equation 7:
The effect of gaussian kernel function is fitting energy function can be confined in certain scope, and this scope is by yardstick
Parameter σ is controlled, and the robustness that larger scale parameter can initialize level set is more preferable, but result may not have use smaller
Parameter it is accurate, the numerical value of general scale parameter is set between 1 and 3.
Regularization term E in energy functionalRIncluding length penalty term and energy penalty term, the effect of length penalty term is to song
The length change of line has effect of contraction so that contour line C keeps as short as possible when gross energy functional reaches minimum.Length is punished
Penalize the formula of item as shown in Equation 8:
It is Heaviside respectively that wherein level set function φ (x, y) zero level collection, which replaces contour line C, H (z) and δ (z),
Function and Dirac functions.H (z) and δ (z) formula is as shown in Equation 9:
The effect of energy penalty term is level set function is kept approximate symbolic measurement in evolutionary process, is avoided
CV models using reinitialize by the way of come slickness and distance function characteristic in keep level set function evolutionary process, and
And amount of calculation can be reduced.Energy penalty term is as shown in Equation 10:
So, the regularization term E in the improved Level Set Models that the embodiment of the present invention proposesRJust by length penalty term and
Energy penalty term is formed, as shown in Equation 11:
ER(φ)=μ L (φ=0)+P (φ) formula 11
In formula 11, μ represents the parameter of control length penalty term, if μ is smaller, level set function can be partitioned into
The less target of area, target on the contrary then larger for area of detection.
By evolution curve C can be replaced by the zero level collection of Lipschitz (Lipchitz) function phi, therefore formula 4
Described in gross energy functional can be further represented as shown in formula 12:
E(c1, c2, f1, f2, φ) and=α ECV(c1, c2, φ) and+β EL(f1, f2, φ) and+ER(φ) formula 12
In formula 12, α and β are to weigh two of global keys and local entity on the occasion of parameter.In actual application, α
It should be set with β numerical value according to the intensity profile situation presented in image.
By evolution level set function φ, image segmentation process is converted to energy functional E minimum process, wherein, water
The EVOLUTION EQUATION of flat collection can be obtained by introducing the calculus of variations.
It is the core concept being improved in the embodiment of the present invention to the CV models of conventional flat set algorithm above, is based on
Level set algorithm in above-described embodiment, above-mentioned steps S3 is described in detail below in conjunction with Fig. 3:
Reference picture 2, step S3 further comprise the steps
Step S301, read original image u0。
Step S302, in original image u0Middle setting initial curve C, initial curve is the initial profile that region growing obtains
Line, the numerical value of following parameter is set:Iterations n, gaussian kernel function parameter σ, the control parameter α of global keys, the control of local entity
Parameter beta, length penalty term parameter μ and control parameter λ processed1And λ2.It should be noted that in above-mentioned steps S4, initial profile line
For the final profile line of upper piece image.
Step S303, evolution level set function φ.
Step S304, whether determined level set function reaches iterations, if reaching iterations, performs step
S305;Otherwise, return to step S303 continues evolution level set function.
Step S305, zero level collection is extracted from level set function φ, as develop obtained final profile line.
Above-mentioned combination Fig. 2 and Fig. 3 respectively illustrates algorithm of region growing and level set according to an embodiment of the invention and calculated
Method, below in conjunction with Fig. 4, the specific of image segmentation is carried out based on algorithm of region growing and level set algorithm to the embodiment of the present invention
Flow chart.
Reference picture 4, it is according to an embodiment of the invention that image segmentation is carried out based on algorithm of region growing and level set algorithm
Flow, including:
Step S401, read image sequence.
Step S402, the algorithm of region growing based on the embodiment of the present invention obtain the initial profile line of first image.This
The scheme of the initial profile line of first image is obtained in inventive embodiments based on algorithm of region growing compared to manually drawing
Mode, susceptibility of the level set function to initial profile curve can be reduced, and then the accuracy of segmentation can be improved.
Step S403, level set movements are carried out based on initial profile line and obtain the final profile line of first image.
Step S404, the initial profile line using the final profile line of a upper image as next image.It is of the invention real
Apply by using initial profile line of the final profile line of a upper image as next image in example, it is both maximized to use
Obtained priori, again improve the Efficiency and accuracy of evolution.
Step S405, developed to obtain the final profile line of next image based on level set function.
Step S406, judges whether the image in image sequence is all handled, if so, then performing step S407;Otherwise, return
Return step S405.
Step S407, read the three-dimensional coordinate of the final profile curve of all images.
Step S408, three-dimensional reconstruction is carried out based on three-dimensional coordinate.
A concrete application example of the invention introduced below:
Split to obtain arteria carotis in the application example of the present invention, such as based on above-mentioned medical image cutting method
Blood vessel, it is illustrated in figure 5 the initial profile line (profile as shown in Figure 5 for the vascular root that application region growth algorithm obtains
Line 502), for initializing the level set function of first image, Fig. 6 be in the result of 200 processed images, with
Four images that machine is chosen, can accurately be partitioned into very much arteria carotis profile (contour line 602 as shown in Figure 6,604,
606 and 608).Fig. 7 shows the topological changeability of level set, can automatically process the crotch of blood vessel and divide (Fig. 7 automatically
Shown in 702,704,706 and 708 be level set movements contour line).Fig. 8 shows the three-dimensional coordinate according to objective contour
The threedimensional model schematic diagram of the arteria carotis of reconstruction.
Fig. 9 diagrammatically illustrates the block diagram of the segmenting device of medical image according to an embodiment of the invention.
Reference picture 9, the segmenting device 900 of medical image according to an embodiment of the invention, including:Acquiring unit 902,
One processing unit 904, second processing unit 906 and cutting unit 908.
Specifically, acquiring unit 902 is used to obtain medical image sequences to be split;First processing units 904 are used for base
First image in the medical image sequences is handled in algorithm of region growing, obtains being used to identify described first
The initial profile curve of the cut zone of image;Second processing unit 906 is used to be based on the initial profile curve, according to level
Set function is handled first image, obtains the final profile curve of first image, and for being directed to institute
Every image in medical image sequences is stated, the initial profile using the final profile curve of previous image as next image
Curve, and every image is handled based on level set function, to obtain the final profile curve of every image;
Cutting unit 908 is split for the final profile curve based on every image to every image.
It should be noted that the detail of each module/unit included in the segmenting device 900 of above-mentioned medical image is
Through being described in detail in the dividing method of corresponding medical image, therefore here is omitted.
Below with reference to Figure 10, it illustrates suitable for for realizing the computer system of the electronic equipment of the embodiment of the present invention
1000 structural representation.The computer system 1000 of electronic equipment shown in Figure 10 is only an example, should not be to the present invention
The function and use range of embodiment bring any restrictions.
As shown in Figure 10, computer system 1000 includes CPU (CPU) 1001, its can according to be stored in only
The journey read the program in memory (ROM) 1002 or be loaded into from storage part 608 in random access storage device (RAM) 1003
Sequence and perform various appropriate actions and processing.In RAM 1003, the various program sums needed for system operatio are also stored with
According to.CPU 1001, ROM 1002 and RAM 1003 are connected with each other by bus 1004.Input/output (I/O) interface 1005
It is connected to bus 1004.
I/O interfaces 1005 are connected to lower component:Importation 1006 including keyboard, mouse etc.;Including such as negative electrode
The output par, c 1007 of ray tube (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.;Storage part including hard disk etc.
1008;And the communications portion 1009 of the NIC including LAN card, modem etc..Communications portion 1009 passes through
Communication process is performed by the network of such as internet.Driver 1010 is also according to needing to be connected to I/O interfaces 1005.It is detachable to be situated between
Matter 1011, such as disk, CD, magneto-optic disk, semiconductor memory etc., it is arranged on as needed on driver 1010, so as to
Storage part 1008 is mounted into as needed in the computer program read from it.
Especially, according to an embodiment of the invention, it may be implemented as computer above with reference to the process of flow chart description
Software program.For example, embodiments of the invention include a kind of computer program product, it includes being carried on computer-readable medium
On computer program, the computer program include be used for execution flow chart shown in method program code.In such reality
To apply in example, the computer program can be downloaded and installed by communications portion 1009 from network, and/or from detachable media
1011 are mounted.When the computer program is performed by CPU (CPU) 1001, perform and limited in the system of the application
Above-mentioned function.
It should be noted that the computer-readable medium shown in the present invention can be computer-readable signal media or meter
Calculation machine readable storage medium storing program for executing either the two any combination.Computer-readable recording medium for example can be --- but not
Be limited to --- electricity, magnetic, optical, electromagnetic, system, device or the device of infrared ray or semiconductor, or it is any more than combination.Meter
The more specifically example of calculation machine readable storage medium storing program for executing can include but is not limited to:Electrical connection with one or more wires, just
Take formula computer disk, hard disk, random access storage device (RAM), read-only storage (ROM), erasable type and may be programmed read-only storage
Device (EPROM or flash memory), optical fiber, portable compact disc read-only storage (CD-ROM), light storage device, magnetic memory device,
Or above-mentioned any appropriate combination.In the present invention, computer-readable recording medium can any include or store journey
The tangible medium of sequence, the program can be commanded the either device use or in connection of execution system, device.And at this
In invention, computer-readable signal media can include in a base band or as carrier wave a part propagation data-signal,
Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including but unlimited
In electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be that computer can
Any computer-readable medium beyond storage medium is read, the computer-readable medium, which can send, propagates or transmit, to be used for
By instruction execution system, device either device use or program in connection.Included on computer-readable medium
Program code can be transmitted with any appropriate medium, be included but is not limited to:Wirelessly, electric wire, optical cable, RF etc., or it is above-mentioned
Any appropriate combination.
Flow chart and block diagram in accompanying drawing, it is illustrated that according to the system of various embodiments of the invention, method and computer journey
Architectural framework in the cards, function and the operation of sequence product.At this point, each square frame in flow chart or block diagram can generation
The part of one module of table, program segment or code, a part for above-mentioned module, program segment or code include one or more
For realizing the executable instruction of defined logic function.It should also be noted that some as replace realization in, institute in square frame
The function of mark can also be with different from the order marked in accompanying drawing generation.For example, two square frames succeedingly represented are actual
On can perform substantially in parallel, they can also be performed in the opposite order sometimes, and this is depending on involved function.Also
It is noted that the combination of each square frame and block diagram in block diagram or flow chart or the square frame in flow chart, can use and perform rule
Fixed function or the special hardware based system of operation are realized, or can use the group of specialized hardware and computer instruction
Close to realize.
Being described in unit involved in the embodiment of the present invention can be realized by way of software, can also be by hard
The mode of part realizes that described unit can also set within a processor.Wherein, the title of these units is in certain situation
Under do not form restriction to the unit in itself.
As on the other hand, present invention also provides a kind of computer-readable medium, the computer-readable medium can be
Included in electronic equipment described in above-described embodiment;Can also be individualism, and without be incorporated the electronic equipment in.
Above computer computer-readable recording medium carries one or more program, and when said one or multiple programs, by one, the electronics is set
During standby execution so that the electronic equipment realizes the dividing method of the medical image as described in above-mentioned embodiment.
Such as, it is possible to achieve as shown in Figure 1:Step S10, obtain medical image sequences to be split;Step S12,
First image in the medical image sequences is handled based on algorithm of region growing, obtains being used to identify described first
Open the initial profile curve of the cut zone of image;Step S14, based on the initial profile curve, according to level set function pair
First image is handled, and obtains the final profile curve of first image;Step S16, for the medical science
Every image in image sequence, the initial profile curve using the final profile curve of previous image as next image,
And every image is handled based on level set function, to obtain the final profile curve of every image;Step
S18, the final profile curve based on every image are split to every image.
And for example, described electronic equipment can realize each step as shown in Fig. 2 to Fig. 4.
It should be noted that although some modules or list of the equipment for action executing are referred in above-detailed
Member, but this division is not enforceable.In fact, according to the embodiment of the present invention, it is above-described two or more
Either the feature of unit and function can embody module in a module or unit.A conversely, above-described mould
Either the feature of unit and function can be further divided into being embodied by multiple modules or unit block.
Through the above description of the embodiments, those skilled in the art is it can be readily appreciated that example described herein is implemented
Mode can be realized by software, can also be realized by way of software combines necessary hardware.Therefore, according to the present invention
The technical scheme of embodiment can be embodied in the form of software product, the software product can be stored in one it is non-volatile
Property storage medium (can be CD-ROM, USB flash disk, mobile hard disk etc.) in or network on, including some instructions are to cause a calculating
Equipment (can be personal computer, server, touch control terminal or network equipment etc.) is performed according to embodiment of the present invention
Method.
Those skilled in the art will readily occur to the present invention its after considering specification and putting into practice invention disclosed herein
Its embodiment.The application be intended to the present invention any modification, purposes or adaptations, these modifications, purposes or
Person's adaptations follow the general principle of the present invention and including undocumented common knowledges in the art of the invention
Or conventional techniques.Description and embodiments are considered only as exemplary, and true scope and spirit of the invention are by following
Claim is pointed out.
It should be appreciated that the invention is not limited in the precision architecture for being described above and being shown in the drawings, and
And various modifications and changes can be being carried out without departing from the scope.The scope of the present invention is only limited by appended claim.
Claims (17)
- A kind of 1. dividing method of medical image, it is characterised in that including:Obtain medical image sequences to be split;First image in the medical image sequences is handled based on algorithm of region growing, obtained described for identifying The initial profile curve of the cut zone of first image;Based on the initial profile curve, first image is handled according to level set function, obtains described first Open the final profile curve of image;For every image in the medical image sequences, using the final profile curve of previous image as next image Initial profile curve, and every image is handled based on level set function, to obtain every image most Whole contour curve;Final profile curve based on every image is split to every image.
- 2. the dividing method of medical image according to claim 1, it is characterised in that based on algorithm of region growing to described First image in medical image sequences is handled, including:Obtain the sub-pixel that user is set in the target area of first image;Storehouse is established, and the sub-pixel is stored in the storehouse;Sub-pixel is taken out from the storehouse successively, wherein, for any sub-pixel taken out from the storehouse, if institute Any neighborhood territory pixel for stating any sub-pixel meets region growing criterion, then using any neighborhood territory pixel as sub-pixel It is stored in the storehouse;It is space-time in the storehouse, the initial profile curve is determined according to the region that all sub-pixels are formed.
- 3. the dividing method of medical image according to claim 2, it is characterised in that also include:According to the gray value of the sub-pixel and the gray value of the neighborhood territory pixel of the sub-pixel, the region growing is calculated The threshold parameter of criterion.
- 4. the dividing method of medical image according to claim 3, it is characterised in that according to the gray scale of the sub-pixel The gray value of the neighborhood territory pixel of value and the sub-pixel, the threshold parameter of the region growing criterion is calculated, including:According to the gray value of the sub-pixel and the gray value of the neighborhood territory pixel of the sub-pixel, the sub-pixel is calculated Gray value and the sub-pixel neighborhood territory pixel gray value between variance;According to the variance, the threshold parameter of the region growing criterion is determined.
- 5. the dividing method of medical image according to claim 2, it is characterised in that also include:Calculate the change between the gray value of the gray value of any sub-pixel and the neighborhood territory pixel of any sub-pixel Change amount;If the gray value of the gray value of any pixel in the neighborhood of any sub-pixel and any sub-pixel it Between variable quantity be less than the threshold parameter of the region growing criterion, it is determined that any pixel meets that the region growing is accurate Then.
- 6. the dividing method of the medical image according to any one of claim 2 to 5, it is characterised in that the neighborhood picture Element is:Four neighborhood territory pixels or eight neighborhood pixel.
- 7. the dividing method of medical image according to claim 1, it is characterised in that based on the level set function and appoint The initial profile curve of one image, any image is handled, including:The initial profile curve is set in any image;Based on the initial profile curve and the evolution parameter set, the level set function is developed;When iterations when the level set function is developed reaches pre-determined number, extracted from the level set function Zero level collection, using the final profile curve as any image.
- 8. the dividing method of medical image according to claim 7, it is characterised in that based on the initial profile curve and The evolution parameter of setting, the level set function is developed, including:Based on the initial profile curve, energy functional model is built;Initial profile curve in the energy functional model is replaced by level set function;Evolution parameter based on setting, equation corresponding to the energy functional model is solved, with to the level set letter Number is developed.
- 9. the dividing method of medical image according to claim 8, it is characterised in that the energy functional model is as follows:E=α ECV+β·EL+ERWherein, E represents the energy functional model;ECVRepresent the energy function of CV models;ELRepresent the energy function of local entity; ERRepresent the energy function of regularization term;α and β represents control parameter.
- 10. the dividing method of medical image according to claim 9, it is characterised in that described in being calculated by equation below The energy function of CV models:ECV(C, c1, c2)=λ1∫outside(C)|I(x)-c1|2dx+λ2∫inside(C)|I(x)-c2|2dx+v|C|Wherein, C represents the initial profile curve;Outside (C) and inside (C) represent the initial profile curve C respectively Exterior domain and inner region;c1And c2Represent constant;λ1、λ2Control parameter is represented with v;I (x) is represented in any image Pixel x gray value.
- 11. the dividing method of medical image according to claim 9, it is characterised in that described in being calculated by equation below The energy function of local entity:EL=∫Ωεx(C, f1, f2)dxWherein, εx(C, f1, f2)=λ1∫inside(C)gσ(x-y)(I(v)-f1(x))2dy+λ2∫outside(C)gσ(x-y)(I(v)-f2 (x))2dy;λ1And λ2Represent control parameter;f1And f (x)2(x) it is near for the ideal of initial profile curve C both sides topography gray scale Like value;gσRepresent gaussian kernel function;I (v) represents the gray value of the pixel υ in any image.
- 12. the dividing method of medical image according to claim 11, it is characterised in that described in being calculated by below equation Gaussian kernel function:<mrow> <msub> <mi>g</mi> <mi>&sigma;</mi> </msub> <mrow> <mo>(</mo> <mi>u</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mn>1</mn> <mrow> <msup> <mrow> <mo>(</mo> <mn>2</mn> <mi>&pi;</mi> <mo>)</mo> </mrow> <mrow> <mi>n</mi> <mo>/</mo> <mn>2</mn> </mrow> </msup> <msup> <mi>&sigma;</mi> <mi>n</mi> </msup> </mrow> </mfrac> <msup> <mi>e</mi> <mrow> <mo>-</mo> <mo>|</mo> <mi>u</mi> <msup> <mo>|</mo> <mn>2</mn> </msup> <mo>/</mo> <mn>2</mn> <msup> <mi>&sigma;</mi> <mn>2</mn> </msup> </mrow> </msup> </mrow>Wherein, u represents constant;σ represents scale parameter;N represents iterations.
- 13. the dividing method of medical image according to claim 9, it is characterised in that described in being calculated by equation below The energy function of regularization term:ER(φ)=μ L (φ=0)+P (φ)Wherein, φ represents level set function;μ represents control parameter;L (φ=0) represents length penalty term,H (z) and δ (z) are represented respectively Heaviside functions and Dirac functions;P (φ) represents energy penalty term,
- 14. the dividing method of the medical image according to any one of claim 1 to 13, it is characterised in that also include:Based on the segmentation result to every image in the medical image sequences, the three-dimensional modeling of medical image is carried out.
- A kind of 15. segmenting device of medical image, it is characterised in that including:Acquiring unit, for obtaining medical image sequences to be split;First processing units, at based on algorithm of region growing to first image in the medical image sequences Reason, obtains the initial profile curve of the cut zone for identifying first image;Second processing unit, for based on the initial profile curve, being carried out according to level set function to first image Processing, the final profile curve of first image, and every image for being directed in the medical image sequences are obtained, Initial profile curve using the final profile curve of previous image as next image, and based on level set function to described Every image is handled, to obtain the final profile curve of every image;Cutting unit, every image is split for the final profile curve based on every image.
- 16. a kind of computer-readable medium, is stored thereon with computer program, it is characterised in that described program is held by processor The dividing method of the medical image as any one of claim 1 to 14 is realized during row.
- 17. a kind of electronic equipment, it is characterised in that including:One or more processors;Storage device, for storing one or more programs, when one or more of programs are by one or more of processing When device performs so that one or more of processors realize the medical image as any one of claim 1 to 14 Dividing method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710733452.4A CN107481252A (en) | 2017-08-24 | 2017-08-24 | Dividing method, device, medium and the electronic equipment of medical image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710733452.4A CN107481252A (en) | 2017-08-24 | 2017-08-24 | Dividing method, device, medium and the electronic equipment of medical image |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107481252A true CN107481252A (en) | 2017-12-15 |
Family
ID=60602512
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710733452.4A Pending CN107481252A (en) | 2017-08-24 | 2017-08-24 | Dividing method, device, medium and the electronic equipment of medical image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107481252A (en) |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108447066A (en) * | 2018-03-19 | 2018-08-24 | 青岛海信医疗设备股份有限公司 | Biliary tract image partition method, terminal, storage medium |
CN108596935A (en) * | 2018-04-24 | 2018-09-28 | 安徽锐捷信息科技有限公司 | A kind of dividing method and device of magnetic resonance image |
CN108830859A (en) * | 2018-04-13 | 2018-11-16 | 中国科学院深圳先进技术研究院 | Based on film dividing method middle in ultrasound image, device, equipment and storage medium |
CN108961278A (en) * | 2018-06-20 | 2018-12-07 | 深圳市旭东数字医学影像技术有限公司 | The method and its system of abdominal wall muscle segmentation based on image data |
CN109740600A (en) * | 2019-01-04 | 2019-05-10 | 上海联影医疗科技有限公司 | Localization method, device, computer equipment and the storage medium of highlighted focal area |
CN109785293A (en) * | 2018-12-22 | 2019-05-21 | 昆明理工大学 | A method of the ultrasound image lesion based on global and local movable contour model is divided automatically |
CN110390667A (en) * | 2019-06-18 | 2019-10-29 | 平安科技(深圳)有限公司 | Lesion extracting method, device, equipment and storage medium based on eyeground OCT image |
CN110705576A (en) * | 2019-09-29 | 2020-01-17 | 慧影医疗科技(北京)有限公司 | Region contour determining method and device and image display equipment |
CN110992370A (en) * | 2019-11-19 | 2020-04-10 | 东软医疗***股份有限公司 | Pancreas tissue segmentation method and device and terminal equipment |
CN111080642A (en) * | 2019-12-31 | 2020-04-28 | 北京推想科技有限公司 | Tissue typing method and device based on medical image and electronic equipment |
CN111105400A (en) * | 2019-12-20 | 2020-05-05 | 中国医学科学院阜外医院 | Valve image processing method, valve image processing device and electronic equipment |
CN111161289A (en) * | 2019-12-26 | 2020-05-15 | 杭州格像科技有限公司 | Method, device and computer program product for improving contour precision of object in image |
CN111429351A (en) * | 2020-03-24 | 2020-07-17 | 珠海嘉润医用影像科技有限公司 | Image interpolation processing method and system |
CN111784716A (en) * | 2020-06-04 | 2020-10-16 | 华中科技大学 | Sequence diagram image segmentation method and system based on ultrasonic CT |
CN113034508A (en) * | 2019-12-25 | 2021-06-25 | 飞依诺科技(苏州)有限公司 | Ultrasonic image focus segmentation method and device and computer equipment |
CN115171204A (en) * | 2022-09-06 | 2022-10-11 | 北京鹰瞳科技发展股份有限公司 | Method for training prediction model for predicting retinal age and related product |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104504708A (en) * | 2014-12-26 | 2015-04-08 | 大连理工大学 | DSA (digital subtraction angiography) cerebrovascular image auto-segmenting method based on adjacent image feature point sets |
CN104599270A (en) * | 2015-01-18 | 2015-05-06 | 北京工业大学 | Breast neoplasms ultrasonic image segmentation method based on improved level set algorithm |
CN107016683A (en) * | 2017-04-07 | 2017-08-04 | 衢州学院 | The level set hippocampus image partition method initialized based on region growing |
-
2017
- 2017-08-24 CN CN201710733452.4A patent/CN107481252A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104504708A (en) * | 2014-12-26 | 2015-04-08 | 大连理工大学 | DSA (digital subtraction angiography) cerebrovascular image auto-segmenting method based on adjacent image feature point sets |
CN104599270A (en) * | 2015-01-18 | 2015-05-06 | 北京工业大学 | Breast neoplasms ultrasonic image segmentation method based on improved level set algorithm |
CN107016683A (en) * | 2017-04-07 | 2017-08-04 | 衢州学院 | The level set hippocampus image partition method initialized based on region growing |
Non-Patent Citations (3)
Title |
---|
于广婷,等: "基于改进水平集的人脑海马图像分割方法", 《计算机工程》 * |
吴珊: "基于活动轮廓模型的肝脏分割算法研究", 《中国优秀硕士学位论文全文数据库(电子期刊)信息科技辑》 * |
王文峰: "腹部CT图像的三维重构与手术规划", 《中国优秀硕士学位论文全文数据库(电子期刊)信息科技辑》 * |
Cited By (28)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108447066A (en) * | 2018-03-19 | 2018-08-24 | 青岛海信医疗设备股份有限公司 | Biliary tract image partition method, terminal, storage medium |
CN108447066B (en) * | 2018-03-19 | 2021-03-09 | 青岛海信医疗设备股份有限公司 | Biliary tract image segmentation method, terminal and storage medium |
CN108830859A (en) * | 2018-04-13 | 2018-11-16 | 中国科学院深圳先进技术研究院 | Based on film dividing method middle in ultrasound image, device, equipment and storage medium |
CN108830859B (en) * | 2018-04-13 | 2022-03-04 | 中国科学院深圳先进技术研究院 | Ultrasound image-based intima-media membrane segmentation method, device, equipment and storage medium |
CN108596935A (en) * | 2018-04-24 | 2018-09-28 | 安徽锐捷信息科技有限公司 | A kind of dividing method and device of magnetic resonance image |
CN108961278B (en) * | 2018-06-20 | 2023-01-31 | 深圳市旭东数字医学影像技术有限公司 | Method and system for abdominal wall muscle segmentation based on image data |
CN108961278A (en) * | 2018-06-20 | 2018-12-07 | 深圳市旭东数字医学影像技术有限公司 | The method and its system of abdominal wall muscle segmentation based on image data |
CN109785293A (en) * | 2018-12-22 | 2019-05-21 | 昆明理工大学 | A method of the ultrasound image lesion based on global and local movable contour model is divided automatically |
CN109785293B (en) * | 2018-12-22 | 2022-09-27 | 昆明理工大学 | Ultrasonic image focus segmentation method based on global and local active contour models |
CN109740600A (en) * | 2019-01-04 | 2019-05-10 | 上海联影医疗科技有限公司 | Localization method, device, computer equipment and the storage medium of highlighted focal area |
CN110390667A (en) * | 2019-06-18 | 2019-10-29 | 平安科技(深圳)有限公司 | Lesion extracting method, device, equipment and storage medium based on eyeground OCT image |
CN110390667B (en) * | 2019-06-18 | 2023-10-20 | 平安科技(深圳)有限公司 | Focus extraction method, device, equipment and storage medium based on fundus OCT image |
WO2020252900A1 (en) * | 2019-06-18 | 2020-12-24 | 平安科技(深圳)有限公司 | Fundus oct picture-based lesion extraction method, apparatus and device, and storage medium |
CN110705576A (en) * | 2019-09-29 | 2020-01-17 | 慧影医疗科技(北京)有限公司 | Region contour determining method and device and image display equipment |
CN110705576B (en) * | 2019-09-29 | 2020-09-22 | 慧影医疗科技(北京)有限公司 | Region contour determining method and device and image display equipment |
CN110992370A (en) * | 2019-11-19 | 2020-04-10 | 东软医疗***股份有限公司 | Pancreas tissue segmentation method and device and terminal equipment |
CN110992370B (en) * | 2019-11-19 | 2023-07-04 | 东软医疗***股份有限公司 | Pancreas tissue segmentation method and device and terminal equipment |
CN111105400A (en) * | 2019-12-20 | 2020-05-05 | 中国医学科学院阜外医院 | Valve image processing method, valve image processing device and electronic equipment |
CN111105400B (en) * | 2019-12-20 | 2023-09-19 | 中国医学科学院阜外医院 | Valve image processing method, valve image processing device and electronic equipment |
CN113034508A (en) * | 2019-12-25 | 2021-06-25 | 飞依诺科技(苏州)有限公司 | Ultrasonic image focus segmentation method and device and computer equipment |
CN113034508B (en) * | 2019-12-25 | 2024-02-13 | 飞依诺科技股份有限公司 | Method and device for segmenting ultrasonic image focus and computer equipment |
CN111161289A (en) * | 2019-12-26 | 2020-05-15 | 杭州格像科技有限公司 | Method, device and computer program product for improving contour precision of object in image |
CN111161289B (en) * | 2019-12-26 | 2023-08-08 | 杭州格像科技有限公司 | Method and device for improving contour precision of object in image |
CN111080642A (en) * | 2019-12-31 | 2020-04-28 | 北京推想科技有限公司 | Tissue typing method and device based on medical image and electronic equipment |
CN111429351A (en) * | 2020-03-24 | 2020-07-17 | 珠海嘉润医用影像科技有限公司 | Image interpolation processing method and system |
CN111784716A (en) * | 2020-06-04 | 2020-10-16 | 华中科技大学 | Sequence diagram image segmentation method and system based on ultrasonic CT |
CN115171204A (en) * | 2022-09-06 | 2022-10-11 | 北京鹰瞳科技发展股份有限公司 | Method for training prediction model for predicting retinal age and related product |
CN115171204B (en) * | 2022-09-06 | 2023-02-21 | 北京鹰瞳科技发展股份有限公司 | Method for training prediction model for predicting retinal age and related product |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107481252A (en) | Dividing method, device, medium and the electronic equipment of medical image | |
CN109242844B (en) | Pancreatic cancer tumor automatic identification system based on deep learning, computer equipment and storage medium | |
CN109409503B (en) | Neural network training method, image conversion method, device, equipment and medium | |
CN106887000B (en) | Gridding processing method and system for medical image | |
CN111047594A (en) | Tumor MRI weak supervised learning analysis modeling method and model thereof | |
CN109166133A (en) | Soft tissue organs image partition method based on critical point detection and deep learning | |
CN107481254A (en) | Processing method, device, medium and the electronic equipment of medical image | |
Nasor et al. | Detection and Localization of Early‐Stage Multiple Brain Tumors Using a Hybrid Technique of Patch‐Based Processing, k‐means Clustering and Object Counting | |
CN107680107B (en) | Automatic segmentation method of diffusion tensor magnetic resonance image based on multiple maps | |
CN112508965A (en) | Automatic contour line drawing system for normal organs in medical image | |
CN107004268A (en) | The automatic region of interest regional partition and registration of the Dynamic constrasted enhancement image of colorectal carcinoma | |
CN105224764B (en) | Bone modeling and simulation method | |
CN109508787A (en) | Neural network model training method and system for ultrasound displacement estimation | |
CN107507189A (en) | Mouse CT image kidney dividing methods based on random forest and statistical model | |
CN116309571B (en) | Three-dimensional cerebrovascular segmentation method and device based on semi-supervised learning | |
WO2020057074A1 (en) | Model training method and device for plaque segmentation, apparatus, and storage medium | |
CN111815593B (en) | Pulmonary nodule domain adaptive segmentation method, device and storage medium based on countermeasure learning | |
CN109300147A (en) | The method and device of blood vessel two-dimensional center line and three-dimensional center line registration | |
CN112927799A (en) | Life cycle analysis system fusing multi-example learning and multi-task depth imaging group | |
CN112017161A (en) | Pulmonary nodule detection method and device based on central point regression | |
JP2024515635A (en) | System and method for reconstructing 3D images from ultrasound and camera images - Patents.com | |
CN113160120A (en) | Liver blood vessel segmentation method and system based on multi-mode fusion and deep learning | |
CN110458842B (en) | Brain tumor segmentation method based on two-channel three-dimensional dense connection network | |
CN110599444A (en) | Device, system and non-transitory readable storage medium for predicting fractional flow reserve of a vessel tree | |
CN111986216B (en) | RSG liver CT image interactive segmentation algorithm based on neural network improvement |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20171215 |