CN108109149A - A kind of coronary artery OCT image automatic division method - Google Patents
A kind of coronary artery OCT image automatic division method Download PDFInfo
- Publication number
- CN108109149A CN108109149A CN201711338640.3A CN201711338640A CN108109149A CN 108109149 A CN108109149 A CN 108109149A CN 201711338640 A CN201711338640 A CN 201711338640A CN 108109149 A CN108109149 A CN 108109149A
- Authority
- CN
- China
- Prior art keywords
- image
- pixel
- coronary artery
- division method
- obtains
- 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
- 238000000034 method Methods 0.000 title claims abstract description 37
- 210000004351 coronary vessel Anatomy 0.000 title claims abstract description 22
- 230000006870 function Effects 0.000 claims abstract description 27
- 230000011218 segmentation Effects 0.000 claims abstract description 24
- 230000000694 effects Effects 0.000 claims abstract description 5
- 230000002093 peripheral effect Effects 0.000 claims abstract description 5
- 238000013507 mapping Methods 0.000 claims description 9
- 206010016654 Fibrosis Diseases 0.000 claims description 8
- 230000004761 fibrosis Effects 0.000 claims description 8
- 150000002632 lipids Chemical class 0.000 claims description 7
- 238000009499 grossing Methods 0.000 claims description 6
- 239000000203 mixture Substances 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 2
- 238000007796 conventional method Methods 0.000 abstract description 2
- 238000002474 experimental method Methods 0.000 abstract description 2
- 238000012014 optical coherence tomography Methods 0.000 description 24
- 206010003211 Arteriosclerosis coronary artery Diseases 0.000 description 2
- 208000029078 coronary artery disease Diseases 0.000 description 2
- 208000026758 coronary atherosclerosis Diseases 0.000 description 2
- 230000002708 enhancing effect Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000000747 cardiac effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000015654 memory Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration by the use of more than one image, e.g. averaging, subtraction
-
- G06T5/70—
-
- G06T5/77—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
Abstract
The invention discloses a kind of coronary artery OCT image automatic division method, step includes：The pretreatment of center oblique line, conduit annulus and peripheral label character is removed to OCT image, obtains pretreated image；Binaryzation simultaneously further obtains endangium image, is then superimposed with gained pretreatment image, obtains denoising image；Adjust the weight of binary image and with denoising image superposition, the superimposed image of high RST region energy must be enhanced；Superimposed image is clustered using k means methods, then transforms to higher dimensional space using radial basis function, then builds GraphCuts energy functions and is further processed, obtains segmentation result.The automated regional segmentation for being related to coronary artery region OCT image automatic division method and can preferably realizing three kinds of patch regions of the present invention, segmentation effect is substantially better than conventional method, image patch region can preferably be preserved, segmentation loss is smaller, and found by many experiments, context of methods has stronger robustness.
Description
Technical field
The present invention relates to medical image dividing methods, specifically a kind of coronary artery OCT image automatic division method.
Background technology
Medical image is a part particularly important in image procossing, in particular with optical interference fault imaging
The appearance of (optical coherence tomography, OCT) technology so that directly observe the shape of coronary atherosclerosis
State is possibly realized.However, accurate understand of OCT image is needed to spend doctor's substantial amounts of time, and different doctor's individuals
Between have stronger subjectivity and individual difference, therefore in Computer Automatic Recognition image heterogeneity become key.
On coronary artery OCT image atherosis region, there are mainly three types of types.Fibrosis patch is that coronary artery is hard
Most common a kind of patch in change, shows as homogeneity, high RST and the region of underdamp；Lipid Plaque show as edge blurry,
High back reflection and strong attenuation region have the fibrous cap of high RST band on the surface of low-signal areas, are the masters that coronary artery is caused to rupture
Want one of factor；Calcified plaque shows as the low signal or non-uniform areas of clear-cut margin, and cardiac stent, which is placed, to be influenced greatly.
Therefore, accurately splitting plaque location and size is and its important.
Main method currently for OCT Image Segmentations be using select come target area, use k-means methods
It is clustered to obtain one of which patch region or obtains patch region using level set gradient method；Or use A-line models
OCT image is analyzed, so that it is determined that patch region.Both approaches or can only identify wherein a certain patch or
It is not high to patch segmentation precision, it is impossible to reach necessary requirement.
The content of the invention
The object of the present invention is to provide a kind of automatic division method in patch region in coronary artery OCT image, to realize
Three kinds of patch regions simultaneously and high-precision automatic segmentation, solve existing automatic Segmentation plaque type on coronary artery
The problem of single or segmentation precision is low.
The purpose of the present invention is according to following technical solution realizations：
A kind of coronary artery OCT image automatic division method, includes the following steps：
A) using one group of coronary artery OCT original image comprising fibrosis patch, calcified plaque and Lipid Plaque as place
Object is managed, the pretreatment of center oblique line, conduit annulus and peripheral label character is removed to OCT original images, obtains pre- place
Image after reason；
B the pretreated image) is changed into polar form, binaryzation is then carried out, obtains binary image；
Minimum value is obtained to the intra-class variance of gained binary image, then extracts each row first of the binary image
One continuous straight line of a non-zero pixels and composition, is then converted to rectangular coordinate system form, obtains endangium image；
Gained endangium image is inserted into A) in pretreated image obtained by step, then by inner membrance interior zone pixel
Value is arranged to 0, obtains the denoising image of removal inner membrance interior noise；
C B) is adjusted) the binaryzation weight of binary image described in step is 0.3, be then added to B) denoising image obtained by step
In, obtain the image of the superposition of high RST region energy enhancing；
D the image of the superposition) is divided into different clusters according to distribution of color using k-means methods, and using European
Distance is as minimum function with definite cluster centre；
E) using radial basis function by D) image after cluster obtained by step transforms to higher dimensional space；
F) build GraphCuts energy functions to through E) step processing image handle, by minimizing the energy function
Reach the allocation optimum of each pixel, realize the purpose for splitting calcified plaque, fibrosis patch and Lipid Plaque,
Obtain segmentation result.
The coronary artery OCT image automatic division method, D) step detailed process is：Use k-means methodsThe image of gained superposition is divided into different clusters according to distribution of color, and uses Europe
Formula distance, which is used as, minimizes function to determine cluster centre C, wherein：The number of regions that K is partitioned into when being cluster sets K=4；l
For the label of pixel, value is 1~K, xlIt is the pixel of l clusters；I is for pixel dimension, SiIt is the pixel in each cluster
Set, μiRepresent SiIn definite barycenter, i.e. cluster centre C.
The coronary artery OCT image automatic division method, E) radial basis function formula in stepWherein, β is the width parameter of radial basis function, for control function radial effect model
It encloses, takes β=0.5；ImFor the pixel set of image.
The coronary artery OCT image automatic division method, F) in step, GraphCuts energy function H (f)=D (f)+
α R (f), wherein,
1. D (f) is the data item for calculating color similarity in cut zone, using the non-Euclidean distance in kernel spaces
As data item, then data item D (f) is evolved into：
Wherein：X and l is respectively the pixel of image and corresponding label, SlIt is cluster set and Label space with L；
For kernel Nonlinear Mappings, ulFor the model parameter of piecewise constant model, mxFor the color vector of image；
2. R (f) is the regular terms for making edge smoothing, using the dissimilar degree using point of proximity and central point as canonical
Item is evolved into as regular terms, then regular terms R (f)：R (f)=∑{x,y}∈NR (f (x), f (y)),
Wherein, { x, y } is a pair of of point of proximity in point set N；R (f (x), f (y))=min (A2, | uf(x)-uf(y)
|2), A is the gray-scale intensity at point set N centers, and f (x), f (y) they are respectively x, y point to the mapping of Label space, uf(x)、uf(y)For
The respectively gray-scale intensity of x, y point；
3. f represents mapping of the pixel to Label space；
4. α is weight coefficient, for controlling edge smoothing degree, α=0.6 is set.
The present invention's is related to coronary artery region OCT image automatic division method and can preferably realize three kinds of patch regions
Automated regional segmentation, segmentation effect is substantially better than conventional method, can preferably preserve image patch region, segmentation loss compared with
It is small, and found by many experiments, context of methods has stronger robustness.
Attached drawing table explanation
Fig. 1 is the flowage structure figure of technical solution of the present invention.
Fig. 2 is OCT original images.
Fig. 3 is OCT original images image after pretreatment.
The endangium image that Fig. 4 is extracted.
Fig. 5 is removal inner membrance internal noise and enhances the image of the superposition of high RST region energy.
Fig. 6 is segmentation result.
Specific embodiment
Embodiment 1：
Test herein using Intel (R) Core (TM) i5-4460 CPU@3.2GHz processors/4GB memories/
It realizes in the computer of 64 bit manipulation systems of Windows7, is operated with 2014 execution of instrument of software Matlab.
The implementation process of the present invention is as shown in Figure 1：
A coronary artery region OCT original images) are obtained, are then pre-processed, preprocessing process includes：Removal is fixed
Center oblique line, conduit annulus and the peripheral label character of position：
1. OCT original images gather：It is acquired using optical coherence tomography scanner (ST.JUDE MEDICAL C7),
One frame of random selection includes fibrosis patch, calcified plaque from 9 coronary atherosclerosis patients (patient is 8 male 1 female)
OCT original images with Lipid Plaque region are as dealing with objects, as shown in Figure 2.
2. using the straight line in Hough transform method detection OCT original images and removing center oblique line, then schemed using OCT
As the characteristic of the conduit annulus of fixed position and size, peripheral label character, conduit annulus and label character are removed, obtained pre-
Treated image, such as Fig. 3.
B pretreated image) is changed into polar form, then using the Otsu algorithms of automatic threshold by image two
Value obtains binary image；
By the way that minimum value is obtained to the intra-class variance in binary image, foreground and background is distinguished, is then extracted each
The pixel of first non-zero of row and the continuous straight line of composition one, finally will be under polar coordinates to get the inner membrance under polar coordinates
Film image is transformed under rectangular coordinate system, so as to extract endangium image, as shown in Figure 4；
Gained endangium image is inserted into A) in pretreated image obtained by step, then by inner membrance interior zone pixel
Value is arranged to 0, obtains the denoising image of removal inner membrance interior noise；
C B) is adjusted) the binaryzation weight of binary image obtained by step is 0.3, be then added to B) denoising figure obtained by step
As in, the image that is superimposed, as shown in figure 5, realizing the purpose of the energy in high RST region in enhancing denoising image.
D the image that gained is superimposed) is divided into different clusters according to distribution of color using k-means methods, and uses Europe
Formula distance is as minimum function with definite cluster centre C：
Wherein：The number of regions that K is partitioned into when being cluster sets K=4；L is the label of pixel, and value is 1~K, xl
It is the pixel of l clusters；I is for pixel dimension, SiIt is the pixel set in each cluster, μiRepresent SiIn definite barycenter,
That is cluster centre C.
E) by radial basis function formula by D) image after step cluster transforms to higher dimensional space：
Wherein, β is the width parameter of radial basis function, for control function radial effect scope, takes β=0.5；ImFor figure
The pixel set of picture.
F GraphCuts energy functions) are built, reach each pixel (label) by minimizing the energy function
Allocation optimum realizes the purpose for splitting calcified plaque, fibrosis patch and Lipid Plaque, obtains segmentation result, such as Fig. 6
It is shown：
In this step, GraphCuts energy functions are made of two parts, first, calculating color similarity in cut zone
Data item D (f) the other is making the regular terms R (f) of edge smoothing, is represented by：
H (f)=D (f)+α R (f)
Wherein：
1. f represents mapping of the pixel to Label space (space that i.e. label is formed)；
2. α is weight coefficient, for controlling edge smoothing degree, α=0.6 is set；
3. using the non-Euclidean distance in kernel spaces as data item, then data item D (f) is evolved into：
Wherein：X and l is respectively the pixel of image and corresponding label, SlIt is cluster set and Label space with L；
For kernel Nonlinear Mappings, ulFor the model parameter of piecewise constant model, mxFor the color vector of image；
4. using the dissimilar degree using point of proximity and central point as regular terms as regular terms, then regular terms R (f)
It is evolved into：
R (f)=∑{x,y}∈NR (f (x), f (y)),
Wherein, { x, y } is a pair of of point of proximity in point set N；R (f (x), f (y))=min (A2, | uf(x)-uf(y)
|2), A is the gray-scale intensity at point set N centers, and f (x), f (y) they are respectively x, y point to the mapping of Label space, uf(x)、uf(y)For
The respectively gray-scale intensity of x, y point.
Patch segmentation precision is evaluated：
Using the patch region of doctor's manual segmentation as " goldstandard ", using the accuracy of overlapping area formula=(A represents this method segmentation result, and B represents doctor's goldstandard) evaluates patch segmentation precision, as a result such as 1 institute of table
Show.
Table 1：Segmentation accuracy
(precision is about 70%~80%), this method compared with other level-set segmentation methods and edge detection dividing method
It can not only identify 3 kinds of regions, and reach higher segmentation level, particularly the segmentation precision of fibrosis patch is reached
To 90%, segmentation result meets doctor's requirement.
Claims (4)
1. a kind of coronary artery OCT image automatic division method, which is characterized in that include the following steps：
A it is right) using the coronary artery OCT original images comprising fibrosis patch, calcified plaque and Lipid Plaque as process object
OCT original images are removed the pretreatment of center oblique line, conduit annulus and peripheral label character, obtain pretreated figure
Picture；
B the pretreated image) is changed into polar form, binaryzation is then carried out, obtains binary image；
Minimum value is obtained to the intra-class variance of gained binary image, each row first for then extracting the binary image are non-
One continuous straight line of zero pixel and composition, is then converted to rectangular coordinate system form, obtains endangium image；
Gained endangium image is inserted into A) in pretreated image obtained by step, then inner membrance interior zone pixel value is set
0 is set to, obtains the denoising image of removal inner membrance interior noise；
C B) is adjusted) the binaryzation weight of binary image described in step is 0.3, be then added to B) in denoising image obtained by step, obtain
The image of the superposition enhanced to high RST region energy；
D the image of the superposition) is divided into different clusters according to distribution of color using k-means methods, and uses Euclidean distance
As minimum function with definite cluster centre；
E) using radial basis function by D) image after cluster obtained by step transforms to higher dimensional space；
F) build GraphCuts energy functions to through E) step processing image handle, reached by minimizing the energy function
The allocation optimum of each pixel is realized the purpose for splitting calcified plaque, fibrosis patch and Lipid Plaque, is obtained
Segmentation result.
2. coronary artery OCT image automatic division method according to claim 1, which is characterized in that
D) step detailed process is：Use k-means methodsThe image that gained is superimposed
It is divided into different clusters according to distribution of color, and uses Euclidean distance as minimum function to determine cluster centre C, wherein：K
For the number of regions being partitioned into during cluster, K=4 is set；L is the label of pixel, and value is 1~K, xlIt is the picture of l clusters
Element；I is for pixel dimension, SiIt is the pixel set in each cluster, μiRepresent SiIn definite barycenter, i.e. cluster centre C.
3. coronary artery OCT image automatic division method according to claim 2, which is characterized in that
E) radial basis function formula in stepWherein, β is the width parameter of radial basis function,
For control function radial effect scope, β=0.5 is taken；ImFor the pixel set of image.
4. coronary artery OCT image automatic division method according to claim 3, which is characterized in that
F) in step, GraphCuts energy function H (f)=D (f)+α R (f), wherein,
1. D (f) is the data item for calculating color similarity in cut zone, using the non-Euclidean distance conduct in kernel spaces
Data item, then data item D (f) be evolved into：
Wherein：X and l is respectively the pixel of image and corresponding label, SlIt is cluster set and Label space with L；For
Kernel Nonlinear Mappings, ulFor the model parameter of piecewise constant model, mxFor the color vector of image；
2. R (f) is the regular terms for making edge smoothing, made using the dissimilar degree using point of proximity and central point as regular terms
For regular terms, then regular terms R (f) is evolved into：R (f)=∑{ x, y } ∈ NR (f (x), f (y)),
Wherein, { x, y } is a pair of of point of proximity in point set N；R (f (x), f (y))=min (A2, | uf(x)-uf(y)|2), A is
The gray-scale intensity at point set N centers, f (x), f (y) are respectively x, y point to the mapping of Label space, uf(x)、uf(y)For be respectively x,
The gray-scale intensity of y points；
3. f represents mapping of the pixel to Label space；
4. α is weight coefficient, for controlling edge smoothing degree, α=0.6 is set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711338640.3A CN108109149A (en) | 2017-12-14 | 2017-12-14 | A kind of coronary artery OCT image automatic division method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711338640.3A CN108109149A (en) | 2017-12-14 | 2017-12-14 | A kind of coronary artery OCT image automatic division method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108109149A true CN108109149A (en) | 2018-06-01 |
Family
ID=62215940
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711338640.3A Pending CN108109149A (en) | 2017-12-14 | 2017-12-14 | A kind of coronary artery OCT image automatic division method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108109149A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109325948A (en) * | 2018-10-09 | 2019-02-12 | 数坤（北京）网络科技有限公司 | A kind of coronary artery dividing method and device based on special area optimization |
| CN109389606A (en) * | 2018-09-30 | 2019-02-26 | 数坤（北京）网络科技有限公司 | A kind of coronary artery dividing method and device |
| CN110942447A (en) * | 2019-10-18 | 2020-03-31 | 平安科技（深圳）有限公司 | OCT image segmentation method, device, equipment and storage medium |
| CN113083724A (en) * | 2021-03-24 | 2021-07-09 | 北京霍里思特科技有限公司 | Ore identification method, detection mechanism and mineral product sorting machine |
| CN113096115A (en) * | 2021-04-28 | 2021-07-09 | 博动医学影像科技（上海）有限公司 | Coronary artery plaque state evaluation method and device and electronic equipment |
| CN113610810A (en) * | 2021-08-09 | 2021-11-05 | 华力创科学(深圳)有限公司 | Blood vessel detection method based on Markov random field |
| CN114820678A (en) * | 2022-06-27 | 2022-07-29 | 天津恒宇医疗科技有限公司 | Automatic extraction method and system for inner contour of blood vessel wall based on OCT image |
| CN114882017A (en) * | 2022-06-30 | 2022-08-09 | 中国科学院大学 | Method and device for detecting thin fiber cap plaque based on intracranial artery image |
| WO2022236995A1 (en) * | 2021-05-12 | 2022-11-17 | 深圳市中科微光医疗器械技术有限公司 | Guided detection and scoring method for calcified plaque, and device and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102800088A (en) * | 2012-06-28 | 2012-11-28 | 华中科技大学 | Automatic dividing method of ultrasound carotid artery plaque |
| CN104978725A (en) * | 2014-04-03 | 2015-10-14 | 上海联影医疗科技有限公司 | Method and device for dividing coronary artery |
| CN105741299A (en) * | 2016-02-02 | 2016-07-06 | 河北大学 | Coronary artery CT angiography image segmentation method |
| CN106780495A (en) * | 2017-02-15 | 2017-05-31 | 深圳市中科微光医疗器械技术有限公司 | Cardiovascular implantation support automatic detection and appraisal procedure and system based on OCT |
| US20170193658A1 (en) * | 2015-12-31 | 2017-07-06 | Acist Medical Systems, Inc. | Semi-automated image segmentation system and method |
-
2017
- 2017-12-14 CN CN201711338640.3A patent/CN108109149A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102800088A (en) * | 2012-06-28 | 2012-11-28 | 华中科技大学 | Automatic dividing method of ultrasound carotid artery plaque |
| CN104978725A (en) * | 2014-04-03 | 2015-10-14 | 上海联影医疗科技有限公司 | Method and device for dividing coronary artery |
| US20170193658A1 (en) * | 2015-12-31 | 2017-07-06 | Acist Medical Systems, Inc. | Semi-automated image segmentation system and method |
| CN105741299A (en) * | 2016-02-02 | 2016-07-06 | 河北大学 | Coronary artery CT angiography image segmentation method |
| CN106780495A (en) * | 2017-02-15 | 2017-05-31 | 深圳市中科微光医疗器械技术有限公司 | Cardiovascular implantation support automatic detection and appraisal procedure and system based on OCT |
Non-Patent Citations (3)
| Title |
|---|
| LAMBROS S. ATHANASIOU 等: "Fully Automated Calcium Detection Using Optical Coherence Tomography", 《2013 35TH ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY (EMBC)》 * |
| 张勃 等: "基于核图割算法的冠脉光学相干断层图像斑块区域分割", 《生物医学工程学杂志》 * |
| 林旭云 等: "线粒体电镜图像二值化增强算法研究", 《中原工学院学报》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109389606A (en) * | 2018-09-30 | 2019-02-26 | 数坤（北京）网络科技有限公司 | A kind of coronary artery dividing method and device |
| CN109325948A (en) * | 2018-10-09 | 2019-02-12 | 数坤（北京）网络科技有限公司 | A kind of coronary artery dividing method and device based on special area optimization |
| CN110942447A (en) * | 2019-10-18 | 2020-03-31 | 平安科技（深圳）有限公司 | OCT image segmentation method, device, equipment and storage medium |
| WO2021073117A1 (en) * | 2019-10-18 | 2021-04-22 | 平安科技（深圳）有限公司 | Oct image segmentation method, apparatus and device, and storage medium |
| CN113083724A (en) * | 2021-03-24 | 2021-07-09 | 北京霍里思特科技有限公司 | Ore identification method, detection mechanism and mineral product sorting machine |
| CN113096115A (en) * | 2021-04-28 | 2021-07-09 | 博动医学影像科技（上海）有限公司 | Coronary artery plaque state evaluation method and device and electronic equipment |
| WO2022236995A1 (en) * | 2021-05-12 | 2022-11-17 | 深圳市中科微光医疗器械技术有限公司 | Guided detection and scoring method for calcified plaque, and device and storage medium |
| CN113610810A (en) * | 2021-08-09 | 2021-11-05 | 华力创科学(深圳)有限公司 | Blood vessel detection method based on Markov random field |
| CN114820678A (en) * | 2022-06-27 | 2022-07-29 | 天津恒宇医疗科技有限公司 | Automatic extraction method and system for inner contour of blood vessel wall based on OCT image |
| CN114882017A (en) * | 2022-06-30 | 2022-08-09 | 中国科学院大学 | Method and device for detecting thin fiber cap plaque based on intracranial artery image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108109149A (en) | A kind of coronary artery OCT image automatic division method | |
| Khan et al. | Classification of melanoma and nevus in digital images for diagnosis of skin cancer | |
| Navarro et al. | Accurate segmentation and registration of skin lesion images to evaluate lesion change | |
| CN109522908B (en) | Image significance detection method based on region label fusion | |
| Sheng et al. | Retinal vessel segmentation using minimum spanning superpixel tree detector | |
| Quellec et al. | Three-dimensional analysis of retinal layer texture: identification of fluid-filled regions in SD-OCT of the macula | |
| US9607391B2 (en) | Image object segmentation using examples | |
| Chakravarty et al. | Joint optic disc and cup boundary extraction from monocular fundus images | |
| Mary et al. | An empirical study on optic disc segmentation using an active contour model | |
| Motta et al. | Vessel optimal transport for automated alignment of retinal fundus images | |
| CN107845098A (en) | Liver cancer image full-automatic partition method based on random forest and fuzzy clustering | |
| Khan et al. | A region growing and local adaptive thresholding-based optic disc detection | |
| Cavalcanti et al. | Macroscopic pigmented skin lesion segmentation and its influence on lesion classification and diagnosis | |
| WO2017017685A1 (en) | Image processing system and method | |
| Riaz et al. | Impact of visual features on the segmentation of gastroenterology images using normalized cuts | |
| Saha et al. | Circular shape constrained fuzzy clustering (CiscFC) for nucleus segmentation in Pap smear images | |
| Wang et al. | Active contours driven by multi-feature Gaussian distribution fitting energy with application to vessel segmentation | |
| Panda et al. | Automated retinal nerve fiber layer defect detection using fundus imaging in glaucoma | |
| Senapati | Bright lesion detection in color fundus images based on texture features | |
| Dharmawan et al. | A new optic disc segmentation method using a modified Dolph-Chebyshev matched filter | |
| Salih et al. | Fast optic disc segmentation using FFT-based template-matching and region-growing techniques | |
| Chopra et al. | Recent studies on segmentation techniques for food recognition: A survey | |
| Karkuzhali et al. | Robust intensity variation and inverse surface adaptive thresholding techniques for detection of optic disc and exudates in retinal fundus images | |
| Lin et al. | Fast segmentation of porcelain images based on texture features | |
| CN109509202B (en) | Cage layer image segmentation method based on improved active contour model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180601 |
|
| RJ01 | Rejection of invention patent application after publication |