CN108416792A - Medical computer tomoscan image dividing method based on movable contour model - Google Patents
Medical computer tomoscan image dividing method based on movable contour model Download PDFInfo
- Publication number
- CN108416792A CN108416792A CN201810038957.3A CN201810038957A CN108416792A CN 108416792 A CN108416792 A CN 108416792A CN 201810038957 A CN201810038957 A CN 201810038957A CN 108416792 A CN108416792 A CN 108416792A
- Authority
- CN
- China
- Prior art keywords
- formula
- function
- indicate
- image
- definition
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
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
- G06T7/149—Segmentation; Edge detection involving deformable models, e.g. active contour models
-
- 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
- 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
- G06T2207/10072—Tomographic images
- G06T2207/10104—Positron emission tomography [PET]
Abstract
The present invention discloses a kind of medical computer tomoscan image dividing method based on movable contour model, belongs to field of medical image processing.First, GAC models and SBGFRLS models are combined using weight function, improve the possibility of zero level collection laboratory medicine image multilayer profile；Secondly, to be further simplified the level set algorithm in model, gaussian filtering technology is introduced, improves the convergence rate of parted pattern；Finally, introduce Hausdorff distance carry out target spacing from measurement, it solves the problems, such as that conventional model is poor to the sunken region segmentation effect of medical image convexity, and then improves the capture ability to multiple target object edge in medical computer tomoscan image, there is higher segmentation precision.
Description
Technical field
The present invention relates to image processing field, especially a kind of fast convergence rate, to initial profile curve position and noise
The high medical computer tomoscan image dividing method based on movable contour model of robustness.
Background technology
Image segmentation is that image processing field is most basic, is also one of sixty-four dollar question.Although numerous researchers are
Propose a variety of image partition methods, the dividing method such as based on Morphological watersheds, the dividing method based on gray threshold, base
Dividing method in edge detection, dividing method based on partial differential equation etc., but not yet occur at present a kind of suitable for institute
There is the image partition method of application scenario.
Constantly bringing forth new ideas and develop with medical imaging device and Examined effect, digitalized image makes computer-aided detection
It surveys and computer-aided diagnosis is possibly realized, and application of the imageological examination in clinical disease diagnosis is further extensive, especially
Increasingly important role is shown in the works in tumour " precisely medical treatment ".In order to which the state of an illness to patient makes accurate judgement, doctor
Life needs effectively to divide the lesions position for including in medical image.Medical image segmentation is to extraction sudden change region, surveys
It measures specific organization and realizes and rebuild three-dimensional basis, be the important step for carrying out focal area discrimination and delineating, and precisely
Whether the height of the basis of medical treatment and key, segmentation quality directly influences diagnosis of the clinician to conditions of patients.However,
It is introduced according to image department doctor, on the one hand, due to the shadow by factors such as sufferer physiology, position movement and acquisition systems
It rings, collected medical image quality is not high sometimes；On the other hand, since the tissue to be divided is different, imaging mechanism is different, point
Segmentation method is often inconsistent.For this purpose, doctor still needed to when clinical treatment is implemented by manual operations be partitioned into patient lesion and
Perienchyma, it is time-consuming bothersome, influence the diagnosis efficiency of clinician.Although moreover, computed tomography (Computed
Tomography, CT) image high resolution, positioning focal area it is more accurate, but have tumor-infiltrated tissue with just
When the density often organized is without significant difference, there are certain difficulty for judgement of the doctor to knub position and boundary.So there is an urgent need to
For certain medical image, practicable dividing method.
Jehan-Besson et al. proposes the medical image cutting method based on extreme learning machine, when reducing segmentation
Between, but precision is relatively low.Wang Yu et al. proposes the multi-threshold Medical image segmentation algorithm based on one-dimensional Otsu, although certain
Segmentation precision height is improved in degree, but it is poor to the segmentation effect in the convex sunken region of depth.The ash having in view of medical image
The characteristics such as degree is unevenly distributed, edge blurry, noise intensity are big, in recent years Variation Model show that increasing application is latent
Power.Variation Model is a kind of model by " internal force " constraint, " external force " driving, has good segmentation effect to medical image.
D.Mumford and J.Shah first proposed Mumford-Shah Variation Models, and taking the lead in will be based on the partial differential mould that image restores
Type extends to image segmentation field.Chan et al. simplifies Mumford-Shah models, and then proposes CV models, advantage
It is that segmentation result is unrelated with initial profile position, can accurately extracts the boundary of image, and to the robustness of noise height.But
It is that, due to introducing level set function, the calculation amount of CV models is higher, it is also necessary to level set function is reinitialized, under gradient
Drop is easily trapped into local minimum.In addition, since CV models are fitted based on global gray scale, it is not particularly suited for intensity profile not
Uniform medical image segmentation.For this purpose, Li et al. people proposes the RSF (Region of the expansible energy term of inclusion region
Scalable Fitting, RSF) model, the inhomogenous image of gray scale can be divided, but to the initial position of active contour and make an uproar
Sound is more sensitive.The global information of CV models is further introduced into GAC (Geodesic Active by Zhang et al.
Contours) in model, it is proposed that level-set segmentation model SBGFRLS (the Selective and based on region
Filtering Regularized Level Set Gaussian) model, take into account the area information and edge for considering image
Information reduces sensibility of the model to initial profile position, but can not effectively divide the gray scales such as medical image unevenness
Even heterogeneous image.
Invention content
The present invention is to provide a kind of fast convergence rate, to first to solve the above-mentioned technical problem present in the prior art
The high medical computer tomoscan image segmentation based on movable contour model of the robustness of beginning contour curve position and noise
Method.
Technical solution of the invention is：A kind of medical computer tomoscan image based on movable contour model point
Segmentation method, it is characterised in that carry out as follows：
Step 1. establishes the active contour parted pattern of medical computer tomoscan image, and level set movements equation is determined
Justice is provided by formula (1)：
(1)
It is describedIndicate the symbol pressure function for controlling curve evolution direction,Indicate image in coordinatePlace
Pixel value,Indicate level set function,Indicate gradient operator,It indicates about initial input imageWeight function,Expression is desired for 0, standard deviationGaussian kernel function,Indicate divergence operator,Table
Show normalized Hausdorff distances,WithIt is scale factor, wherein weight functionDefinition provided by formula (2)：
(2)
It is describedExpression is desired for 0, standard deviationGaussian kernel function, "" indicate convolution operation, it is normalized
The definition of Hausdorff distances is provided by formula (3)：
(3)
It is describedIndicate the Hausdorff distances between half-tone information in the inside and outside regional area of active contour curve, definition by
Formula (4) provides：
(4)
It is describedIndicate two discrete point setsWithDistance, definition provided by formula (5)：
(5)
It is describedIndicate the pixel set in active contour curvilinear inner region,Indicate active contour curved exterior region
Pixel set, coordinateThe pixel value at place is provided by formula (6)-formula (7)：
(6)
(7)
It is describedIndicate that Heaviside functions, definition are provided by formula (8)：
(8)
Step 2. inputs medical computer tomoscan image to be split, correlated Gaussian kernel function, and profit are set
Weight function is calculated with formula (2)；
If step 3.For CT images or MRI image, then step 3.1 is transferred to；IfFor PET image, it is transferred to step
3.2；
Step 3.1 is calculated using formula (9)The value of function：
(9)
It is describedWithThe average pixel value for indicating the inside and outside region of active contour respectively, defines respectively by formula
(10) it is provided with formula (11)：
(10)
(11)
It is transferred to step 4；
Step 3.2 is calculated using formula (12)The value of function：
(12)
It is describedExpression judges that the gray threshold of tumor region, definition are provided by formula (13)：
(13)
It is describedIndicate that the derived function of Weibull probability-distribution functions, definition are provided by formula (14)：
(14)
It is describedFor the form parameter of probability-distribution function,For the scale parameter of probability-distribution function；
Step 4. initializes level set function, and enable,；
Step 5. calculates the normalization Hausdorff distances of the pixel value in the inside and outside region of active contour using formula (3)
It calculates；
Step 6. updates level set function using finite difference calculus and formula (1)；
Step 7. checks whether stable convergence stops iteration to evolution curve if stable convergence, and algorithm terminates；Otherwise, it is transferred to
Step 3.
The present invention has the following advantages that compared with prior art：First, it is improved using normalized Hausdorff distances
The convergence rate of segmentation；Second, introduce the evolutionary process that gaussian filtering simplifies level set；Third is improved using weight function
Zero level collection examines the ability of object multilayer profile, while also improving to the areas depth Tu Xian in computed tomography images
The capture ability in domain and multiple target object edge.
Description of the drawings
Fig. 1 is segmentation result comparison diagram of the embodiment of the present invention with other methods to CT or MRI image.
Fig. 2 is segmentation result comparison diagram of the embodiment of the present invention with other methods to PET image.
Specific implementation mode
The medical computer tomoscan image dividing method based on movable contour model of the present invention, in accordance with the following steps
It carries out：
Step 1. establishes the active contour parted pattern of medical computer tomoscan image, and level set movements equation is determined
Justice is provided by formula (1)：
(1)
It is describedIndicate the symbol pressure function for controlling curve evolution direction,Indicate image in coordinatePlace
Pixel value,Indicate level set function,Indicate gradient operator,It indicates about initial input imageWeight function,Expression is desired for 0, standard deviationGaussian kernel function,Indicate divergence operator,
Indicate normalized Hausdorff distances,WithIt is scale factor, wherein weight functionDefinition given by formula (2)
Go out：
(2)
It is describedExpression is desired for 0, standard deviationGaussian kernel function, "" indicate convolution operation, it is normalized
The definition of Hausdorff distances is provided by formula (3)：
(3)
It is describedIndicate the Hausdorff distances between half-tone information in the inside and outside regional area of active contour curve, definition by
Formula (4) provides：
(4)
It is describedIndicate two discrete point setsWithDistance, definition provided by formula (5)：
(5)
It is describedIndicate the pixel set in active contour curvilinear inner region,Indicate active contour curved exterior region
Pixel set, coordinateThe pixel value at place is provided by formula (6)-formula (7)：
(6)
(7)
It is describedIndicate that Heaviside functions, definition are provided by formula (8)：
(8)
Step 2. inputs medical computer tomoscan image to be split, correlated Gaussian kernel function, and profit are set
Weight function is calculated with formula (2)；
If step 3.For CT images or MRI image, then step 3.1 is transferred to；IfFor PET image, it is transferred to step
Rapid 3.2；
Step 3.1 is calculated using formula (9)The value of function：
(9)
It is describedWithThe average pixel value for indicating the inside and outside region of active contour respectively, defines respectively by formula
(10) it is provided with formula (11)：
(10)
(11)
It is transferred to step 4；
Step 3.2 is calculated using formula (12)The value of function：
(12)
It is describedExpression judges that the gray threshold of tumor region, definition are provided by formula (13)：
(13)
It is describedIndicate that the derived function of Weibull probability-distribution functions, definition are provided by formula (14)：
(14)
It is describedFor the form parameter of probability-distribution function,For the scale parameter of probability-distribution function；
Step 4. initializes level set function, and enable,；
Step 5. calculates the normalization Hausdorff distances of the pixel value in the inside and outside region of active contour using formula (3)
It calculates；
Step 6. updates level set function using finite difference calculus and formula (1)；
Step 7. checks whether stable convergence stops iteration to evolution curve if stable convergence, and algorithm terminates；Otherwise, it is transferred to
Step 3.
The embodiment of the present invention and other methods are more as shown in Figure 1 to the segmentation result of CT or MRI image：Divide from left to right
It Wei not (a) original image；(b) segmentation result of the embodiment of the present invention；(c) Jehan-Besson propose based on extreme learning machine
Segmentation result；(c) the multi-threshold segmentation result based on one-dimensional Otsu that Wang Yu is proposed.
The embodiment of the present invention and other methods are more as shown in Figure 2 to the segmentation result of PET image：It is respectively from left to right
(a) original image；(b) segmentation result of the embodiment of the present invention；(c) point based on extreme learning machine that Jehan-Besson is proposed
Cut result；(c) the multi-threshold segmentation result based on one-dimensional Otsu that Wang Yu is proposed.
The embodiment of the present invention divides 3 width CT or MRI image with other methods（Fig. 1）Required iterations are compared with the time
As shown in table 1.
The embodiment of the present invention divides 3 width CT or MRI image with other methods（Fig. 1）Error rate comparison it is as shown in table 2.
The embodiment of the present invention divides 3 width PET images with other methods（Fig. 2）Required iterations and time comparison such as table
Shown in 3.
The embodiment of the present invention divides 3 width PET images with other methods（Fig. 2）Error rate comparison it is as shown in table 4.
The iterations of 3 width images are compared with the time in 1 Fig. 1 of table（Unit：Second）
The segmentation error rate comparison of 3 width images in 2 Fig. 1 of table
The iterations of 3 width images are compared with the time in 3 Fig. 2 of table（Unit：Second）
The segmentation error rate comparison of 3 width images in 4 Fig. 2 of table
。
Comparing result shows：The present invention can more accurately divide that contrast is low, gray scale is non-uniform in a short time
Heterogeneous medical computer tomoscan image.
Claims (1)
1. a kind of medical computer tomoscan image dividing method based on movable contour model, it is characterised in that by following step
It is rapid to carry out：
Step 1. establishes the active contour parted pattern of medical computer tomoscan image, and level set movements equation is determined
Justice is provided by formula (1)：
(1)
It is describedIndicate the symbol pressure function for controlling curve evolution direction,Indicate image in coordinatePlace
Pixel value,Indicate level set function,Indicate gradient operator,It indicates about initial input image's
Weight function,Expression is desired for 0, standard deviationGaussian kernel function,Indicate divergence operator,Indicate normalization
Hausdorff distances,WithIt is scale factor, wherein weight functionDefinition provided by formula (2)：
(2)
It is describedExpression is desired for 0, standard deviationGaussian kernel function, "" indicate convolution operation, it is normalized
The definition of Hausdorff distances is provided by formula (3)：
(3)
It is describedIt indicates the Hausdorff distances between half-tone information in the inside and outside regional area of active contour curve, defines by public affairs
Formula (4) provides：
(4)
It is describedIndicate two discrete point setsWithDistance, definition provided by formula (5)：
(5)
It is describedIndicate the pixel set in active contour curvilinear inner region,Indicate the picture in active contour curved exterior region
Element set, coordinateThe pixel value at place is provided by formula (6)-formula (7)：
(6)
(7)
It is describedIndicate that Heaviside functions, definition are provided by formula (8)：
(8)
Step 2. inputs medical computer tomoscan image to be split, correlated Gaussian kernel function is set, and utilizes
Formula (2) calculates weight function；
If step 3.For CT images or MRI image, then step 3.1 is transferred to；IfFor PET image, it is transferred to step
3.2；
Step 3.1 is calculated using formula (9)The value of function：
(9)
It is describedWithThe average pixel value for indicating the inside and outside region of active contour respectively, defines respectively by formula (10)
It is provided with formula (11)：
(10)
(11)
It is transferred to step 4；
Step 3.2 is calculated using formula (12)The value of function：
(12)
It is describedExpression judges that the gray threshold of tumor region, definition are provided by formula (13)：
(13)
It is describedIndicate that the derived function of Weibull probability-distribution functions, definition are provided by formula (14)：
(14)
It is describedFor the form parameter of probability-distribution function,For the scale parameter of probability-distribution function；
Step 4. initializes level set function, and enable,；
Step 5. calculates the normalization Hausdorff distances of the pixel value in the inside and outside region of active contour using formula (3)
It calculates；
Step 6. updates level set function using finite difference calculus and formula (1)；
Step 7. checks whether stable convergence stops iteration to evolution curve if stable convergence, and algorithm terminates；Otherwise, it is transferred to
Step 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810038957.3A CN108416792B (en) | 2018-01-16 | 2018-01-16 | Medical computed tomography image segmentation method based on active contour model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810038957.3A CN108416792B (en) | 2018-01-16 | 2018-01-16 | Medical computed tomography image segmentation method based on active contour model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108416792A true CN108416792A (en) | 2018-08-17 |
| CN108416792B CN108416792B (en) | 2021-07-06 |
Family
ID=63125713
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810038957.3A Expired - Fee Related CN108416792B (en) | 2018-01-16 | 2018-01-16 | Medical computed tomography image segmentation method based on active contour model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108416792B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110335287A (en) * | 2019-07-15 | 2019-10-15 | 北华航天工业学院 | The extracting method and device of Architectural drawing data |
| CN116740768A (en) * | 2023-08-11 | 2023-09-12 | 南京诺源医疗器械有限公司 | Navigation visualization method, system, equipment and storage medium based on nasoscope |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102024260A (en) * | 2010-12-30 | 2011-04-20 | 浙江工商大学 | Local Gamma fitting-based active contour SAR image segmentation method |
| US20120308110A1 (en) * | 2011-03-14 | 2012-12-06 | Dongguk University, Industry-Academic Cooperation Foundation | Automation Method For Computerized Tomography Image Analysis Using Automated Calculation Of Evaluation Index Of Degree Of Thoracic Deformation Based On Automatic Initialization, And Record Medium And Apparatus |
| CN103065309A (en) * | 2012-12-26 | 2013-04-24 | 辽宁师范大学 | Image segmentation method based on simplified local binary fitting (LBF) model |
| CN105761274A (en) * | 2016-03-21 | 2016-07-13 | 辽宁师范大学 | Medical image segmentation method combining edge and area information |
| CN106570867A (en) * | 2016-10-18 | 2017-04-19 | 浙江大学 | ACM (Active Contour Model) image rapid segmentation method based on gray scale morphological energy method |
| CN106803260A (en) * | 2016-12-28 | 2017-06-06 | 辽宁师范大学 | Infrared ship activity of imagination contours segmentation method based on the convex optimization of local entropy |
| CN106997596A (en) * | 2017-04-01 | 2017-08-01 | 太原理工大学 | A kind of Lung neoplasm dividing method of the LBF movable contour models based on comentropy and joint vector |
| CN107154047A (en) * | 2017-04-24 | 2017-09-12 | 天津大学 | Multi-mode brain tumor image blend dividing method and device |
-
2018
- 2018-01-16 CN CN201810038957.3A patent/CN108416792B/en not_active Expired - Fee Related
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102024260A (en) * | 2010-12-30 | 2011-04-20 | 浙江工商大学 | Local Gamma fitting-based active contour SAR image segmentation method |
| US20120308110A1 (en) * | 2011-03-14 | 2012-12-06 | Dongguk University, Industry-Academic Cooperation Foundation | Automation Method For Computerized Tomography Image Analysis Using Automated Calculation Of Evaluation Index Of Degree Of Thoracic Deformation Based On Automatic Initialization, And Record Medium And Apparatus |
| CN103065309A (en) * | 2012-12-26 | 2013-04-24 | 辽宁师范大学 | Image segmentation method based on simplified local binary fitting (LBF) model |
| CN105761274A (en) * | 2016-03-21 | 2016-07-13 | 辽宁师范大学 | Medical image segmentation method combining edge and area information |
| CN106570867A (en) * | 2016-10-18 | 2017-04-19 | 浙江大学 | ACM (Active Contour Model) image rapid segmentation method based on gray scale morphological energy method |
| CN106803260A (en) * | 2016-12-28 | 2017-06-06 | 辽宁师范大学 | Infrared ship activity of imagination contours segmentation method based on the convex optimization of local entropy |
| CN106997596A (en) * | 2017-04-01 | 2017-08-01 | 太原理工大学 | A kind of Lung neoplasm dividing method of the LBF movable contour models based on comentropy and joint vector |
| CN107154047A (en) * | 2017-04-24 | 2017-09-12 | 天津大学 | Multi-mode brain tumor image blend dividing method and device |
Non-Patent Citations (4)
| Title |
|---|
| HONGWEI JI等: "ACM-Based Automatic Liver Segmentation From 3-D CT Images by Combining Multiple Atlases and Improved Mean-Shift Techniques", 《IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS》 * |
| 方玲玲等: "图像分割的活动轮廓模型研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
| 杨振森等: "基于改进GAC模型的二值水平集前列腺超声图像自动分割算法", 《中国科学技术大学学报》 * |
| 王相海等: "活动轮廓模型的图像分割方法综述", 《模式识别与人工智能》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110335287A (en) * | 2019-07-15 | 2019-10-15 | 北华航天工业学院 | The extracting method and device of Architectural drawing data |
| CN116740768A (en) * | 2023-08-11 | 2023-09-12 | 南京诺源医疗器械有限公司 | Navigation visualization method, system, equipment and storage medium based on nasoscope |
| CN116740768B (en) * | 2023-08-11 | 2023-10-20 | 南京诺源医疗器械有限公司 | Navigation visualization method, system, equipment and storage medium based on nasoscope |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108416792B (en) | 2021-07-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111008984B (en) | Automatic contour line drawing method for normal organ in medical image | |
| Chung et al. | Automatic lung segmentation with juxta-pleural nodule identification using active contour model and Bayesian approach | |
| Zhou et al. | Automated lung segmentation and smoothing techniques for inclusion of juxtapleural nodules and pulmonary vessels on chest CT images | |
| Thirion et al. | Deformation analysis to detect and quantify active lesions in three-dimensional medical image sequences | |
| 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 | |
| Hassan et al. | Joint segmentation and quantification of chorioretinal biomarkers in optical coherence tomography scans: A deep learning approach | |
| JP2002523123A (en) | Method and system for lesion segmentation and classification | |
| CN104268873A (en) | Breast tumor partition method based on nuclear magnetic resonance images | |
| Wang et al. | Automatic approach for lung segmentation with juxta-pleural nodules from thoracic CT based on contour tracing and correction | |
| Jebarani et al. | A novel hybrid K-means and GMM machine learning model for breast cancer detection | |
| US20210271914A1 (en) | Image processing apparatus, image processing method, and program | |
| Hong et al. | Automatic lung nodule matching on sequential CT images | |
| CN113168912A (en) | Determining growth rate of objects in 3D data sets using deep learning | |
| EP2208183A2 (en) | Computer-aided detection (cad) of a disease | |
| US7835555B2 (en) | System and method for airway detection | |
| WO2018219818A1 (en) | Quantified aspects of lesions in medical images | |
| CN111899850A (en) | Medical image information processing method, display method and readable storage medium | |
| CN108416792A (en) | Medical computer tomoscan image dividing method based on movable contour model | |
| CN109215035B (en) | Brain MRI hippocampus three-dimensional segmentation method based on deep learning | |
| Fu et al. | Advancements of medical image enhancement in healthcare applications | |
| Ou et al. | A sub-pixel accurate quantification of joint space narrowing progression in rheumatoid arthritis | |
| CN112116623B (en) | Image segmentation method and device | |
| WO2007033170A1 (en) | System and method for polyp detection in tagged or non-tagged stool images | |
| Bhakta et al. | Lung tumor segmentation and staging from ct images using fast and robust fuzzy C-Means clustering | |
| Hu et al. | Image manifold revealing for breast lesion segmentation in DCE-MRI |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210706 Termination date: 20220116 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |