CN102542556A - Method for automatically extracting ultrasonic breast tumor image - Google Patents
Method for automatically extracting ultrasonic breast tumor image Download PDFInfo
- Publication number
- CN102542556A CN102542556A CN2010106133884A CN201010613388A CN102542556A CN 102542556 A CN102542556 A CN 102542556A CN 2010106133884 A CN2010106133884 A CN 2010106133884A CN 201010613388 A CN201010613388 A CN 201010613388A CN 102542556 A CN102542556 A CN 102542556A
- Authority
- CN
- China
- Prior art keywords
- image
- tumor
- model
- algorithm
- automatically
- 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
Images
Abstract
The invention relates to the field of signal processing of biomedicine, in particular to a method for automatically extracting an ultrasonic breast tumor image. The method comprises the following steps of: 1, selecting a target image, namely marking a square frame by a user through a mouse so as to contain the tumor in the square frame; 2, automatically extracting an edge of the ultrasonic tumor image, namely suppressing noise of a cut image through a speckle reducing anisotropic diffusion (SRAD) algorithm; 3, obtaining an image I1 subjected to noise suppression through the step (2), and automatically segmenting the tumor image by utilizing an improved geometrical active contour model, namely 3a, pre-processing the image, 3b, optimizing the model, 3c, extracting the edge of the image, 3d, updating the model, and 3e, converging the image. A novel energy function is provided to improve the original model, so that the model is more suitable for lesion extraction of medical ultrasonic tumor image, and the accuracy and practicality of the algorithm are improved further.
Description
Technical field
The invention belongs to the processing of biomedical signals field, be specifically related to tumour ultrasonoscopy extraction method.
Background technology
Breast cancer has become women's No.1 killer, and its morbidity number significantly rises with the speed of average annual 3%-5%, and increasingly serious trend is arranged.Research shows that if inspection in time in early days, cancer can be cured, and cure rate is up to more than 92%.It is thus clear that the early detection of tumor of breast is to the crucial effects that healed the sick.Based on the detection technique of ultrasonoscopy be with the fastest developing speed in the medical science, use one of tumor disease detection technique the most widely.Yet because the special mechanism of its imaging, ultrasonoscopy tumor of breast focus inspection problem also is not resolved so far, is one of current main research focus, is a classic problem yet.To this problem, the researchist has proposed a large amount of partitioning algorithms so far both at home and abroad, yet up to the present the shortcoming that these algorithms all exist and are of limited application, limitation is stronger does not also exist a kind of general focus detection method.
In recent years, in the image segmentation field, show actively based on the movable contour model method (like the snake model) at edge, all obtained in many aspects using widely.But they have following shortcoming usually:
(1) relatively more responsive to noise and clutter;
(2) in weak edge the border leakage phenomenon takes place easily;
(3) relatively harsher to the requirement of starting condition.
Chan and Vese are at article " Chan, T. F.; Vese, L. A.; Active contours without edges. Image Processing; IEEE Transactions; Vol.10, pp.266 – 277,2001 " in; proposed a kind of geometric active contour model-Chan-Vese model based on the zone of classics, it has greatly improved the above-mentioned shortcoming of Snake model.In this model, initial profile can be arranged on any position of target area, makes the practicality of model be further strengthened; Yet, for ultrasonoscopy, because it receives the serious interference of speckle noise in imaging process; Therefore, at the tumor boundaries near zone, speckle noise tends to form the bulk of some catastrophe points or sudden change; Thereby, caused the Chan-Vese model when extracting lesion boundary, to be easy to that speckle noise is regarded as the edge and extracted.In addition, the Chan-Vese model is to multiobject image Segmentation, and it is inaccurate to occur the location, edge easily.
Summary of the invention
The present invention seeks to defective, a kind of ultrasonoscopy tumor of breast extraction method is provided to above-mentioned cutting techniques existence.
Order of the present invention can realize not having artificial tumor of breast edge extracting of intervening; Can not only extract the tumor focus zone apace; And improve accuracy greatly; Thereby establish the good technical basis for the differentiation of tumor of breast and computer-aided diagnosis, help advancing the application of Medical Image Processing technology in the tumor of breast clinical diagnosis, the useful information of tumor region is provided for numerous doctors.
The present invention realizes through following technical scheme, specifically may further comprise the steps and carries out.
1. choose target image: the user utilizes mouse to draw and gets a square frame, tumour is included within this square frame, and its purpose one is in order to cut out irrelevant information, to reduce and disturb; The 2nd, in order to improve the real-time of system.
2. tumour ultrasonoscopy edge extracts automatically: adopt SRAD anisotropy broadcast algorithm, carry out squelch to cutting out the image that obtains; The outstanding advantage applies of this algorithm exists: can effectively remove speckle noise and preserving edge information, can strengthen the step-like edge simultaneously; Therefore, spot (Speckle) interference of noise that adopts this algorithm to reduce as much as possible to be caused by ultrasonic imaging mechanism is for subsequent treatment provides good basis; The iterative equation of SRAD algorithm is as follows:
Wherein,
Be the image after cutting out;
Be coefficient of diffusion;
The supporting domain of presentation video,
For
The border, and
For
Outer normal vector;
is the output image after the iteration each time;
3. obtain through the image after the squelch by step (2)
I1 ,Utilize improved geometric active contour model that tumor image is cut apart automatically:
The pre-service of 3a image: the expansion and the caustic solution that utilize geometric shape are to subimage
Carry out pre-service, generate the initial active outline line automatically
, i.e. zero level collection, and generation symbolic distance function S DF, i.e. level set function
,The computing formula that expansion and caustic solution adopt is following:
Wherein
Expression has the structural element of definite shape and size;
I1 doesImage after the squelch;
The 3b model optimization: in order to improve the huge problem of calculated amount that the original geometry active contour develops, system has adopted the arrowband algorithm, promptly only considers the near zone of zero level collection, only upgrades the SDF in this narrowband region at every turn, thereby improves the work efficiency of algorithm;
3c Edge extraction: extract problems such as inaccurate to the noise immunity difference of ultrasonoscopy and to multiple goal in order to improve traditional Chan-Vese model; The gradient information of systems incorporate image and area grayscale information propose a new energy term based on gradient:
Where,
for the re-initialization of the level set function;
and
, respectively pixels inside the target area average and average gradient;
and
, respectively pixels outside the target area average and average gradient;
and
for the adjustable parameters; and there:
The 3d model modification: when zero level collection curve near or when touching the border, arrowband, according to the evolution formula, upgrade level set function automatically, and recomputate new narrowband region;
Convergence of 3e model and criterion: when whether the inspection iteration restrained, if convergence does not then forward step 3b to, otherwise then calculating stopped, and zero level collection curve stops to develop, and the Rule of judgment of getting iteration convergence is:
Wherein,
is the level set function that the n time iteration obtains;
is time step, and
grid sum that
satisfied in expression.
Advantage of the invention and effect
Advantage of the invention and effect
The present invention compared with prior art has the following advantages:
1, the present invention has adopted advanced geometric shape method auto-initiation level set function, can not only avoid artificial participation effectively, has improved the automatization level of this system.
2, the present invention utilizes the arrowband method to improve original geometric active contour model, has greatly reduced calculated amount, has improved the real-time performance of system.
3, the present invention proposes a new energy function master pattern is improved, make that this model is more suitable for extracting in the focus of medical ultrasonic image tumour, further improved algorithm accuracy and practicality.
Accompanying drawing and explanation thereof
Fig. 1 is a process flow diagram of the present invention.
Fig. 2 is the present invention and the contrast and experiment figure of original geometry movable contour model algorithm on the noisy image of multiple goal;
Wherein, 2-a is former figure; 2-b is the experimental result of Li; 2-c is an experimental result of the present invention.
Fig. 3 is directed against the comparative test result figure of ultrasonoscopy tumor of breast extraction for the present invention and original geometry movable contour model algorithm;
Wherein, 3-a is former figure; 3-b is the experimental result of Li; 3-c is an experimental result of the present invention.
Fig. 4-a is the ultrasonoscopy that contains the stringiness tumor of breast that diagnostic ultrasonic equipment collects, the region that the tumour that the white rectangle square frame is got for the user draws is general; Fig. 4-b gets the subimage that the back forms for drawing; Fig. 4-c is the image through obtaining after the SRAD squelch, and by the automatic zero level collection curve that generates of this image; Fig. 4-d carries out in the automatic focus extraction the automatic evolutionary process of zero level collection curve for the improvement Chan-Vese model algorithm that adopts the present invention to propose; Fig. 4-e is that zero level collection curve evolvement finally stops, and the contour curve that obtains promptly extracts and obtained the tumor of breast edge; The final extraction that Fig. 4-f obtains for this diagnostic ultrasonic equipment utilization the present invention is displayed map as a result.
The present invention brings forward a new energy function, has realized the effective improvement to original Chan-Vese model, makes that the Chan-Vese model is more suitable for extracting in the focus of medical ultrasonic image tumour.Compare through simulation study and experiment, the present invention all obviously is superior to the Chan-Vese model on extraction performance and time loss, concrete visible lab diagram 2 and Fig. 3.Wherein, Fig. 2 is the present invention and the contrast and experiment figure of Chan-Vese model algorithm on the noisy image of multiple goal; Fig. 3 is directed against the comparative test result figure of ultrasonoscopy tumor of breast extraction for the present invention and Chan-Vese model algorithm.In addition, the present invention need not manual intervention in the leaching process of tumor focus, and automatization level is higher.Fig. 4 is loaded on the diagnostic ultrasonic equipment for the present invention, carries out the instance graph of clinical assistant diagnosis.In the selection of this instance parameter;
;
;
=
=0.4,
; The SRAD iterations is set to 5 times, and the iteration stopping number of times of algorithm is 50 times.
Concrete embodiment:
The present invention is primarily aimed at that extraction and the specialized designs of ultrasonoscopy tumor of breast implement.On the basis of the characteristics of fully having studied ultrasonic imaging mechanism and ultrasonoscopy; The present invention has used spot noise reduction anisotropy diffusion (SRAD) algorithm ultrasonoscopy has been carried out necessity ground noise remove, and has kept the information at tumor region edge effectively and strengthened edge contour; Following closely be, adopt improved geometric active contour model (Chan-Vese model) that the image after the denoising is handled, extract the focus zone of tumor of breast.Obviously, the entire process process need not manual intervention, and the present invention can automatically extract and obtain final result.
With reference to Fig. 1, the present invention is based on the medical ultrasonic image tumour extraction that improves movable contour model and comprise:
Step 1: cut out, obtain area-of-interest.
The present invention only needs the user probably to confirm the position of tumour, promptly can utilize mouse on the medical image that shows, to draw the approximate range of getting tumour and get final product, and system just can automatically propose important informations such as concrete shape and the size of tumour.
Step 2: squelch, it is readable to improve ultrasonoscopy, is convenient to subsequent treatment.
The speckle noise of ultrasonoscopy has reduced ultrasonic image quality widely, has had a strong impact on the subsequent treatment of ultrasonoscopy, especially to the extraction and the identification of tumor focus.The present invention introduces SRAD anisotropy broadcast algorithm, can effectively remove speckle noise and preserving edge information, can strengthen the step-like edge simultaneously.Its iterative equation is as follows:
Wherein coefficient of diffusion c (q) is:
In the formula, q is the instantaneous coefficient operator of being calculated by local variance; Q0 (t) is the spot scale coefficient, is used to control level and smooth degree.Experiment shows that q0 (t) gets [0,1], and effect is good.Here, get q0 (t)=0.4.Instantaneous coefficient operator q is defined as:
This operator has comprised gradient operator and Laplace operator, is used for detecting the edge of spot image.Obtain higher value with high-contrast profile place on the edge of, and obtain smaller value at homogeneous area.
Step 3: system extracts borderline tumor automatically, need not man-machine interactively.
If the image that obtains behind the noise suppression does
I1, next, utilize improved geometric active contour model, realize that the edge of tumor of breast extracts automatically:
1) utilize expansion and corrosion in the geometric shape that the image I after the denoising is handled; Automatically generate initial active outline line
; And then; Can calculate the symbolic distance function, just level set function
.It is following with the definition of erosion operation to expand:
Wherein
expression has the structural element of definite shape and size.
2) for improving real-time, system has adopted the arrowband algorithm that model is optimized, and only considers that promptly all pixels are that 4 pixels are with interior zone to zero level collection curve distance; In iterative computation after this, only upgrade the SDF in this narrowband region at every turn, thereby improve the work efficiency of algorithm.
3) native system makes full use of the gradient information and the area grayscale information of image, proposes a more efficiently energy function:
Where,
and
respectively pixels inside the target area average and average gradient;
and
, respectively pixels outside the target area average and the average gradient.
is illustrated in the Grad that point
is located,
be the level set function after reinitializing.
Utilize the variational method to this total energy function minimization, obtain level set function
evolutionary process and be:
For the numerical solution of level set function
EVOLUTION EQUATION, adopt the implicit iterative solution.Can prove for the above-mentioned time-based level set function that obtains by the variation minimization
EVOLUTION EQUATION; Its implied format iterative solution method is unconditional stability; Therefore; Time step
can suitably strengthen; With the evolution of acceleration curve, the present invention gets
.
4) according to the evolution formula, when zero level collection curve near or when touching the border, arrowband, system will upgrade level set function automatically and recomputate new narrowband region.
When 5) whether the inspection iteration restrains, if convergence does not then forward step 2 to), otherwise then calculating stops, zero level collection curve stops to develop, and at this moment, this curve will drop on the edge of tumour exactly, thereby obtain the extraction result of final tumour profile.For fear of unnecessary iterative computation, the Rule of judgment of getting iteration convergence is:
Wherein,
is the level set function that the n time iteration obtains, and
grid sum that
satisfied in expression.
Below verify the validity and the practicality of the inventive method through emulation experiment and concrete clinical practice.The method that is compared is the method that people such as Li propose, concrete list of references " T. F. Chan, L. A. Vese; Active contours without edges, IEEE Trans. Image Processing, 2001; vol.10, no.2, pp.266-277. "
Fig. 2 and Fig. 3 are respectively noisy multiple goal composograph and medical ultrasonic tumor of breast image, and the comparative result of experiment.As can be seen from the figure, the present invention can correctly split target, compares the scheduling algorithm with Li, has higher accuracy rate, and has good real time performance, and is as shown in table 1.
Two kinds of methods of table 1 are cut apart the comparison of required time.
Fig. 2 | Fig. 3 | |
The Li method | 11.54 | 12.33 |
The present invention | 0.97 | 1.25 |
Fig. 4 is loaded on the diagnostic ultrasonic equipment for the present invention, carries out the instance graph of clinical assistant diagnosis.The result of clinical testing shows that the present invention has very high practicality, and the important information of tumour can be provided for the clinician practically.
Claims (1)
1. tumor of breast ultrasonoscopy extraction method, carry out according to the following steps:
(1) choose target image: the user utilizes mouse to cut out to include the block scheme picture of pending tumor of breast;
(2) tumour ultrasonoscopy edge extracts automatically: adopt SRAD anisotropy broadcast algorithm, carry out squelch to cutting out the image that obtains, the iterative equation of SRAD algorithm is following:
Wherein,
Be the image after cutting out;
Be coefficient of diffusion;
The supporting domain of presentation video,
For
The border, and
For
Outer normal vector;
(3) obtain through the image after the squelch by step (2)
I1 ,Utilize improved geometric active contour model that tumor image is cut apart automatically:
The pre-service of 3a image: the expansion and the caustic solution that utilize geometric shape are to subimage
Carry out pre-service, generate the initial active outline line automatically
, i.e. zero level collection, and generation symbolic distance function S DF, i.e. level set function
,The computing formula that expansion and caustic solution adopt is following:
Wherein
Expression has the structural element of definite shape and size;
I1 doesImage after the squelch;
3b model optimization: adopted the arrowband algorithm, promptly only considered the near zone of zero level collection, upgraded the SDF in this narrowband region again, thereby improved the work efficiency of algorithm;
3c Edge extraction: the gradient information of combining image and area grayscale information; Adopt minimization of energy function
to calculate, make zero level collection curve develop along the borderline tumor direction:
Where,
for the re-initialization of the level set function;
and
, respectively pixels inside the target area average and average gradient;
and
are outside of the target area of the pixel average and the average gradient,
and
for the adjustable parameters; and there:
The 3d model modification: when zero level collection curve near or when touching the border, arrowband, according to the evolution formula, upgrade level set function automatically, and recomputate new narrowband region;
Convergence of 3e model and criterion
:When whether the inspection iteration restrained, if convergence does not then forward step 3b to, otherwise then calculating stopped, and zero level collection curve stops to develop, and the Rule of judgment of getting iteration convergence is:
)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201010613388.4A CN102542556B (en) | 2010-12-30 | 2010-12-30 | Method for automatically extracting ultrasonic breast tumor image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201010613388.4A CN102542556B (en) | 2010-12-30 | 2010-12-30 | Method for automatically extracting ultrasonic breast tumor image |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102542556A true CN102542556A (en) | 2012-07-04 |
CN102542556B CN102542556B (en) | 2014-10-08 |
Family
ID=46349379
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201010613388.4A Expired - Fee Related CN102542556B (en) | 2010-12-30 | 2010-12-30 | Method for automatically extracting ultrasonic breast tumor image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102542556B (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103679685A (en) * | 2012-09-11 | 2014-03-26 | 北京三星通信技术研究有限公司 | Image processing system and image processing method |
CN104091330A (en) * | 2014-06-27 | 2014-10-08 | 汕头职业技术学院 | Doppler ultrasound color pixel density computing method |
CN104091331A (en) * | 2014-06-27 | 2014-10-08 | 深圳市开立科技有限公司 | Method, device and system for segmenting ultrasonic focus image |
CN104376564A (en) * | 2014-11-24 | 2015-02-25 | 西安工程大学 | Method for extracting rough image edge based on anisotropism Gaussian directional derivative filter |
CN106023231A (en) * | 2016-06-07 | 2016-10-12 | 首都师范大学 | Method for automatically detecting cattle and sheep in high resolution image |
CN106108932A (en) * | 2016-07-21 | 2016-11-16 | 四川大学 | Full-automatic kidney region of interest extraction element and method |
CN106570868A (en) * | 2015-10-10 | 2017-04-19 | 中国科学院深圳先进技术研究院 | Three-dimensional liver tumor semi-automatic segmentation method based on level set |
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 |
CN106651892A (en) * | 2016-12-21 | 2017-05-10 | 福建师范大学 | Edge detection method |
CN110264461A (en) * | 2019-06-25 | 2019-09-20 | 南京工程学院 | Microcalciffcation point automatic testing method based on ultrasonic tumor of breast image |
CN113324727A (en) * | 2019-07-16 | 2021-08-31 | 中国人民解放军空军工程大学 | Schlieren image processing method for compressed corner supersonic flow field structure |
WO2023205896A1 (en) * | 2022-04-28 | 2023-11-02 | Afx Medical Inc. | Systems and methods for detecting structures in 3d images |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1471034A (en) * | 2002-07-24 | 2004-01-28 | 中国科学院自动化研究所 | Medical image segmentation method based on horizontal collection and watershed method |
US20060158447A1 (en) * | 2005-01-14 | 2006-07-20 | Mcgraw Tim | System and method for fast tensor field segmentation |
-
2010
- 2010-12-30 CN CN201010613388.4A patent/CN102542556B/en not_active Expired - Fee Related
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1471034A (en) * | 2002-07-24 | 2004-01-28 | 中国科学院自动化研究所 | Medical image segmentation method based on horizontal collection and watershed method |
US20060158447A1 (en) * | 2005-01-14 | 2006-07-20 | Mcgraw Tim | System and method for fast tensor field segmentation |
Non-Patent Citations (3)
Title |
---|
BING LI ET AL.: "Active Contour External Force Using Vector Field Convolution for Image Segmentation", 《IEEE TRANSACTIONS ON IMAGE PROCESSING》, vol. 16, no. 8, 31 August 2007 (2007-08-31), pages 2096 - 2106, XP011187313, DOI: doi:10.1109/TIP.2007.899601 * |
MARIA MERCEDE CERIMELE ET AL.: "Coastline Detection from SAR Images by Level Set Model", 《LNCS》, vol. 5716, 31 December 2009 (2009-12-31), pages 364 - 373, XP019128164 * |
NOURA AZZABOU ET AL.: "Spatio-temporal Speckle Reduction in Ultrasound Sequences", 《LNCS》, vol. 5241, 31 December 2008 (2008-12-31), pages 951 - 958, XP019104999 * |
Cited By (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103679685A (en) * | 2012-09-11 | 2014-03-26 | 北京三星通信技术研究有限公司 | Image processing system and image processing method |
CN103679685B (en) * | 2012-09-11 | 2018-03-27 | 北京三星通信技术研究有限公司 | Image processing system and image processing method |
CN104091330A (en) * | 2014-06-27 | 2014-10-08 | 汕头职业技术学院 | Doppler ultrasound color pixel density computing method |
CN104091331A (en) * | 2014-06-27 | 2014-10-08 | 深圳市开立科技有限公司 | Method, device and system for segmenting ultrasonic focus image |
CN104091331B (en) * | 2014-06-27 | 2015-07-29 | 深圳开立生物医疗科技股份有限公司 | A kind of ultrasonic focus image partition method, Apparatus and system |
CN104091330B (en) * | 2014-06-27 | 2017-01-11 | 汕头职业技术学院 | Doppler ultrasound color pixel density computing method |
CN104376564A (en) * | 2014-11-24 | 2015-02-25 | 西安工程大学 | Method for extracting rough image edge based on anisotropism Gaussian directional derivative filter |
CN104376564B (en) * | 2014-11-24 | 2018-04-24 | 西安工程大学 | Method based on anisotropic Gaussian directional derivative wave filter extraction image thick edge |
CN106570868A (en) * | 2015-10-10 | 2017-04-19 | 中国科学院深圳先进技术研究院 | Three-dimensional liver tumor semi-automatic segmentation method based on level set |
CN106023231A (en) * | 2016-06-07 | 2016-10-12 | 首都师范大学 | Method for automatically detecting cattle and sheep in high resolution image |
CN106108932A (en) * | 2016-07-21 | 2016-11-16 | 四川大学 | Full-automatic kidney region of interest extraction element and method |
CN106108932B (en) * | 2016-07-21 | 2018-08-07 | 四川大学 | Full-automatic kidney region of interest extraction element and method |
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 |
CN106570867B (en) * | 2016-10-18 | 2019-03-29 | 浙江大学 | Movable contour model image fast segmentation method based on gray scale morphology energy method |
CN106651892A (en) * | 2016-12-21 | 2017-05-10 | 福建师范大学 | Edge detection method |
CN106651892B (en) * | 2016-12-21 | 2019-09-17 | 福建师范大学 | A kind of edge detection method |
CN110264461A (en) * | 2019-06-25 | 2019-09-20 | 南京工程学院 | Microcalciffcation point automatic testing method based on ultrasonic tumor of breast image |
CN110264461B (en) * | 2019-06-25 | 2020-10-27 | 南京工程学院 | Automatic micro-calcification point detection method based on ultrasonic breast tumor image |
CN113324727A (en) * | 2019-07-16 | 2021-08-31 | 中国人民解放军空军工程大学 | Schlieren image processing method for compressed corner supersonic flow field structure |
CN113324727B (en) * | 2019-07-16 | 2023-05-05 | 中国人民解放军空军工程大学 | Schlieren image processing method for compressed corner supersonic flow field structure |
WO2023205896A1 (en) * | 2022-04-28 | 2023-11-02 | Afx Medical Inc. | Systems and methods for detecting structures in 3d images |
Also Published As
Publication number | Publication date |
---|---|
CN102542556B (en) | 2014-10-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102542556B (en) | Method for automatically extracting ultrasonic breast tumor image | |
Dai et al. | A novel approach of lung segmentation on chest CT images using graph cuts | |
CN108053417B (en) | lung segmentation device of 3D U-Net network based on mixed rough segmentation characteristics | |
CN111091573B (en) | CT image pulmonary vessel segmentation method and system based on deep learning | |
CN102682449B (en) | Automatic segmentation method based on self-adaptive external force level set for magnetic resonance images (MRIs) of soft tissue and realization method thereof | |
CN103020969B (en) | A kind of disposal route of CT image liver segmentation and system | |
CN103150723B (en) | The stomach CT image lymph node detection system of Shape-based interpolation and ellipse fitting and method | |
CN111105424A (en) | Lymph node automatic delineation method and device | |
CN105740945A (en) | People counting method based on video analysis | |
CN102360495B (en) | Pulmonary nodule segmentation method based on average intensity projection and translation gaussian model | |
CN103400134B (en) | Non-contact type finger vein sample area-of-interest exacting method | |
CN103714547B (en) | Image registration method combined with edge regions and cross-correlation | |
CN101964118A (en) | Three-dimensional segmentation method for intravascular ultrasound image sequence | |
CN103473767A (en) | Segmentation method and system for abdomen soft tissue nuclear magnetism image | |
CN101826209A (en) | Canny model-based method for segmenting three-dimensional medical image | |
CN103035009A (en) | Pulmonary nodule edge rebuilding and partitioning method based on computed tomography (CT) image | |
CN103093474B (en) | Three-dimensional mammary gland ultrasound image partition method based on homoplasmon and partial energy | |
CN110363775A (en) | A kind of image partition method based on domain type variation level set | |
CN102426699B (en) | Level set synthetic aperture radar (SAR) image segmentation method based on edge and regional information | |
CN101430789B (en) | Image edge detection method based on Fast Slant Stack transformation | |
CN103971369B (en) | The optic disk localization method of retinal images | |
CN102881021B (en) | Based on the aorta petal ultrasonic image division method of probability distribution and continuous max-flow | |
CN102509273B (en) | Tumor segmentation method based on homogeneous pieces and fuzzy measure of breast ultrasound image | |
CN102663749A (en) | Ultrasonic image lesion segmentation technology based on phase information | |
CN104376559A (en) | Medical image segmentation method based on improved range adjustment level set algorithm |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20141008 Termination date: 20191230 |