CN102397070B - Method for fully-automatically segmenting and quantifying left ventricle of cardiac magnetic resonance image - Google Patents
Method for fully-automatically segmenting and quantifying left ventricle of cardiac magnetic resonance image Download PDFInfo
- Publication number
- CN102397070B CN102397070B CN 201110027847 CN201110027847A CN102397070B CN 102397070 B CN102397070 B CN 102397070B CN 201110027847 CN201110027847 CN 201110027847 CN 201110027847 A CN201110027847 A CN 201110027847A CN 102397070 B CN102397070 B CN 102397070B
- Authority
- CN
- China
- Prior art keywords
- left ventricle
- center
- magnetic resonance
- area
- resonance image
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The invention discloses a method for fully-automatically segmenting and quantifying the left ventricle of a cardiac magnetic resonance image, comprising the following specific steps of: carrying out automatic denoising and edge enhancement processing on a 4D (four-dimensional) cardiac magnetic resonance image first; automatically and preprimarily determining the center of the left ventricle by utilizing Hough transform, and implementing a region growing technology by taking the center of the left ventricle as a starting seed point to provide a left ventricle full-voxel blood region, and taking the mass center of the left ventricle full-voxel blood region as a center of the left ventricle in the current layer; finding out the center of each layer by using a seed propagation technology; implementing a region growing technology based on an iterative falling threshold by taking the center of the left ventricle of each layer as a starting seed point to automatically provide a left ventricle blood region of each layer, and calculating the area of the left ventricle blood region of each layer; automatically segmenting the top of the left ventricle according to the time-space continuity of the area of the left ventricle and calculating the area of the top of the left ventricle; positioning the bottom of the left ventricle according to the area and shape time-space continuity of the left ventricle, and automatically segmenting the bottom of the left ventricle of the heart by adopting a region growing technology which is constricted by the left ventricle shape with time-space continuity, and calculating the area of the bottom of the left ventricle; and finally, realizing whole segmentation of the left ventricle image. The method disclosed by the invention is of a fully-automatic process without any manual intervene.
Description
Technical field
The invention belongs to the technical field of nuclear magnetic resonance, concrete refer to a kind of automatic accurate location 4D cardiac magnetic resonance image left ventricle, automatically identify left ventricle bottom and tip position, thereby realize the method for the whole cardiac magnetic resonance image left ventricle of full-automatic dividing quantification.
Background technology
In recent years, heart disease has become the healthy No.1 killer of harm humans.To reduce the deaths from heart disease rate in order improving the quality of living, now to have developed a large amount of technology and be used for the clinical heart disease of diagnosing perspectively and treat.Become the conventional effective means of current clinical diagnosis heart disease, formulation therapeutic scheme by Medical Imaging Technology accurate evaluation heart disease.Nuclear magnetic resonance have not damaged, soft tissue contrast high, afford a wide field of vision, characteristics such as fault imaging arbitrarily, be widely used in clinical diagnose.Show according to the study, the cardiac magnetic resonance imaging precision is high, favorable repeatability, and application clinically is more and more wider, has become the golden standard of estimating cardiac function, estimating cardiac muscle variation, detection myocardial scar and congenital heart disease.Use the cardiac magnetic resonance image not only can observe the morphosis of heart, can also estimate the functional status of heart, can help the doctor to make correct judgement to cardiac structure and function.
it is the pump housing of systemic blood circulation due to left ventricle, Indices of Left Ventricular Function, for example the easypro contracting volume of left ventricle (comprises end-diastolic dimension and end-systolic volume, the basic index of estimating the ventricle morphological function), left ventricle often rich output (pumps through aortic valve from left ventricle and enters aortal blood volume, the important indicator of reflection heart contraction intensity and speed) and left ventricular ejection fraction (left ventricle stroke volume and end-diastolic volume ratio, one of important indicator of assess cardiac pumping function), it is the important references of clinical diagnosis heart disease and curative effect, quantizing Indices of Left Ventricular Function also just becomes the conventional means of clinical diagnosis and the employing for the treatment of heart disease.
Cutting apart left ventricle is the prerequisite that quantizes Indices of Left Ventricular Function.In the clinical practice of standard, ventricle is cut apart all by experienced doctor's hand drawing.Yet the clinical image data volume is very large, manually cuts apart generally only to diastasis and end-systolic image, and has subjective differences when the myocardial structural of describing as complexity such as girder flesh, papillary muscless.This shows, cut apart ventricle by hand very consuming time, inefficiency, labor intensity is large, and is repeatable poor.Therefore, automatical and efficient Accurate Segmentation quantify left ventricle be always current research emphasis and focus.
Up to now, the left ventricle partitioning algorithm has developed to get comparative maturity.These algorithms comprise traditional edge extracting, region growing, and some are based on the method for particular theory, as level set algorithm, genetic algorithm etc.Yet, these algorithms are nearly all only effective when cutting apart the left ventricle of being surrounded fully by cardiac muscle, roughly also there is the following problem: can not automatic precision truly have effect location left ventricle, can not automatically identify bottom and the top of left ventricle, can not automatically extract left ventricle bottom blood volume.That is to say, before cutting apart quantify left ventricular function parameter, still need to manually locate left ventricle, manually determine the bottom of left ventricle and the position at top, manual closed bottom left ventricle etc. also need the bad image of some segmentation effects of manual correction simultaneously.In addition, although used some business softwares to replace traditional purely manually cut apart in clinical, due to the limitation that is subject to algorithm, still need a large amount of manual interventions just can complete the functional parameter quantification of left ventricle.Therefore, still be badly in need of clinically accuracy and repeatability that a kind of reliable and effective full-automatic method further improves heart function parameter, increase work efficiency, alleviate working strength.
Summary of the invention
The objective of the invention is for above-mentioned the deficiencies in the prior art part, a kind of automatic location 4D cardiac magnetic resonance image left ventricle is provided, automatically identifies bottom left ventricle and tip position, thereby realize that full-automatic dividing quantizes the method for cardiac magnetic resonance image left ventricle.
Realization of the present invention is completed by following technical scheme:
A kind of full-automatic dividing quantizes the method for cardiac magnetic resonance image left ventricle, it is characterized in that, the concrete implementation step of the method is:
(1) 4D cardiac magnetic resonance image is carried out automatic denoising and edge enhancement process;
(2) tentatively determine the left ventricle center, implement region growth technique take this center as initial seed point and propose all plain blood regions of left ventricle and calculate this regional barycenter, and it is defined as current residual aspect left ventricle center; Enforcement seminal propagation technology estimation residue aspect left ventricle center;
(3) the left ventricle center of determining in the step (2) is as initial seed points, adopts the region growth technique based on the iteration falling-threshold value automatically propose territory, left ventricular blood liquid zone and calculate respective area to left ventricle middle part aspect;
(4) the left ventricle center of determining in the step (2) is as initial seed points, middle part aspect employing is automatically extracted each tomographic image in the blood regions that is sometimes gone up mutually and calculates respective area based on the region growth technique of iteration falling-threshold value to left ventricle, automatically locate the heart left ventricle top according to the space and time continuous of extracting the region area variation, and proofread and correct the area at estimation left ventricle top;
(5) according to left ventricle area and left ventricle bottom, shape space and time continuous location, utilize the region growth technique auto Segmentation heart left ventricle bottom that is subjected to shape constraining and automatic calibration is carried out in the zone that is partitioned into;
(6) calculate Indices of Left Ventricular Function according to the area of left ventricle: comprise left ventricle relax contracting volume, often win output and ejection fraction, and describe to fill curve.
Preferably, in described step (1), anisotropic diffusion method is adopted in the processing of 4D cardiac magnetic resonance image.
Preferably, in described step (2) before proposing all plain blood regions of left ventricle, to left ventricle middle part aspect diastasis and end-systole subtraction image carrying out the Hough conversion.
Preferably, seminal propagation technology in described step (2) refers to for the residue aspect, with the center of the contiguous aspect of determining the center of the residue aspect initial seed point as this residue aspect, implement region growth technique extraction whole blood voxel left ventricle regional, and this regional barycenter is defined as the left ventricle center of current residual aspect.
Preferably, adopt in described step (4) first from left ventricle middle part aspect and carry out left ventricle center breeding along the left ventricle top-direction, carry out take this center as initial seed points and automatically extract each tomographic image in the blood regions that is sometimes gone up mutually based on the region growth technique of iteration falling-threshold value method.
Described falling-threshold value can be obtained by following formula:
th=u
br
In above-mentioned formula, μ
bBe the average of blood regions, r is a variable, and initial value is 1.0, increase progressively with step-length 0.05 in iterative process, thereby threshold value th descends in the process of iteration.
Preferably, working as certain one deck in described step (4) has part phase place growth district area generation transition, adopts following formula estimated area according to time continuity:
Wherein, s represents the s layer, and the time phase of transition occurs for p representative, and q represents from the nearest time phase that area transition does not occur of p, and ms represents the number of plies of left ventricle intermediate surface.
Preferably, in described step (4), transition all occurs in all phase place growth district areas of certain one deck,
The ventricle top is considered as circular cone or elliptic cone, adopts following formula estimated area according to spatial continuity:
Wherein, s represents the s layer, and the time phase of transition occurs in the p representative.
Preferably, adopt in described step (5) first from left ventricle middle part aspect and carry out left ventricle center breeding along the left ventricle bottom direction, carry out take this center as initial seed points and automatically extract each tomographic image in the blood regions that is sometimes gone up mutually based on the region growth technique of iteration falling-threshold value method.
Advantage of the present invention is, the inventive method can realize fast accurately that the automatic location of the whole left ventricle part of cardiac magnetic resonance image cuts apart.Compare with current existing method, the present invention adopts full automatic auto Segmentation quantization method, can automatically complete the cutting apart and the calculating of the functional parameter of left ventricle of left ventricle, without any need for manual intervention, functional parameter all comprises, the left ventricle contracting volume that relaxes, ejection fraction, often rich output, full curve, because former method is all to need the people to cut apart base of heart and top by hand for going, have very strong subjectivity, the result that different people obtains has very large diversity.
Description of drawings
Fig. 1 is a curve chart, shows the change in volume of extracting the zone;
Fig. 2 is a left ventricle middle part original image and cut apart image (left be original image, right for cutting apart image) in embodiment;
Fig. 3 is area and the variation (left is spatial continuity, the right side be time continuity) of shape on space-time of left ventricle in embodiment;
Fig. 4 is Zhang San's dimension curve figure, shows left ventricle area space and time continuous;
Fig. 5 is a left ventricle top original image and cut apart image (left be original image, right for cutting apart image) in embodiment;
Fig. 6 is a left ventricle bottom original image and cut apart image (left be original image, right for cutting apart image) in embodiment;
Fig. 7 is left ventricular filling curve in embodiment (phase when transverse axis is left ventricle, the longitudinal axis are left ventricular volume).
The specific embodiment
Feature of the present invention and other correlated characteristic are described in further detail by embodiment below in conjunction with accompanying drawing:
The inventive method accurately automatically quantizes 4D cardiac magnetic resonance image (CMRI) Indices of Left Ventricular Function efficiently, without any need for manual intervention.Following instance step-by-step instructions the inventive method is located left ventricle automatically, automatically determines left ventricle top and bottom position, the specific operation process of auto Segmentation quantify left ventricle.
The magnetic resonance imaging data of the present embodiment collection is the cardiac magnetic resonance imaging data.Data from GE Signa1.5T magnetic resonance imaging system, selected imaging sequence is the SSFP sequence.Concrete imaging parameters: TR3.3-4.5ms, TE1.1-2.0ms, flip angle 55-60, matrix size 192 * 192 256 * 256, image size 256 * 256 receives bandwidth 125kHz, the visual field (FOV) 290-400 * 240-360, bed thickness and interlamellar spacing are respectively 6-8mm and 2-4mm (10mm altogether), and the left ventricle of each data has the 6-10 layer, phase during the 20-28 heart.
(1) iteration falling-threshold value method is extracted blood
At first begin execution area growth technology from seed points and extract the whole blood voxel, calculate average and the standard deviation (μ in whole blood sample zone
bAnd σ
b).Then with μ
bCarry out a series of region growth technique based on continuous falling-threshold value (th) for initial threshold and extract a series of growth districts until should break through away from cardiac muscle suddenly in the zone.Falling-threshold value can be by μ continuously
bObtain with a variable, specifically adopt following formula:
th=u
b/r
①
In above-mentioned formula, r is a variable, and initial value is 1.0, increase progressively with step-length 0.05 in iterative process, thereby threshold value th descends in the process of iteration.Accompanying drawing 1(a) having described to extract regional area from the angle of region area changes until sudden change along with the threshold value of continuous reduction is continuous.Transverse axis represents variable r, and the longitudinal axis represents area (representing with the voxel number).According to threshold value and the σ before the volume sudden change
bCan find one just when being used for cutting apart left ventricle.
Fig. 2 has shown a left ventricle middle part original image in embodiment and has cut apart image based on what continuous falling-threshold value method was obtained.
(2) cardiac magnetic resonance image left ventricle center breeding
In heart middle part part, as starting point, current aspect is carried out region growth technique take adjacent layer left ventricle center, the barycenter of growth district is defined as the center of current aspect left ventricle; Be the center of current aspect left ventricle take the center of contiguous aspect left ventricle separately in the top of heart and bottom.
(3) the left ventricle top is cut apart and the area estimation
The left ventricle area is continually varying in time and space.Fig. 3 has illustrated area and the seriality of shape on space-time of left ventricle.Fig. 4 has shown space and time continuous with the left ventricular side product of embodiment value.As shown in Figure 4, the number of phases when y axle represents, the number of data layers that the representative of x axle collects, the z axle represents the growth district area.Phase when this embodiment one has 12 layers 20, in figure, 12 batten curves represent 12 layers, each represents the growth district area variation of phase at any time that every one deck extracts.Cut apart automatically the heart top according to the area change space and time continuous of extracting the zone, and estimate the area of these positions according to the situation that transition occurs at the top.Two kinds of situations are arranged:
(a) certain one deck has part phase place growth district area generation transition, adopts following formula estimated area according to time continuity:
②
Wherein, s represents the s layer, and the time phase of transition occurs for p representative, and q represents from the nearest time phase that area transition does not occur of p, and ms represents the number of plies of left ventricle intermediate surface.
(b) transition all occurs in all phase place growth district areas of certain one deck, the ventricle top is considered as circular cone or ellipse is sewed, and adopts following formula estimated area according to spatial continuity:
③
Here, s represents the s layer, and the time phase of transition occurs in the p representative.
Fig. 5 has shown a left ventricle top original image in embodiment and has cut apart image according to what space and time continuous obtained.
(4) location, left ventricle bottom and estimation
The bottom position that the area change regional according to extraction and gravity center shift time space seriality are determined left ventricle.And utilize the region growth technique auto Segmentation heart left ventricle bottom that is subjected to shape constraining and the zone that is partitioned into is proofreaied and correct.Fig. 6 has shown a left ventricle bottom original image in embodiment and has cut apart image according to what space and time continuous obtained.
(5) left ventricular filling curve
After completing the full-automatic dividing left ventricle, can quantize the left ventricular function parameter according to the area that is partitioned into the zone, comprise left ventricle relax contracting volume, often win output and ejection fraction, and describe to fill curve.Fig. 7 has described embodiment left ventricular filling curve, phase when transverse axis is, and the longitudinal axis is left ventricular mass.
Above-described is only the preferred embodiment of the present invention; should be pointed out that for the person of ordinary skill of the art, under the prerequisite that does not break away from the invention design; can also make some similar distortion and improvement, within these also should be considered as protection scope of the present invention.
Claims (8)
1. the method for a full-automatic dividing quantification cardiac magnetic resonance image left ventricle, is characterized in that, the concrete implementation step of the method is:
(1) 4D cardiac magnetic resonance image is carried out automatic denoising and edge enhancement process;
(2) tentatively determine the left ventricle center, implement region growth technique take this center as initial seed point and propose all plain blood regions of left ventricle and calculate this regional barycenter, and barycenter that should the zone is defined as current aspect left ventricle center; Enforcement seminal propagation technology estimation residue aspect left ventricle center;
(3) the left ventricle center of determining in the step (2) is as initial seed points, adopts the region growth technique based on the iteration falling-threshold value automatically propose territory, left ventricular blood liquid zone and calculate respective area to left ventricle middle part aspect;
(4) the left ventricle center of determining in the step (2) is as initial seed points, middle part aspect employing is automatically extracted each tomographic image in the blood regions that is sometimes gone up mutually and calculates respective area based on the region growth technique of iteration falling-threshold value to left ventricle, automatically locate the heart left ventricle top according to the space and time continuous of extracting the region area variation, and proofread and correct the area at estimation left ventricle top;
(5) according to left ventricle area and left ventricle bottom, shape space and time continuous location, utilize the region growth technique auto Segmentation heart left ventricle bottom that is subjected to shape constraining and automatic calibration is carried out in the zone that is partitioned into;
(6) calculate Indices of Left Ventricular Function according to the area of left ventricle: comprise left ventricle relax contracting volume, often win output and ejection fraction, and describe to fill curve.
2. a kind of full-automatic dividing according to claim 1 quantizes the method for cardiac magnetic resonance image left ventricle, it is characterized in that, in described step (1), anisotropic diffusion method is adopted in the processing of 4D cardiac magnetic resonance image.
3. a kind of full-automatic dividing according to claim 2 quantizes the method for cardiac magnetic resonance image left ventricle, it is characterized in that, in described step (2) before proposing all plain blood regions of left ventricle, to left ventricle middle part aspect diastasis and end-systole subtraction image carrying out the Hough conversion.
4. a kind of full-automatic dividing according to claim 1 quantizes the method for cardiac magnetic resonance image left ventricle, it is characterized in that, seminal propagation technology in described step (2) refers to for the residue aspect, with the center of the contiguous aspect of determining the center of the residue aspect initial seed point as this residue aspect, implement region growth technique extraction whole blood voxel left ventricle regional, and this regional barycenter is defined as the left ventricle center of current residual aspect.
5. a kind of full-automatic dividing according to claim 4 quantizes the method for cardiac magnetic resonance image left ventricle, it is characterized in that, adopt in described step (4) first from left ventricle middle part aspect and carry out left ventricle center breeding along the left ventricle top-direction, carry out take this center as initial seed points and automatically extract each tomographic image in the blood regions that is sometimes gone up mutually based on the region growth technique of iteration falling-threshold value method;
Described falling-threshold value can be obtained by following formula:
th＝u
b/r
In above-mentioned formula, μ
bBe the average of blood regions, r is a variable, and initial value is 1.0, increase progressively with step-length 0.05 in iterative process, thereby threshold value th descends in the process of iteration.
6. a kind of full-automatic dividing according to claim 5 quantizes the method for cardiac magnetic resonance image left ventricle, it is characterized in that, working as certain one deck in described step (4) has part phase place growth district area generation transition, adopts following formula estimated area according to time continuity:
Wherein, s represents the s layer, and the time phase of transition occurs for p representative, and q represents from the nearest time phase that area transition does not occur of p, and ms represents the number of plies of left ventricle intermediate surface.
7. a kind of full-automatic dividing according to claim 5 quantizes the method for cardiac magnetic resonance image left ventricle, it is characterized in that, in described step (4), transition all occurs in all phase place growth district areas of certain one deck, the ventricle top is considered as circular cone or elliptic cone, adopts following formula estimated area according to spatial continuity:
Wherein, s represents the s layer, and the time phase of transition occurs in the p representative.
8. according to claim 6 or 7 described a kind of full-automatic dividings quantize the method for cardiac magnetic resonance image left ventricle, it is characterized in that, adopt in described step (5) first from left ventricle middle part aspect and carry out left ventricle center breeding along the left ventricle bottom direction, carry out take this center as initial seed points and automatically extract each tomographic image in the blood regions that is sometimes gone up mutually based on the region growth technique of iteration falling-threshold value method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110027847 CN102397070B (en) | 2011-01-26 | 2011-01-26 | Method for fully-automatically segmenting and quantifying left ventricle of cardiac magnetic resonance image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110027847 CN102397070B (en) | 2011-01-26 | 2011-01-26 | Method for fully-automatically segmenting and quantifying left ventricle of cardiac magnetic resonance image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102397070A CN102397070A (en) | 2012-04-04 |
| CN102397070B true CN102397070B (en) | 2013-06-19 |
Family
ID=45879982
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201110027847 Expired - Fee Related CN102397070B (en) | 2011-01-26 | 2011-01-26 | Method for fully-automatically segmenting and quantifying left ventricle of cardiac magnetic resonance image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102397070B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6001280B2 (en) * | 2012-03-09 | 2016-10-05 | 学校法人金沢医科大学 | Method for manufacturing heart correction net |
| CN103810688B (en) * | 2012-11-06 | 2017-08-11 | 上海联影医疗科技有限公司 | A kind of automatic Partitioning method of left ventricle |
| CN103908279B (en) * | 2013-01-07 | 2017-07-28 | 上海联影医疗科技有限公司 | A kind of recognition methods of heart state time phase |
| CN103400398B (en) * | 2013-07-31 | 2015-11-04 | 东北大学 | A kind of heart automatic positioning method based on chest spacer |
| CN103854287B (en) * | 2014-03-11 | 2017-06-06 | 深圳市旭东数字医学影像技术有限公司 | The method and its device of the meniscus segmentation based on MRI |
| CN104200465B (en) | 2014-08-20 | 2017-08-25 | 沈阳东软医疗系统有限公司 | The dividing method and device of cardiac three-dimensional image |
| US9949643B2 (en) | 2014-10-18 | 2018-04-24 | International Business Machines Corporation | Automatic visualization of regional functional parameters of left ventricle from cardiac imaging |
| CN105741310B (en) * | 2016-03-21 | 2018-05-04 | 东北大学 | A kind of cardiac left-ventricle image segmenting system and method |
| CN106709930B (en) | 2016-12-29 | 2020-03-31 | 上海联影医疗科技有限公司 | Method and device for segmenting interested volume of three-dimensional medical image |
| CN110766693B (en) * | 2018-09-06 | 2022-06-21 | 北京连心医疗科技有限公司 | Method for jointly predicting radiotherapy structure position based on multi-model neural network |
| CN109741395B (en) * | 2018-12-14 | 2021-07-23 | 北京市商汤科技开发有限公司 | Dual-chamber quantification method and device, electronic equipment and storage medium |
| CN110322414B (en) * | 2019-07-05 | 2021-08-10 | 北京探境科技有限公司 | Image data online quantitative correction method and system based on AI processor |
| CN110731777B (en) * | 2019-09-16 | 2023-07-25 | 平安科技（深圳）有限公司 | Left ventricle measurement method and device based on image recognition and computer equipment |
| CN112308845B (en) * | 2020-11-03 | 2021-07-02 | 赛诺威盛科技(北京)股份有限公司 | Left ventricle segmentation method and device and electronic equipment |
| CN112881467B (en) * | 2021-03-15 | 2023-04-28 | 中国空气动力研究与发展中心超高速空气动力研究所 | Large-size composite material damage imaging and quantitative identification method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101596111A (en) * | 2008-05-23 | 2009-12-09 | 西门子公司 | In the nuclear magnetic resonance of heart film to the automatic location of left ventricle |
| CN101849240A (en) * | 2007-05-21 | 2010-09-29 | 康奈尔大学 | Method for segmenting objects in images |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006102125A (en) * | 2004-10-05 | 2006-04-20 | Toshiba Corp | Method for identifying region of interest with time in medical image processing apparatus and medical image processing apparatus |
| US20100215238A1 (en) * | 2009-02-23 | 2010-08-26 | Yingli Lu | Method for Automatic Segmentation of Images |
-
2011
- 2011-01-26 CN CN 201110027847 patent/CN102397070B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101849240A (en) * | 2007-05-21 | 2010-09-29 | 康奈尔大学 | Method for segmenting objects in images |
| CN101596111A (en) * | 2008-05-23 | 2009-12-09 | 西门子公司 | In the nuclear magnetic resonance of heart film to the automatic location of left ventricle |
Non-Patent Citations (1)
| Title |
|---|
| JP特开2006-102125A 2006.04.20 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102397070A (en) | 2012-04-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102397070B (en) | Method for fully-automatically segmenting and quantifying left ventricle of cardiac magnetic resonance image | |
| CN111798462B (en) | Automatic delineation method of nasopharyngeal carcinoma radiotherapy target area based on CT image | |
| CN104992430B (en) | Full automatic three-dimensional liver segmentation method based on convolutional neural networks | |
| CN109584254A (en) | A kind of heart left ventricle's dividing method based on the full convolutional neural networks of deep layer | |
| CN110047082B (en) | Deep learning-based pancreatic neuroendocrine tumor automatic segmentation method and system | |
| CN102048550B (en) | Method for automatically generating liver 3D (three-dimensional) image and accurately positioning liver vascular domination region | |
| CN107578413B (en) | Method, apparatus, equipment and the readable storage medium storing program for executing of retinal images layering | |
| CN110599499B (en) | MRI image heart structure segmentation method based on multipath convolutional neural network | |
| CN110874860A (en) | Target extraction method of symmetric supervision model based on mixed loss function | |
| JP2019532789A (en) | Mapping of mammary artery calcification | |
| CN105139377A (en) | Rapid robustness auto-partitioning method for abdomen computed tomography (CT) sequence image of liver | |
| CN110516759A (en) | A kind of soft tissue sarcoma based on machine learning shifts risk forecasting system | |
| CN103914845B (en) | The method obtaining initial profile in Ultrasound Image Segmentation based on active contour model | |
| Cristobal-Huerta et al. | Automated quantification of epicardial adipose tissue in cardiac magnetic resonance imaging | |
| CN104517315A (en) | Method and system for reconstructing bilateral ureters based on interactive region growing method | |
| Mosbech et al. | Automatic segmentation of abdominal adipose tissue in MRI | |
| CN110136139A (en) | Dental nerve dividing method in facial CT image based on shape feature | |
| CN112950611A (en) | Liver blood vessel segmentation method based on CT image | |
| CN109544565A (en) | A kind of surface of a wound image partition method based on DC-GMRF level set | |
| CN110060234B (en) | Method and device for calculating kidney GFR value based on SPECT (single photon emission computed tomography) tomographic image | |
| CN115762787B (en) | Eyelid disease operation curative effect evaluation method and system | |
| CN111429404A (en) | Imaging system and method for detecting cardiovascular and cerebrovascular vessels | |
| CN110610502A (en) | Automatic aortic arch region positioning and segmentation method based on CT image | |
| CN104835148B (en) | A kind of automatic division method of retina cryptomere oedema | |
| Supriyanto et al. | Automatic non invasive kidney volume measurement based on ultrasound image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| ASS | Succession or assignment of patent right |
Owner name: JIAXING YIFU SOFTWARE CO., LTD. Free format text: FORMER OWNER: WUXI ZHOUSHI MEDICAL SOFTWARE DEVELOPMENT CO., LTD. Effective date: 20130417 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| COR | Change of bibliographic data |
Free format text: CORRECT: ADDRESS; FROM: 214213 WUXI, JIANGSU PROVINCE TO: 314000 JIAXING, ZHEJIANG PROVINCE |
|
| TA01 | Transfer of patent application right |
Effective date of registration: 20130417 Address after: Jiaxing City, Zhejiang province 314000 South Road No. 1369 Building No. 8 Room 309 Branch Center Applicant after: Wuxi Zhoushi Medical Software Development Co.,Ltd. Applicant after: East China Normal University Address before: 925 room 8, R & D building, Wen Zhuang Road, Yixing Economic Development Zone, Jiangsu 214213, China Applicant before: Wuxi Zhoushi Medical Software Development Co.,Ltd. Applicant before: East China Normal University |
|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130619 Termination date: 20140126 |