CN113379682B - Heart MRI image coupling level set segmentation method and system - Google Patents

Heart MRI image coupling level set segmentation method and system Download PDF

Info

Publication number
CN113379682B
CN113379682B CN202110557989.6A CN202110557989A CN113379682B CN 113379682 B CN113379682 B CN 113379682B CN 202110557989 A CN202110557989 A CN 202110557989A CN 113379682 B CN113379682 B CN 113379682B
Authority
CN
China
Prior art keywords
segmentation
level set
myocardial
coupling level
constraint
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.)
Active
Application number
CN202110557989.6A
Other languages
Chinese (zh)
Other versions
CN113379682A (en
Inventor
杨聪
马建红
王瑞娟
朱英琳
赵博学
郭朔
邓奎
李双
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhengzhou University
Original Assignee
Zhengzhou University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Zhengzhou University filed Critical Zhengzhou University
Priority to CN202110557989.6A priority Critical patent/CN113379682B/en
Publication of CN113379682A publication Critical patent/CN113379682A/en
Application granted granted Critical
Publication of CN113379682B publication Critical patent/CN113379682B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • 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
    • G06T2207/20—Special algorithmic details
    • G06T2207/20081—Training; Learning
    • 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
    • G06T2207/20—Special algorithmic details
    • G06T2207/20084—Artificial neural networks [ANN]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30048—Heart; Cardiac

Abstract

The invention provides a heart MRI image coupling level set segmentation method and a system, wherein the method comprises a rough segmentation stage of an inner membrane and an outer membrane of a myocardium based on a convolutional neural network, and a fine segmentation stage of the inner membrane and the outer membrane of the myocardium based on a coupling level set method; in the rough segmentation stage, a U-Net frame is adopted to respectively construct segmentation models of the myocardial inner membrane and the myocardial outer membrane; in the fine segmentation stage, based on a coupling level set model of distance smooth change constraint, a constraint condition of distance smooth change is set, and the contours of the myocardial intima and the myocardial adventitia are respectively extracted by using two level set functions in combination with the length, rule constraint and balloon force term of a level set. According to the method, the deep learning model is used, and a level set method based on distance constraint is combined, so that automatic and accurate segmentation of the cardiac MRI image can be realized, the time for a clinician to manually segment the image is saved, and the image segmentation efficiency is greatly improved.

Description

Heart MRI image coupling level set segmentation method and system
Technical Field
The invention belongs to the technical field of medical image processing, and particularly relates to a method and a system for segmenting a coupling level set of a cardiac MRI image.
Background
The segmentation of the cardiac magnetic resonance image is a precondition for realizing quantitative calculation of cardiac function indexes, and the segmentation of the cardiac MRI image mainly has the following difficulties: in the basal plane section, a part of the myocardial contour disappears under the influence of the left ventricular outflow tract; the tissues such as the trabecular spine, the papillary muscles and the like in the ventricular blood pool have similar gray distribution with the cardiac muscle, so that the outline of the cardiac muscle intima is fuzzy; in the top plane slice, the endocardial contour of the myocardium is very blurred due to the small area of the left ventricle, which is also affected by the partial volume effect. Currently, the mainstream cardiac MRI image segmentation algorithms include algorithms based on dynamic programming, algorithms based on image registration, algorithms based on graph theory, and algorithms based on active contour models. However, the existing cardiac MRI image segmentation scheme algorithm usually extracts cardiac tissues separately, neglects the spatial position relationship between cardiac tissue structures, and leads the segmentation result to deviate from the anatomical structure.
Disclosure of Invention
In order to solve the above problem, it is necessary to provide a cardiac MRI image coupling level set segmentation method.
The invention provides a heart MRI image coupling level set segmentation method, which comprises a rough segmentation stage of an inner membrane and an outer membrane of a heart muscle based on a convolutional neural network and a fine segmentation stage of the inner membrane and the outer membrane of the heart muscle based on a coupling level set method;
in the rough segmentation stage, a U-Net frame is adopted to respectively construct segmentation models of the myocardial intima and the myocardial adventitia; in the fine segmentation stage, based on a coupling level set model of distance smooth change constraint, a constraint condition of distance smooth change is set, and the contours of the myocardial intima and the myocardial adventitia are respectively extracted by using two level set functions in combination with the length, rule constraint and balloon force term of a level set.
Based on the above, when modeling the endocardium and the epicardium by using U-Net, the U-Net comprises two parts of down sampling and up sampling;
the up-sampling process and the down-sampling process are symmetrical, and 3 times of sampling operation is carried out on the characteristic diagram respectively, wherein 2 times of convolution operation is carried out before each down-sampling.
Based on the above, the loss function of the U-Net model is constructed based on the Dice coefficient, which is defined as the formula:
Figure BDA0003078024760000021
wherein T is the real contour of the segmentation target, P is the prediction result of the segmentation target, and T, P belongs to [0,1 ]] w×h W and h represent the size of the segmented image;
subtracting the Dice coefficient from 1 as a loss function of the U-Net model, and defining the loss function as follows: loss Dice (T, P) =1-Dice (T, P), where Dice (T, P) represents the Dice coefficient values of the segmentation result P and the true contour T.
Based on the above, let phi i (x, y) =0 and phi o (x, y) =0 zero level sets of the two level set functions represent endocardium and adventitia of myocardium, respectively, E ci For the distance-smoothly varying constraint, λ i ,α i ,μ i ,β i Is a length term E L Balloon force E A Item and rule item E R The coefficients of (c);
for myocardial intimal segmentation, the coupled level set model formula based on distance smooth change constraint is defined as:
E i (φ i ；φ o )＝λ i E L (φ i )+α i E A (φ i )+μ i E R (φ i )+β i E Ci (φ i ；φ o )；
wherein, the first and the second end of the pipe are connected with each other,
Figure BDA0003078024760000022
Figure BDA0003078024760000023
Figure BDA0003078024760000031
Figure BDA0003078024760000032
δ (x) is the derivative of the Heaviside function;
the coupling level set segmentation function for the endocardium of the myocardium was obtained as:
Figure BDA0003078024760000033
similarly, the coupling level set segmentation function of the myocardial adventitia is obtained as follows:
Figure BDA0003078024760000034
the invention provides a heart MRI image coupling level set segmentation system in a second aspect, which comprises a rough segmentation unit and a fine segmentation unit;
the rough segmentation unit adopts a U-Net frame to respectively construct segmentation models of the myocardial inner membrane and the myocardial outer membrane, and carries out rough segmentation on the myocardial inner membrane and the myocardial outer membrane;
the precise segmentation unit sets constraint conditions of distance smooth change based on a coupling level set model of distance smooth change constraint, combines the length of a level set, rule constraint and a balloon force term, and respectively extracts the contours of the endocardium and the epicardium of the myocardium by using two level set functions to realize precise segmentation of the endocardium and the epicardium of the myocardium.
A third aspect of the present invention provides a terminal, including a processor, a memory, and a cardiac MRI image coupling level set segmentation method program stored in the memory, where when the cardiac MRI image coupling level set segmentation method program is executed by the processor, the steps of the cardiac MRI image coupling level set segmentation method are implemented.
A fourth aspect of the invention provides a computer readable storage medium having stored thereon computer instructions which, when executed by a processor, carry out the steps of the cardiac MRI image coupling level set segmentation method as described.
Compared with the prior art, the method has outstanding substantive characteristics and remarkable progress, and particularly, the method can realize automatic and accurate segmentation of the cardiac MRI image by using a deep learning model and combining a level set method based on distance constraint, save the time for a clinician to manually segment the image and greatly improve the efficiency of image segmentation.
Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a block flow diagram of a cardiac MRI image coupling level set segmentation method of the present invention.
Fig. 2 is a situation that often occurs in the conventional coarse segmentation stage.
Fig. 3 is a diagram of the actual segmentation effect of the present invention.
Detailed Description
In order that the above objects, features and advantages of the present invention can be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings. It should be noted that the embodiments and features of the embodiments of the present application may be combined with each other without conflict.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those specifically described herein, and therefore the scope of the present invention is not limited by the specific embodiments disclosed below.
Example 1
As shown in fig. 1, the present embodiment proposes a cardiac MRI image coupling level set segmentation method, which includes a coarse segmentation stage of the endocardium and the adventitia of the myocardium based on a convolutional neural network, and a fine segmentation stage of the endocardium and the adventitia of the myocardium based on a coupling level set method;
in the rough segmentation stage, a U-Net frame is adopted to respectively construct segmentation models of the myocardial inner membrane and the myocardial outer membrane; in the fine segmentation stage, based on a coupling level set model of distance smooth change constraint, a constraint condition of distance smooth change is set, and the contours of the myocardial intima and the myocardial adventitia are respectively extracted by using two level set functions in combination with the length, rule constraint and balloon force term of a level set.
Specifically, when modeling the inner membrane and the outer membrane of the myocardium by using U-Net, the U-Net comprises a down-sampling part and an up-sampling part;
the up-sampling process and the down-sampling process are symmetrical, and 3 times of sampling operation is carried out on the characteristic diagram respectively, wherein 2 times of convolution operation is carried out before each down-sampling.
The loss function of the U-Net model is constructed based on a Dice coefficient, and the Dice coefficient is defined as a formula:
Figure BDA0003078024760000051
wherein T is the real contour of the segmentation target, P is the prediction result of the segmentation target, and T, P belongs to [0,1 ]] w×h W and h represent the size of the segmented image;
when the Dice coefficient is used as the loss function, in order to minimize the loss function, 1 minus the Dice coefficient is defined as the loss function of the U-Net model: loss Dice (T, P) =1-Dice (T, P), where Dice (T, P) represents the Dice coefficient values of the segmentation result P and the true contour T. It can be seen that the higher the Dice value of the whole segmentation result, the smaller the loss value of the whole function is, and finally the purpose of feeding back and updating the whole model parameter is achieved.
Let phi i (x, y) =0 and phi o (x, y) =0 zero level sets of the two level set functions represent endocardium and adventitia of myocardium, respectively, E ci For the distance-smoothly varying constraint, λ i ,α i ,μ i ,β i Is a length term E L Balloon force E A Item and rule item E R The coefficients of (c);
for myocardial intimal segmentation, the coupled level set model formula based on distance smooth change constraint is defined as:
E i (φ i ；φ o )＝λ i E L (φ i )+α i E A (φ i )+μ i E R (φ i )+β i E Ci (φ i ；φ o )；
wherein the content of the first and second substances,
Figure BDA0003078024760000061
Figure BDA0003078024760000062
Figure BDA0003078024760000063
Figure BDA0003078024760000064
δ (x) is the derivative of the Heaviside function;
the coupling level set segmentation function for the endocardium of the myocardium was obtained as:
Figure BDA0003078024760000065
similarly, the coupling level set segmentation function of the myocardial adventitia is obtained as follows:
Figure BDA0003078024760000066
it should be noted that the invention adopts a U-Net frame to construct a segmentation model of the myocardial intima and adventitia, and then adopts a level set model based on distance smooth change constraint to optimize a coarse segmentation result, so that the model based on the convolutional neural network needs to go through conventional stages such as training, verification and testing. Data can be collected from a network public data set, preprocessing including image scaling, clipping and label conversion is performed on an original image, then training, verification and testing are performed, and finally a segmentation function of the myocardial intima and the myocardial adventitia is obtained, and the training, verification and testing processes are not described again.
Results of the experiment
As shown in fig. 2, the black solid line simulates the endocardium and the white solid line simulates the epicardium. The first row of simulated partial endocardium is outside the adventitia and the second row of simulated partial epicardium is further from the intima. Neither of these situations fits the heart anatomy, but often occurs during the coarse segmentation phase. Fig. 3 is a practical segmentation result of the present invention, and it can be seen that the level set method based on distance coupling can better correct the curves of the deviated portions in both cases.
Example 2
The embodiment provides a cardiac MRI image coupling level set segmentation system, which comprises a coarse segmentation unit and a fine segmentation unit;
the rough segmentation unit adopts a U-Net frame to respectively construct segmentation models of the myocardial inner membrane and the myocardial outer membrane, and carries out rough segmentation on the myocardial inner membrane and the myocardial outer membrane;
the precise segmentation unit sets constraint conditions of distance smooth change based on a coupling level set model of distance smooth change constraint, combines the length of a level set, rule constraint and a balloon force term, and respectively extracts the contours of the endocardium and the epicardium of the myocardium by using two level set functions to realize precise segmentation of the endocardium and the epicardium of the myocardium.
It should be noted that, for convenience and simplicity of description, the specific working process of the cardiac MRI image coupling level set segmentation system described above may refer to the corresponding process of the method described above, and is not described herein again.
Example 3
This embodiment provides a terminal, including a processor, a memory, and a cardiac MRI image coupling level set segmentation method program stored in the memory, where the cardiac MRI image coupling level set segmentation method program, when executed by the processor, implements the steps of the cardiac MRI image coupling level set segmentation method according to embodiment 1.
Example 4
The present embodiment provides a computer readable storage medium having stored thereon computer instructions which, when executed by a processor, implement the steps of the cardiac MRI image coupling level set segmentation method according to embodiment 1.
In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to the related descriptions of other embodiments for parts that are not described or illustrated in a certain embodiment.
Those of ordinary skill in the art will appreciate that the various illustrative modules and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
In the embodiments provided in the present application, it should be understood that the disclosed system and method may be implemented in other ways. For example, the above-described system embodiments are merely illustrative, and for example, the division of the above-described units is only one logical functional division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated module may be stored in a computer-readable storage medium if it is implemented in the form of a software functional unit and sold or used as a separate product. Based on such understanding, all or part of the flow in the method of the embodiments described above may be implemented by a computer program, which may be stored in a computer-readable storage medium and can implement the steps of the embodiments of the methods described above when the computer program is executed by a processor. The computer program includes computer program code, and the computer program code may be in a source code form, an object code form, an executable file or some intermediate form.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily think of the changes or substitutions within the technical scope of the present invention, and shall cover the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.

Claims (6)

1. A heart MRI image coupling level set segmentation method is characterized in that:
the method comprises a rough segmentation stage of the endocardium and the adventitia of the myocardium based on a convolutional neural network and a fine segmentation stage of the endocardium and the adventitia of the myocardium based on a coupling level set method;
in the rough segmentation stage, a U-Net frame is adopted to respectively construct segmentation models of the myocardial inner membrane and the myocardial outer membrane; in the fine segmentation stage, based on a coupling level set model of distance smooth change constraint, setting constraint conditions of distance smooth change, and extracting the contours of the myocardial intima and the myocardial adventitia by using two level set functions in combination with the length, rule constraint and balloon force term of a level set;
the loss function of the U-Net model is constructed based on a Dice coefficient, and the Dice coefficient is defined as a formula:
Figure FDA0003813622860000011
wherein T is the real contour of the segmented target, and P is the scorePrediction result of target cutting, T, P ∈ [0,1 ∈ [ ]] w×h W and h represent the size of the segmented image;
subtracting the Dice coefficient from 1 as a loss function of the U-Net model, and defining the loss function as follows: loss Dice (T, P) =1-Dice (T, P), where Dice (T, P) represents Dice coefficient values of the segmentation result P and the true contour T;
let phi i (x, y) =0 and phi o (x, y) =0 zero level sets of the two level set functions represent endocardium and adventitia of myocardium, respectively, E ci For the distance-smoothly varying constraint, λ i ,α i ,μ i ,β i Is a length term E L Balloon force E A Item and rule item E R The coefficients of (c);
the coupling level set model formula based on the distance smooth change constraint is defined as:
E i (φ i ；φ o )＝λ i E L (φ i )+α i E A (φ i )+μ i E R (φ i )+β i E Ci (φ i ；φ o )
Figure FDA0003813622860000012
Figure FDA0003813622860000013
Figure FDA0003813622860000014
Figure FDA0003813622860000021
wherein δ (x) is the derivative of the Heaviside function;
the coupling level set segmentation function for the endocardium of the myocardium was obtained as:
Figure FDA0003813622860000022
the coupled level set segmentation function for the myocardial adventitia was obtained as:
Figure FDA0003813622860000023
2. the cardiac MRI image coupling level set segmentation method as set forth in claim 1, wherein: when modeling the endocardium and the epicardium by using U-Net, the U-Net comprises a down-sampling part and an up-sampling part;
the up-sampling process and the down-sampling process are symmetrical, and 3 times of sampling operation is carried out on the characteristic diagram respectively, wherein 2 times of convolution operation is carried out before each down-sampling.
3. A cardiac MRI image coupling level set segmentation system, characterized by: the device comprises a rough segmentation unit and a fine segmentation unit;
the rough segmentation unit adopts a U-Net frame to respectively construct segmentation models of the myocardial inner membrane and the myocardial outer membrane, and carries out rough segmentation on the myocardial inner membrane and the myocardial outer membrane;
the precise segmentation unit sets constraint conditions of distance smooth change based on a coupling level set model of distance smooth change constraint, and extracts the contours of the myocardial intima and the myocardial adventitia by using two level set functions in combination with the length, the rule constraint and the balloon force term of a level set to realize precise segmentation of the myocardial intima and the myocardial adventitia;
the loss function of the U-Net model is constructed based on a Dice coefficient, and the Dice coefficient is defined as a formula:
Figure FDA0003813622860000024
wherein T is the real contour of the segmentation target, P is the prediction result of the segmentation target, T, P belongs to [0,1 ]] w×h W and h represent the size of the segmented image;
subtracting the Dice coefficient from 1 as a loss function of the U-Net model, and defining the loss function as follows: loss Dice (T, P) =1-Dice (T, P), where Dice (T, P) represents Dice coefficient values of the segmentation result P and the true contour T;
let phi i (x, y) =0 and phi o (x, y) =0 zero level sets of the two level set functions represent endocardium and adventitia of myocardium, respectively, E ci For the distance-smoothly varying constraint, λ i ,α i ,μ i ,β i Is a length term E L Balloon force E A Item and rule item E R The coefficients of (c);
for myocardial intimal segmentation, the coupling level set model formula based on the distance smooth variation constraint is defined as:
E i (φ i ；φ o )＝λ i E L (φ i )+α i E A (φ i )+μ i E R (φ i )+β i E Ci (φ i ；φ o )；
wherein the content of the first and second substances,
Figure FDA0003813622860000031
Figure FDA0003813622860000032
Figure FDA0003813622860000033
Figure FDA0003813622860000034
δ (x) is the derivative of the Heaviside function;
the coupling level set segmentation function for the endocardium of the myocardium was obtained as:
Figure FDA0003813622860000035
similarly, the coupling level set segmentation function of the myocardial adventitia is obtained as follows:
Figure FDA0003813622860000036
4. a cardiac MRI image coupling level set segmentation system as set forth in claim 3, wherein: when modeling the endocardium and the epicardium by using U-Net, the U-Net comprises a down-sampling part and an up-sampling part;
the up-sampling process and the down-sampling process are symmetrical, and 3 times of sampling operation is carried out on the characteristic diagram respectively, wherein 2 times of convolution operation is carried out before each down-sampling.
5. A terminal, characterized by: comprising a processor, a memory and a cardiac MRI image coupling level set segmentation method program stored in the memory, which when executed by the processor, carries out the steps of the cardiac MRI image coupling level set segmentation method according to any one of claims 1-2.
6. A computer-readable storage medium having stored thereon computer instructions, characterized in that: the computer instructions, when executed by a processor, perform the steps of the cardiac MRI image coupling level set segmentation method as set forth in any one of claims 1-2.
CN202110557989.6A 2021-05-21 2021-05-21 Heart MRI image coupling level set segmentation method and system Active CN113379682B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110557989.6A CN113379682B (en) 2021-05-21 2021-05-21 Heart MRI image coupling level set segmentation method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110557989.6A CN113379682B (en) 2021-05-21 2021-05-21 Heart MRI image coupling level set segmentation method and system

Publications (2)

Publication Number Publication Date
CN113379682A CN113379682A (en) 2021-09-10
CN113379682B true CN113379682B (en) 2022-10-04

Family

ID=77571561

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110557989.6A Active CN113379682B (en) 2021-05-21 2021-05-21 Heart MRI image coupling level set segmentation method and system

Country Status (1)

Country Link
CN (1) CN113379682B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114419032B (en) * 2022-03-14 2022-06-21 深圳科亚医疗科技有限公司 Method and device for segmenting the endocardium and/or the epicardium of the left ventricle of the heart

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1524247A (en) * 2001-05-17 2004-08-25 О A variational approach for the segmentation of the left ventricle in mr cardiac images
CN108230342A (en) * 2017-12-29 2018-06-29 电子科技大学 A kind of left and right ventricles level-set segmentation methods based on cardiac anatomy
CN109272512A (en) * 2018-09-25 2019-01-25 南昌航空大学 A kind of method of outer membrane in automatic segmentation left ventricle
CN111127504A (en) * 2019-12-28 2020-05-08 中国科学院深圳先进技术研究院 Heart medical image segmentation method and system for atrial septal occlusion patient
CN111640120A (en) * 2020-04-09 2020-09-08 之江实验室 Pancreas CT automatic segmentation method based on significance dense connection expansion convolution network

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10699414B2 (en) * 2018-04-03 2020-06-30 International Business Machines Corporation Image segmentation based on a shape-guided deformable model driven by a fully convolutional network prior

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1524247A (en) * 2001-05-17 2004-08-25 О A variational approach for the segmentation of the left ventricle in mr cardiac images
CN108230342A (en) * 2017-12-29 2018-06-29 电子科技大学 A kind of left and right ventricles level-set segmentation methods based on cardiac anatomy
CN109272512A (en) * 2018-09-25 2019-01-25 南昌航空大学 A kind of method of outer membrane in automatic segmentation left ventricle
CN111127504A (en) * 2019-12-28 2020-05-08 中国科学院深圳先进技术研究院 Heart medical image segmentation method and system for atrial septal occlusion patient
CN111640120A (en) * 2020-04-09 2020-09-08 之江实验室 Pancreas CT automatic segmentation method based on significance dense connection expansion convolution network

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
Distance regularized two level sets for segmentation of left and right ventricles from cine-MRI;Yu Liu等;《Magnetic Resonance Imaging》;20160630;第34卷(第5期);第699-706页 *
Left ventricle segmentation via two-layer level sets with circular shape constraint;CongYang等;《Magnetic Resonance Imaging》;20170531;第202-213页 *
基于心脏电影磁共振图像的左心肌分割新进展;王慧等;《中国生物医学工程学报》;20200430;第39卷(第2期);第238-246页 *
心血管疾病的计算机辅助诊断关键技术研究;刘畅;《中国优秀硕士学位论文全文数据库医药卫生科技辑》;20210215;E062-24 *

Also Published As

Publication number Publication date
CN113379682A (en) 2021-09-10

Similar Documents

Publication Publication Date Title
US11055851B2 (en) Multi-class image segmentation method
DE112019000708T5 (en) SYSTEM FOR SEGMENTING ANATOMICAL STRUCTURES IN THE HEART-CTA USING COMPLETE CONVOLUTIONAL NEURAL NETWORKS
Nikan et al. PWD-3DNet: a deep learning-based fully-automated segmentation of multiple structures on temporal bone CT scans
Mansoor et al. Deep learning guided partitioned shape model for anterior visual pathway segmentation
CN109461495A (en) A kind of recognition methods of medical image, model training method and server
CN110827216A (en) Multi-generator generation countermeasure network learning method for image denoising
CN109754403A (en) Tumour automatic division method and system in a kind of CT image
CN110120051A (en) A kind of right ventricle automatic division method based on deep learning
CN108664976B (en) Super-pixel-based fuzzy spectral clustering brain tumor image automatic segmentation method
CN114581662B (en) Brain tumor image segmentation method, system, device and storage medium
CN110570394B (en) Medical image segmentation method, device, equipment and storage medium
CN114693561A (en) Nuclear magnetic resonance image processing method and system based on convolutional neural network
CN110706225A (en) Tumor identification system based on artificial intelligence
CN113379682B (en) Heart MRI image coupling level set segmentation method and system
CN112233132A (en) Brain magnetic resonance image segmentation method and device based on unsupervised learning
WO2022086910A1 (en) Anatomically-informed deep learning on contrast-enhanced cardiac mri
CN113436203A (en) Automatic ventricular membrane segmentation method
Alves et al. Extracting lungs from ct images using fully convolutional networks
CN110288581B (en) Segmentation method based on model for keeping shape convexity level set
CN117495876A (en) Coronary artery image segmentation method and system based on deep learning
Khader et al. Adaptive preprocessing for generalization in cardiac MR image segmentation
CN111814891A (en) Medical image synthesis method, device and storage medium
CN115100115A (en) Organ image segmentation method, device, equipment and storage medium
CN113724320A (en) Shape prior-based kidney image segmentation method
CN115775219A (en) Medical image segmentation method, system, electronic device, and medium

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