東華大學圖書館 |

Automatic Mesh-based Segmentation of Multiple Organs in MR Images.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Automatic Mesh-based Segmentation of Multiple Organs in MR Images./
作者:	Moheb Pour, Majid Reza.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2018,
面頁冊數:	122 p.
附註:	Source: Masters Abstracts International, Volume: 80-09.
Contained By:	Masters Abstracts International80-09.
標題:	Biomedical engineering. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=13839148

Automatic Mesh-based Segmentation of Multiple Organs in MR Images.
Moheb Pour, Majid Reza.

Automatic Mesh-based Segmentation of Multiple Organs in MR Images. - Ann Arbor : ProQuest Dissertations & Theses, 2018 - 122 p.

Source: Masters Abstracts International, Volume: 80-09.

Thesis (Ph.D.)--Ecole Polytechnique, Montreal (Canada), 2018.

This item must not be sold to any third party vendors.

Segmentation of multiple anatomical structures in MR images is often required for biomedical engineering applications such as clinical simulation, image-guided surgery, treatment planning, etc. Moreover, there is a growing need for automatic segmentation of multiple organs and complex structures from this medical imaging modality. Many successful multi-object segmentation attempts were introduced for CT images. However in the case of MR images it is a more challenging task due to intensity inhomogeneity and variability of anatomy appearance. Therefore, state-of-the-art in multi-object MR segmentation is very inferior to that of CT images.In literature dealing with MR image segmentation, the region-based approaches are sensitive to noise and non-uniformity in the input image. The edge-based approaches are challenging to group the edge information into a coherent closed contour. The atlas-based techniques can be problematic for complicated structures with anatomical variability. Deformable models are among the most popular methods for automatic detection of different organs in MR images. However they still have an important limitation which is that they are sensitive to initial position and shape of the model. An unsuitable initialization may provide failure to capture the true boundaries of the objects. On the other hand, a useful aim for an automatic multi-object MR segmentation is to provide a model which promotes understanding of the structural features of the distinct objects within the MR images. The current automatic initialization methods which have used different descriptors are not completely successful in extracting multiple objects from MR images and we need to find richer information that is available from edges. In this regard, anisotropic adaptive meshes seem to be a potential solution to the aforesaid limitation. Anisotropic adaptive meshes constructed from MR images contain higher level, abstract information about the anatomical structures of the organs within the image retained as the elements shape and orientation. Existing methods for constructing adaptive meshes based on image features have a practical limitation where manifest itself in inadequate mesh elements alignment to inclined edges in the image. Therefore, we also have to enhance mesh adaptation process to provide a better mesh-based representation. In this Ph.D. project, considering the highlighted limitations we are going to present a novel method for automatic segmentation of multiple organs in MR images by incorporating mesh adaptation techniques. In our progress, first, we improve an anisotropic adaptation process for the meshes that are constructed from MR images where the mesh elements align adequately to the image content and improve mesh anisotropy along edges in all directions. Then the resulting adaptive meshes are used for initialization of multiple active models which leads to extract initial object boundaries close to the true boundaries of multiple objects simultaneously. Finally, the Vector Field Convolution method is utilized to guide curve evolution towards the object boundaries to obtain the final segmentation results and present a better performance in terms of speed and accuracy.Subjects--Topical Terms:

535387
Biomedical engineering.

Automatic Mesh-based Segmentation of Multiple Organs in MR Images.
LDR:07375nmm a2200337 4500 001 2209670
005 20191104073802.5
008 201008s2018 ||||||||||||||||| ||eng d
035 $a (MiAaPQ)AAI13839148
035 $a (MiAaPQ)EPMontreal3258
035 $a AAI13839148
040 $a MiAaPQ $c MiAaPQ
100 1 $a Moheb Pour, Majid Reza. $3 3436768
245 1 0 $a Automatic Mesh-based Segmentation of Multiple Organs in MR Images.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2018
300 $a 122 p.
500 $a Source: Masters Abstracts International, Volume: 80-09.
500 $a Publisher info.: Dissertation/Thesis.
500 $a Advisor: Guibault, Francois;Cheriet, Farida.
502 $a Thesis (Ph.D.)--Ecole Polytechnique, Montreal (Canada), 2018.
506 $a This item must not be sold to any third party vendors.
520 $a Segmentation of multiple anatomical structures in MR images is often required for biomedical engineering applications such as clinical simulation, image-guided surgery, treatment planning, etc. Moreover, there is a growing need for automatic segmentation of multiple organs and complex structures from this medical imaging modality. Many successful multi-object segmentation attempts were introduced for CT images. However in the case of MR images it is a more challenging task due to intensity inhomogeneity and variability of anatomy appearance. Therefore, state-of-the-art in multi-object MR segmentation is very inferior to that of CT images.In literature dealing with MR image segmentation, the region-based approaches are sensitive to noise and non-uniformity in the input image. The edge-based approaches are challenging to group the edge information into a coherent closed contour. The atlas-based techniques can be problematic for complicated structures with anatomical variability. Deformable models are among the most popular methods for automatic detection of different organs in MR images. However they still have an important limitation which is that they are sensitive to initial position and shape of the model. An unsuitable initialization may provide failure to capture the true boundaries of the objects. On the other hand, a useful aim for an automatic multi-object MR segmentation is to provide a model which promotes understanding of the structural features of the distinct objects within the MR images. The current automatic initialization methods which have used different descriptors are not completely successful in extracting multiple objects from MR images and we need to find richer information that is available from edges. In this regard, anisotropic adaptive meshes seem to be a potential solution to the aforesaid limitation. Anisotropic adaptive meshes constructed from MR images contain higher level, abstract information about the anatomical structures of the organs within the image retained as the elements shape and orientation. Existing methods for constructing adaptive meshes based on image features have a practical limitation where manifest itself in inadequate mesh elements alignment to inclined edges in the image. Therefore, we also have to enhance mesh adaptation process to provide a better mesh-based representation. In this Ph.D. project, considering the highlighted limitations we are going to present a novel method for automatic segmentation of multiple organs in MR images by incorporating mesh adaptation techniques. In our progress, first, we improve an anisotropic adaptation process for the meshes that are constructed from MR images where the mesh elements align adequately to the image content and improve mesh anisotropy along edges in all directions. Then the resulting adaptive meshes are used for initialization of multiple active models which leads to extract initial object boundaries close to the true boundaries of multiple objects simultaneously. Finally, the Vector Field Convolution method is utilized to guide curve evolution towards the object boundaries to obtain the final segmentation results and present a better performance in terms of speed and accuracy.
520 $a La segmentation de structures anatomiques multiples dans des images de resonance magnetique (RM) est souvent requise dans des applications de genie biomedical telles que la simulation numerique, la chirurgie guidee par l'image, la planification de traitements, etc. De plus, il y a un besoin croissant pour une segmentation automatique d'organes multiples et de structures complexes a partir de cette modalite d'imagerie. Il existe plusieurs techniques de segmentation multi-objets qui ont ete appliquees avec succes sur des images de tomographie axiale a rayons-X (CT). Cependant, dans le cas des images RM cette tache est plus difficile en raison de l'inhomogeneite des intensites dans ces images et de la variabilite dans l'apparence des structures anatomiques. Par consequent, l'etat de l'art sur la segmentation multi-objets sur des images RM est beaucoup plus faible que celui sur les images CT. Parmi les travaux qui portent sur la segmentation d'images RM, les approches basees sur la segmentation de regions sont sensibles au bruit et la non uniformite de l'intensite dans les images. Les approches basees sur les contours ont de la difficulte a regrouper les informations sur les contours de sorte a produire un contour ferme coherent. Les techniques basees sur les atlas peuvent avoir des problemes en presence de structures complexes avec une grande variabilite anatomique. Les modeles deformables representent une des methodes les plus populaire pour la detection automatique de differents organes dans les images RM. Cependant, ces modeles souffrent encore d'une limitation importante qui est leur sensibilite a la position initiale et la forme du modele. Une initialisation inappropriee peut conduire a un echec dans l'extraction des frontieres des objets. D'un autre cote, le but ultime d'une segmentation automatique multi-objets dans les images RM est de produire un modele qui peut aider a extraire les caracteristiques structurelles d'organes distincts dans les images. Les methodes d'initialisation automatique actuelles qui utilisent differents descripteurs ne reussissent pas completement l'extraction d'objets multiples dans les images RM. Nous avons besoin d'exploiter une information plus riche qui se trouve dans les contours des organes. Dans ce contexte les maillages adaptatifs anisotropiques semblent etre une solution potentielle au probleme souleve. Les maillages adaptatifs anisotropiques construits a partir des images RM contiennent de l'information a un plus haut niveau d'abstraction representant les elements, d'une orientation et d'une forme donnee, qui constituent les differents organes dans l'image. Les methodes existantes pour la construction de maillages adaptatifs sont basees sur les intensites dans l'image et possedent une limitation pratique qui est l'alignement inadequat des elements du maillage en presence de contours inclines dans l'image. Par consequent, nous avons aussi besoin d'ameliorer le processus d'adaptation de maillage pour produire une meilleure representation de l'image basee sur un maillage.
590 $a School code: 1105.
650 4 $a Biomedical engineering. $3 535387
650 4 $a Medical imaging. $3 3172799
650 4 $a Computer engineering. $3 621879
690 $a 0464
690 $a 0541
690 $a 0574
710 2 $a Ecole Polytechnique, Montreal (Canada). $3 1018606
773 0 $t Masters Abstracts International $g 80-09.
790 $a 1105
791 $a Ph.D.
792 $a 2018
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=13839148