A novel combined level set model for automatic MR image segmentation

Medical image processing comes along with object segmentation, which is one of the most important tasks in that field. Nevertheless, noise and intensity inhomogeneity in magnetic resonance images challenge the segmentation procedure. The level set method has been widely used in object detection. The...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Li Jianzhang, Nebelung Sven, Rath Björn, Tingart Markus, Eschweiler Jörg
Formato: article
Lenguaje:EN
Publicado: De Gruyter 2020
Materias:
mri
R
Acceso en línea:https://doaj.org/article/7e84993e1fd945a09f88c303acc1fb33
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:7e84993e1fd945a09f88c303acc1fb33
record_format dspace
spelling oai:doaj.org-article:7e84993e1fd945a09f88c303acc1fb332021-12-05T14:10:42ZA novel combined level set model for automatic MR image segmentation2364-550410.1515/cdbme-2020-3006https://doaj.org/article/7e84993e1fd945a09f88c303acc1fb332020-09-01T00:00:00Zhttps://doi.org/10.1515/cdbme-2020-3006https://doaj.org/toc/2364-5504Medical image processing comes along with object segmentation, which is one of the most important tasks in that field. Nevertheless, noise and intensity inhomogeneity in magnetic resonance images challenge the segmentation procedure. The level set method has been widely used in object detection. The flexible integration of energy terms affords the level set method to deal with variable difficulties. In this paper, we introduce a novel combined level set model that mainly cooperates with an edge detector and a local region intensity descriptor. The noise and intensity inhomogeneities are eliminated by the local region intensity descriptor. The edge detector helps the level set model to locate the object boundaries more precisely. The proposed model was validated on synthesized images and magnetic resonance images of in vivo wrist bones. Comparing with the ground truth, the proposed method reached a Dice similarity coefficient of > 0.99 on all image tests, while the compared segmentation approaches failed the segmentations. The presented combined level set model can be used for the object segmentation in magnetic resonance images.Li JianzhangNebelung SvenRath BjörnTingart MarkusEschweiler JörgDe Gruyterarticlelevel set methodmrisegmentationintensity inhomogeneityMedicineRENCurrent Directions in Biomedical Engineering, Vol 6, Iss 3, Pp 20-23 (2020)
institution DOAJ
collection DOAJ
language EN
topic level set method
mri
segmentation
intensity inhomogeneity
Medicine
R
spellingShingle level set method
mri
segmentation
intensity inhomogeneity
Medicine
R
Li Jianzhang
Nebelung Sven
Rath Björn
Tingart Markus
Eschweiler Jörg
A novel combined level set model for automatic MR image segmentation
description Medical image processing comes along with object segmentation, which is one of the most important tasks in that field. Nevertheless, noise and intensity inhomogeneity in magnetic resonance images challenge the segmentation procedure. The level set method has been widely used in object detection. The flexible integration of energy terms affords the level set method to deal with variable difficulties. In this paper, we introduce a novel combined level set model that mainly cooperates with an edge detector and a local region intensity descriptor. The noise and intensity inhomogeneities are eliminated by the local region intensity descriptor. The edge detector helps the level set model to locate the object boundaries more precisely. The proposed model was validated on synthesized images and magnetic resonance images of in vivo wrist bones. Comparing with the ground truth, the proposed method reached a Dice similarity coefficient of > 0.99 on all image tests, while the compared segmentation approaches failed the segmentations. The presented combined level set model can be used for the object segmentation in magnetic resonance images.
format article
author Li Jianzhang
Nebelung Sven
Rath Björn
Tingart Markus
Eschweiler Jörg
author_facet Li Jianzhang
Nebelung Sven
Rath Björn
Tingart Markus
Eschweiler Jörg
author_sort Li Jianzhang
title A novel combined level set model for automatic MR image segmentation
title_short A novel combined level set model for automatic MR image segmentation
title_full A novel combined level set model for automatic MR image segmentation
title_fullStr A novel combined level set model for automatic MR image segmentation
title_full_unstemmed A novel combined level set model for automatic MR image segmentation
title_sort novel combined level set model for automatic mr image segmentation
publisher De Gruyter
publishDate 2020
url https://doaj.org/article/7e84993e1fd945a09f88c303acc1fb33
work_keys_str_mv AT lijianzhang anovelcombinedlevelsetmodelforautomaticmrimagesegmentation
AT nebelungsven anovelcombinedlevelsetmodelforautomaticmrimagesegmentation
AT rathbjorn anovelcombinedlevelsetmodelforautomaticmrimagesegmentation
AT tingartmarkus anovelcombinedlevelsetmodelforautomaticmrimagesegmentation
AT eschweilerjorg anovelcombinedlevelsetmodelforautomaticmrimagesegmentation
AT lijianzhang novelcombinedlevelsetmodelforautomaticmrimagesegmentation
AT nebelungsven novelcombinedlevelsetmodelforautomaticmrimagesegmentation
AT rathbjorn novelcombinedlevelsetmodelforautomaticmrimagesegmentation
AT tingartmarkus novelcombinedlevelsetmodelforautomaticmrimagesegmentation
AT eschweilerjorg novelcombinedlevelsetmodelforautomaticmrimagesegmentation
_version_ 1718371794386157568