TU Darmstadt / ULB / TUbiblio

Automatic Segmentation of Structures in CT Head and Neck Images using a Coupled Shape Model

Jung, Florian and Knapp, Oliver and Wesarg, Stefan (2016):
Automatic Segmentation of Structures in CT Head and Neck Images using a Coupled Shape Model.
In: MIDAS Journal, p. 3, [Article]

Abstract

The common approach to do a fully automatic segmentation of multiple struc tures is an atlas or multi-atlas based solution. These already have proven to be suitable for the segmentation of structures in the head and neck area and provide very accurate segmentation results, but can struggle with challenging cases with unnatural postures, where the registration of the reference patient(s) is extremely difficult. Therefore, we propose an coupled shape model (CoSMo) algorithm for the segmentation relevant structures in parallel. The model adaptation to a test image is done with respect to the appearance of its items and the trained articulation space. Even on very challenging data sets with unnatural postures, which occur far more often than expected, the model adaptation algorithm succeeds. The approach is based on an articulated atlas citeSteger2012a, that is trained from a set of manually labeled training samples. Furthermore, we have combined the initial solution with statistical shape models citeKirschner2011 to represent structures with high shape variation. CoSMo is not tailored to specifc structures or regions. It can be trained from any set of given gold standard segmentations and makes it thereby very generic.

Item Type: Article
Erschienen: 2016
Creators: Jung, Florian and Knapp, Oliver and Wesarg, Stefan
Title: Automatic Segmentation of Structures in CT Head and Neck Images using a Coupled Shape Model
Language: English
Abstract:

The common approach to do a fully automatic segmentation of multiple struc tures is an atlas or multi-atlas based solution. These already have proven to be suitable for the segmentation of structures in the head and neck area and provide very accurate segmentation results, but can struggle with challenging cases with unnatural postures, where the registration of the reference patient(s) is extremely difficult. Therefore, we propose an coupled shape model (CoSMo) algorithm for the segmentation relevant structures in parallel. The model adaptation to a test image is done with respect to the appearance of its items and the trained articulation space. Even on very challenging data sets with unnatural postures, which occur far more often than expected, the model adaptation algorithm succeeds. The approach is based on an articulated atlas citeSteger2012a, that is trained from a set of manually labeled training samples. Furthermore, we have combined the initial solution with statistical shape models citeKirschner2011 to represent structures with high shape variation. CoSMo is not tailored to specifc structures or regions. It can be trained from any set of given gold standard segmentations and makes it thereby very generic.

Journal or Publication Title: MIDAS Journal
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Mathematical and Applied Visual Computing
Date Deposited: 03 May 2019 07:23
Export:
Suche nach Titel in: TUfind oder in Google

Optionen (nur für Redakteure)

View Item View Item