Evaluation of Carotid Artery Segmentation with Centerline Detection and Active Contours without Edges Algorithm

Abstract : The main contribution of this article is a new method of segmentation of carotid artery based on original authors inner path finding algorithm and active contours without edges segmentation method for vessels wall detection. Instead of defining new force to being minimized or intensity metric we decide to find optimal weight of image – dependent forces. This allows our method to be easily reproduced and applied in other software solutions. We judge the quality of segmentation by dice coefficient between manual segmentation done by a specialist and automatic segmentation performed by our algorithm. We did not find any other publication in which such approach for carotid artery bifurcation region segmentation has been proposed or investigated. The proposed algorithm has shown to be reliable method for that task. The dice coefficient at the level of 0.949(0.050 situates our algorithm among best state of the art methods for those solutions. That type of segmentation is the main step performed before sophisticated semantic analysis of complex image patterns utilized by cognitive image and scene understanding methods. The complete diagnostic record (Electronic Health Record – EHR) obtained that way consists private biometric data and its safety is essential for personal and homeland security.
Type de document :
Communication dans un congrès
Gerald Quirchmayr; Josef Basl; Ilsun You; Lida Xu; Edgar Weippl. International Cross-Domain Conference and Workshop on Availability, Reliability, and Security (CD-ARES), Aug 2012, Prague, Czech Republic. Springer, Lecture Notes in Computer Science, LNCS-7465, pp.468-478, 2012, Multidisciplinary Research and Practice for Information Systems. 〈10.1007/978-3-642-32498-7_35〉
Liste complète des métadonnées

Littérature citée [14 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01542461
Contributeur : Hal Ifip <>
Soumis le : lundi 19 juin 2017 - 17:01:40
Dernière modification le : mardi 20 juin 2017 - 01:06:35
Document(s) archivé(s) le : vendredi 15 décembre 2017 - 22:33:01

Fichier

978-3-642-32498-7_35_Chapter.p...
Fichiers produits par l'(les) auteur(s)

Licence


Distributed under a Creative Commons Paternité 4.0 International License

Identifiants

Citation

Tomasz Hachaj, Marek Ogiela. Evaluation of Carotid Artery Segmentation with Centerline Detection and Active Contours without Edges Algorithm. Gerald Quirchmayr; Josef Basl; Ilsun You; Lida Xu; Edgar Weippl. International Cross-Domain Conference and Workshop on Availability, Reliability, and Security (CD-ARES), Aug 2012, Prague, Czech Republic. Springer, Lecture Notes in Computer Science, LNCS-7465, pp.468-478, 2012, Multidisciplinary Research and Practice for Information Systems. 〈10.1007/978-3-642-32498-7_35〉. 〈hal-01542461〉

Partager

Métriques

Consultations de la notice

84

Téléchargements de fichiers

28