Biological cells classification using bio-inspired descriptor in a boosting k-NN framework

Abstract : High-content imaging is an emerging technology for the analysis and quantification of biological phenomena. Thus, classifying a huge number of cells or quantifying markers from large sets of images by experts is a very time-consuming and poorly reproducible task. In order to overcome such limitations, we propose a supervised method for automatic cell classification. Our approach consists of two steps: the first one is an indexing stage based on specific bio-inspired features relying on the distribution of contrast information on segmented cells. The second one is a supervised learning stage that selects the prototypical samples best representing the cell categories. These prototypes are used in a leveraged k-NN framework to predict the class of unlabeled cells. In this paper we have tested our new learning algorithm on cellular images acquired for the analysis of pathologies. In order to evaluate the automatic classification performances, we tested our algorithm on the HEp2 Cells dataset of (Foggia et al, CBMS 2010). Results are very promising, showing classification precision larger than 96% on average, thus suggesting our method as a valuable decision-support tool in such cellular imaging applications.
Type de document :
Communication dans un congrès
CBMS - 25th International Symposium on Computer-Based Medical Systems, Jun 2012, Rome, Italy. IEEE, pp.1-6, 2012, <10.1109/CBMS.2012.6266359>
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-00958860
Contributeur : Estelle Nivault <>
Soumis le : jeudi 13 mars 2014 - 14:54:34
Dernière modification le : jeudi 21 janvier 2016 - 15:58:04
Document(s) archivé(s) le : vendredi 13 juin 2014 - 11:46:07

Fichier

cbms.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

UNICE | CEA | I3S | DSV

Citation

Wafa Bel Haj Ali, Paolo Piro, Dario Giampaglia, Thierry Pourcher, Michel Barlaud. Biological cells classification using bio-inspired descriptor in a boosting k-NN framework. CBMS - 25th International Symposium on Computer-Based Medical Systems, Jun 2012, Rome, Italy. IEEE, pp.1-6, 2012, <10.1109/CBMS.2012.6266359>. <hal-00958860>

Partager

Métriques

Consultations de
la notice

157

Téléchargements du document

385