Active Learning For Real Time Detection Of Polyps In Videocolonoscopy - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2016

Active Learning For Real Time Detection Of Polyps In Videocolonoscopy

Quentin Angermann
Aymeric Histace
Olivier Romain

Résumé

In this paper a method to perform real-time detection of polyps in videocolonoscopy is introduced. Polyps are at the origins of colorectal cancer which is one of the deadliest disease in the world. Many methods to improve detection of polyps have been proposed so far. But performance of these methods strongly depends on the available computational resources and, until now, are not able to perform real-time detection during a standard exam. The proposed method, based on active learning, is able to solve these issues. Most precisely, this approach allows to detect approximately 90% of polyps on a freely available database introduced to the community in 2012, for a F2 score of 65%, and matches real-time constraint by making possible the analysis of a frame in only 0.023s (average value) on a standard computer not necessarily dedicated to that kind of application.
Fichier principal
Vignette du fichier
ArticleHAL.pdf (951.26 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01344018 , version 1 (22-07-2016)

Identifiants

Citer

Quentin Angermann, Aymeric Histace, Olivier Romain. Active Learning For Real Time Detection Of Polyps In Videocolonoscopy. Medical Image Understanding and Analysis Conference, British Machine Vision Association, Jul 2016, Loughborough, United Kingdom. pp.182-187, ⟨10.1016/j.procs.2016.07.017⟩. ⟨hal-01344018⟩
197 Consultations
241 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More