Automatic graph cut segmentation of multiple sclerosis lesions

Laurence Catanese 1 Olivier Commowick 1 Christian Barillot 1, *
* Corresponding author
1 VisAGeS - Vision, Action et Gestion d'informations en Santé
INSERM - Institut National de la Santé et de la Recherche Médicale : U746, Inria Rennes – Bretagne Atlantique , IRISA-D5 - SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE
Abstract : A fully automated segmentation algorithm for Multiple Sclerosis (MS) lesions is presented. Our method includes two main steps: the detection of lesions by graph cut initialized with a robust Expectation-Maximization (EM) algorithm and the application of rules to remove false positives. Our algorithm will be tested on the ISBI 2015 challenge longitudinal data. For each patient, a unique parameter set is used to run the algorithm.
Complete list of metadatas

https://www.hal.inserm.fr/inserm-01304109
Contributor : Olivier Commowick <>
Submitted on : Tuesday, April 19, 2016 - 10:41:50 AM
Last modification on : Monday, August 5, 2019 - 5:56:07 PM
Long-term archiving on : Tuesday, November 15, 2016 - 6:14:29 AM

File

ISBI_Challenge_Catanese.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inserm-01304109, version 1

Citation

Laurence Catanese, Olivier Commowick, Christian Barillot. Automatic graph cut segmentation of multiple sclerosis lesions. ISBI Longitudinal Multiple Sclerosis Lesion Segmentation Challenge, Apr 2015, New York, United States. ⟨inserm-01304109⟩

Share

Metrics

Record views

304

Files downloads

237