Skip to Main content Skip to Navigation
Journal articles

Prostate segmentation on T2 MRI using Optimal Surface Detection

Abstract : This paper deals with a T2 MRI prostate segmentation method. We assume to have an initial surface mesh obtained interactively or after a first rough segmentation. The surface of the prostate is then searched within the initial mesh neighborhood using the Optimal Surface Detection algorithm (OSD). This algorithm is based on the construction of a directed graph from the information obtained around the initial mesh. The optimal surface is then obtained by a graph cut. Three different cost functions for the graph have been explored, one based on the local gradient, another on a statistical model of shape and a third on a model of gradient profile. The parameters of this method have been tuned on 33 different T2 MRI volumes.
Document type :
Journal articles
Complete list of metadata

Cited literature [6 references]  Display  Hide  Download
Contributor : Jean-Louis Dillenseger Connect in order to contact the contributor
Submitted on : Friday, November 1, 2013 - 11:58:05 AM
Last modification on : Wednesday, March 9, 2022 - 9:16:03 AM
Long-term archiving on: : Sunday, February 2, 2014 - 4:28:02 AM


Files produced by the author(s)



Ke Wu, Carole Garnier, Huazhong Shu, Jean-Louis Dillenseger. Prostate segmentation on T2 MRI using Optimal Surface Detection. Innovation and Research in BioMedical engineering, Elsevier Masson, 2013, 34 (4-5), pp.287-290. ⟨10.1016/j.irbm.2013.07.005⟩. ⟨inserm-00879165⟩



Record views


Files downloads