DocumentCode :
594670
Title :
Graph cut energy minimization in a probabilistic learning framework for 3D prostate segmentation in MRI
Author :
Ghose, Sarbani ; Mitra, Joydeep ; Oliver, Arnau ; Marti, Robert ; Llado, Xavier ; Freixenet, J. ; Vilanova, J.C. ; Sidibe, Desire ; Meriaudeau, Fabrice
Author_Institution :
Univ. de Bourgogne, Dijon, France
fYear :
2012
fDate :
11-15 Nov. 2012
Firstpage :
125
Lastpage :
128
Abstract :
Variations in inter-patient prostate shape, and size and imaging artifacts in magnetic resonance images (MRI) hinders automatic accurate prostate segmentation. In this paper we propose a graph cut based energy minimization of the posterior probabilities obtained in a supervised learning schema for automatic 3D segmentation of the prostate in MRI. A probabilistic classification of the prostate voxels is achieved with a probabilistic atlas and a random forest based learning framework. The posterior probabilities are combined to obtain the likelihood of a voxel being prostate. Finally, 3D graph cut based energy minimization in the stochastic space provides segmentation of the prostate. The proposed method achieves a mean Dice similarity coefficient (DSC) value of 0.91±0.04 and 95% mean Haus-dorff distance (HD) of 4.69±2.62 voxels when validated with 15 prostate volumes of a public dataset in a leave-one-patient-out validation framework. The model achieves statistically significant t-test p-value<;0.0001 in mean DSC and mean HD values compared to some of the works in literature.
Keywords :
biomedical MRI; graph theory; image classification; image segmentation; inference mechanisms; learning (artificial intelligence); medical image processing; minimisation; probability; shape recognition; stochastic processes; 3D automatic accurate prostate segmentation; 3D graph cut energy minimization; DSC; Dice similarity coefficient; HD; Hausdorff distance; MRI; imaging artifact; interpatient prostate shape variation; leave-one-patient-out validation framework; magnetic resonance image; posterior probability; probabilistic atlas; probabilistic learning framework; probabilistic prostate voxel classification; random forest based learning framework; stochastic space; supervised learning schema; Image segmentation; Magnetic resonance imaging; Minimization; Probabilistic logic; Shape; Training; Vegetation;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Pattern Recognition (ICPR), 2012 21st International Conference on
Conference_Location :
Tsukuba
ISSN :
1051-4651
Print_ISBN :
978-1-4673-2216-4
Type :
conf
Filename :
6460088
Link To Document :
بازگشت