Segmentation of Brain MR Images with Bias Field Correction.
Author(s)
Kim, Seung-Gu
Ng, Shu-Kay
J. McLachlan, Geoffrey
Wang, Deming
Griffith University Author(s)
Year published
2003
Metadata
Show full item recordAbstract
We consider a statistical model-based approach to the segmentation of magnetic resonance (MR) images with bias field correction. The proposed method of penalized maximum likelihood is implemented via the expectationconditional maximization (ECM) algorithm, using an approximation to the E-step based on a fractional weight version of the iterated conditional modes (ICM) algorithm. A Markov random field (MRF) is adopted to model the spatial dependence between neighouring voxels. The approach is illustrated using some simulated and real MR data.We consider a statistical model-based approach to the segmentation of magnetic resonance (MR) images with bias field correction. The proposed method of penalized maximum likelihood is implemented via the expectationconditional maximization (ECM) algorithm, using an approximation to the E-step based on a fractional weight version of the iterated conditional modes (ICM) algorithm. A Markov random field (MRF) is adopted to model the spatial dependence between neighouring voxels. The approach is illustrated using some simulated and real MR data.
View less >
View less >
Conference Title
Proceedings of the 2003 APRS Workshop on Digital Image Computing (WDIC 2003)
Publisher URI
Subject
Multi-Disciplinary