A RVR-based Method for Bias Field Estimation in Brain Magnetic Resonance Images Segmentation

Jinwei Wang,Dexing Kong
DOI: https://doi.org/10.1007/978-3-642-29305-4_101
2013-01-01
Abstract:This paper presents a relevance vector regression (RVR) based parametric approach to the bias field estimation in brain magnetic resonance (MR) image segmentation. Segmentation is a very important and challenging task in brain analysis, while the bias field existed in the images can significantly deteriorate the performance. Most of current parametric bias field correction techniques use a pre-set linear combination of low degree basis functions, the coefficients and the basis function types of which completely determine the field. The proposed RVR method can automatically determine the best combination for the bias field, resulting in a good segmentation in the presence of noise by combining with spatial constrained fuzzy C-means (SCFCM) segmentation. Experiments on simulated T1 images show the efficiency.
What problem does this paper attempt to address?