Rough segmentation of coherent local intensity for bias induced 3-D MR brain images

作者：

Highlights：

• A new segmentation algorithm, termed as CoLoRS (Coherent Local Intensity Rough Segmentation), is proposed for brain MR volumes corrupted with bias field artifact.

• It judiciously integrates the merits of coherent local intensity clustering and the theory of rough sets for simultaneous segmentation and bias field correction.

• The dual-region concept is used to represent the neighborhood structure, according to the spatial locations of voxels within neighborhood.

• The segmentation in fuzzy approximation spaces handles overlapping partitions and addresses vagueness in tissue class definition.

摘要

•A new segmentation algorithm, termed as CoLoRS (Coherent Local Intensity Rough Segmentation), is proposed for brain MR volumes corrupted with bias field artifact.•It judiciously integrates the merits of coherent local intensity clustering and the theory of rough sets for simultaneous segmentation and bias field correction.•The dual-region concept is used to represent the neighborhood structure, according to the spatial locations of voxels within neighborhood.•The segmentation in fuzzy approximation spaces handles overlapping partitions and addresses vagueness in tissue class definition.

论文关键词：Magnetic resonance image,Segmentation,Intensity inhomogeneity,Rough sets,Coherent local intensity clustering

论文评审过程：Received 2 February 2018, Revised 23 July 2019, Accepted 31 July 2019, Available online 1 August 2019, Version of Record 20 August 2019.

论文官网地址：https://doi.org/10.1016/j.patcog.2019.106997