A label fusion method using conditional random fields with higher-order potentials: Application to hippocampal segmentation

作者：

Highlights：

• A label fusion method based on minimizing an energy using graph-cuts is presented.

• The discrete energy is defined on unary, pairwise and higher-order potentials.

• Two available databases of T1-weighted magnetic resonance images are used.

• We compare several label fusion methods in the hippocampal automatic segmentation.

• The scripts for running are available at https://www.nitrc.org/projects/lf_crf/.

摘要

Highlights•A label fusion method based on minimizing an energy using graph-cuts is presented.•The discrete energy is defined on unary, pairwise and higher-order potentials.•Two available databases of T1-weighted magnetic resonance images are used.•We compare several label fusion methods in the hippocampal automatic segmentation.•The scripts for running are available at https://www.nitrc.org/projects/lf_crf/.

论文关键词：Atlas-based segmentation,Image registration,Label fusion,Graph cuts,Global optimization,Hippocampal segmentation,Magnetic resonance imaging

论文评审过程：Received 25 September 2014, Revised 21 January 2015, Accepted 26 April 2015, Available online 4 May 2015, Version of Record 9 June 2015.

论文官网地址：https://doi.org/10.1016/j.artmed.2015.04.005