Brain tissue MR-image segmentation via optimum-path forest clustering

• Published in: Computer vision and image understanding Vol. 116; no. 10; pp. 1047 - 1059
• Main Authors: Cappabianco, Fábio A.M., Falcão, Alexandre X., Yasuda, Clarissa L., Udupa, Jayaram K.
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier Inc 01.10.2012; Elsevier
• Subjects: Applied sciences; Artificial intelligence; Biological and medical sciences; Brain; Brain tissue segmentation; Clustering; Computer science; control theory; systems; Exact sciences and technology; Field inhomogeneity/bias correction; Forests; Graph-based methods; Image clustering; Investigative techniques, diagnostic techniques (general aspects); Labels; Magnetic resonance images; Medical image analysis; Medical sciences; Nervous system; Pattern recognition. Digital image processing. Computational geometry; Radiodiagnosis. Nmr imagery. Nmr spectrometry; Robustness; Segmentation; Segmentation evaluation; Strategy; Segmentation evaluation; Brain tissue segmentation; Field inhomogeneity/bias correction; Magnetic resonance images; Medical image analysis; Image clustering; Graph-based methods; Cluster analysis; Brain; Image processing; Bias; Central nervous system; Optimization; Encephalon; Tissue; Completeness; Voxel; Classification; Robustness; Cluster; Graph theory; Anatomy; Nuclear magnetic resonance imaging; Image segmentation; Image analysis; Medical imagery; Tridimensional image; Random set; Graph method
• ISSN: 1077-3142; 1090-235X
• DOI: 10.1016/j.cviu.2012.06.002

► A novel brain MRI inhomogeneity correction method based on WM voxels. ► Automatic brain tissue segmentation by an optimum-path forest algorithm. ► Segmentation evaluation by delineation operating characteristic curve. ► Parameter independent and optimal comparison of brain tissue segmentation methods. We present an accurate and fast approach for MR-image segmentation of brain tissues, that is robust to anatomical variations and takes an average of less than 1min for completion on modern PCs. The method first corrects voxel values in the brain based on local estimations of the white-matter intensities. This strategy is inspired by other works, but it is simple, fast, and very effective. Tissue classification exploits a recent clustering approach based on the motion of optimum-path forest (OPF), which can find natural groups such that the absolute majority of voxels in each group belongs to the same class. First, a small random set of brain voxels is used for OPF clustering. Cluster labels are propagated to the remaining voxels, and then class labels are assigned to each group. The experiments used several datasets from three protocols (involving normal subjects, phantoms, and patients), two state-of-the-art approaches, and a novel methodology which finds the best choice of parameters for each method within the operational range of these parameters using a training dataset. The proposed method outperformed the compared approaches in speed, accuracy, and robustness.