White matter lesion extension to automatic brain tissue segmentation on MRI

• Published in: NeuroImage (Orlando, Fla.) Vol. 45; no. 4; pp. 1151 - 1161
• Main Authors: de Boer, Renske, Vrooman, Henri A., van der Lijn, Fedde, Vernooij, Meike W., Ikram, M. Arfan, van der Lugt, Aad, Breteler, Monique M.B., Niessen, Wiro J.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.05.2009; Elsevier Limited
• Subjects: Age; Aged; Aged, 80 and over; Algorithms; Automation; Brain; Brain - pathology; Brain tissue segmentation; Confidence intervals; Data imaging; Demyelinating Diseases - pathology; Female; Humans; Image Enhancement - methods; Image Interpretation, Computer-Assisted - methods; Magnetic Resonance Imaging - methods; Male; Medical imaging; Medical research; Methods; MRI; Nerve Fibers, Myelinated - pathology; NMR; Nuclear magnetic resonance; Pattern Recognition, Automated - methods; Reproducibility of Results; Sensitivity and Specificity; Studies; White matter hyperintensities; White matter lesions; White matter lesions; Brain tissue segmentation; White matter hyperintensities; MRI
• ISSN: 1053-8119; 1095-9572; 1095-9572
• DOI: 10.1016/j.neuroimage.2009.01.011

A fully automated brain tissue segmentation method is optimized and extended with white matter lesion segmentation. Cerebrospinal fluid (CSF), gray matter (GM) and white matter (WM) are segmented by an atlas-based k-nearest neighbor classifier on multi-modal magnetic resonance imaging data. This classifier is trained by registering brain atlases to the subject. The resulting GM segmentation is used to automatically find a white matter lesion (WML) threshold in a fluid-attenuated inversion recovery scan. False positive lesions are removed by ensuring that the lesions are within the white matter. The method was visually validated on a set of 209 subjects. No segmentation errors were found in 98% of the brain tissue segmentations and 97% of the WML segmentations. A quantitative evaluation using manual segmentations was performed on a subset of 6 subjects for CSF, GM and WM segmentation and an additional 14 for the WML segmentations. The results indicated that the automatic segmentation accuracy is close to the interobserver variability of manual segmentations.