Title :
Automated topologically correct cortical surfaces from MR images
Author :
Shattuck, D. ; Leahy, R.
Author_Institution :
Signal & Image Process. Inst., Univ. of Southern California, Los Angeles, CA, USA
Abstract :
We present an automated method for generating cortical surface representations from magnetic resonance (MR) images. The method is based on a sequence of low-level operations. The brain is extracted from an MR volume, image non-uniformity is estimated and corrected by examining local properties of the brain volume, and a Bayesian classifier labels the bias-corrected image with tissue classes. The white matter surface is selected and automatically edited using a graph-based algorithm to produce a typologically spherical volume
Keywords :
Bayes methods; biological tissues; biomedical MRI; brain; graph theory; image classification; image sequences; medical image processing; rendering (computer graphics); Bayesian classifier; MR images; MR volume; automated topologically correct cortical surfaces; bias-corrected image; brain volume; cortical surface representations; graph-based algorithm; image non-uniformity; low-level operations; magnetic resonance images; sequence; tissue classes; typologically spherical volume; white matter surface; Bayesian methods; Brain; Cerebral cortex; Image processing; Image segmentation; Labeling; Lattices; Magnetic resonance; Signal processing; Surface reconstruction;
Conference_Titel :
[Engineering in Medicine and Biology, 1999. 21st Annual Conference and the 1999 Annual Fall Meetring of the Biomedical Engineering Society] BMES/EMBS Conference, 1999. Proceedings of the First Joint
Conference_Location :
Atlanta, GA
Print_ISBN :
0-7803-5674-8
DOI :
10.1109/IEMBS.1999.804311
