Evaluating Automatic Brain Tissue Classifiers

S. Bouix, L. Ungar, C. C. Dickey, R. W. McCarley, M. E. Shenton
Proceedings of the 7th International Conference on Medical Image Computing and Computer Assisted Intervention MICCAI 2004
Pages 1038-1039
2004

Download full paper

Abstract

We present a quantitative evaluation of MR brain images segmentation. Five classi ers were tested. The task was to classify an MR image into four di erent classes: background, cortical spinal uid, gray matter and white matter. The performance was rated by rst estimating a ground truth (EGT) using STAPLE and then analyzing the volume di erences as well as the Dice similarity measure between each of the 5 classifers.

Reference

Bouix S, Ungar L, Dickey CC, McCarley RW, Shenton ME. Evaluating automatic brain tissue classifiers. In Proceedings of the 7th International Conference on Medical Image Computing and Computer Assisted Intervention MICCAI 2004. Springer-Verlag GmbH, 2004;1038-1039.

Research area

segmentation
© 2013 Psychiatry Neuroimaging Laboratory | Last updated 04.15.2013