Two Methods for Validating Brain Tissue Classifiers

M. Martin-Fernandez, S. Bouix, L. Ungar, R. W. McCarley, M. E. Shenton
Lecture Notes in Computer Science
Volume 3749, Pages 515-522
2005

Download full paper

Abstract

In this paper, we present an evaluation of seven automatic brain tissue classifiers based on level of agreements. A number of agree- ment measures are explained, and we show how they can be used to compare different segmentation techniques. We use the Simultaneous Truth and Performance Level Estimation (STAPLE) of Warfield et al. but also introduce a novel evaluation technique based on the Williams' index. The methods are evaluated using these two techniques on a pop- ulation of forty subjects, each having an SPGR scan and a co-registered T2 weighted scan. We provide an interpretation of the results and show how similar the output of the STAPLE analysis and Williams' index are. When no ground truth is required, we recommend the use of Williams index as it is easy and fast to compute.


Reference

Martin-Fernandez M, Bouix S, Ungar L, McCarley RW, Shenton ME. Two methods for validating brain tissue classifiers. In J Duncan, G Gerig, eds., Lecture Notes in Computer Science, volume 3749. Palm Springs, CA, USA: Springer-Verlag Berlin Heidelberg, 2005;515-522.

Grants

NIH/NIMH 2K01 MH01110, NIH RO1 MH50747, U54 EB005149, NIH/NIMH 2RO1 MH40799, VA Merit Award(MES) 2000-2008, VA Merit Award(RWM) 1998-2009, REAP, Fulbright

Research area

segmentation
© 2013 Psychiatry Neuroimaging Laboratory | Last updated 04.15.2013