Automatic intensity-based tissue classification sets requirements for the q
uality of multispectral magnetic resonance (MR) images. Tests for evaluatin
g the separability of tissue classes, and on the other hand class distances
required to obtain reliable classification, are presented in this study. I
ntraslice, interslice and interpatient training schemes for 5-nn classifica
tion were considered. Interslice training was utilized in classification of
images from 10 patients with ischemic stroke giving results of satisfactor
y but highly variable quality. Based on the experience with these data sets
, similar tests are recommended before imaging a large patient series in or
der to avoid extra manual work and to obtain reliable classification result
s. (C) 2001 Elsevier Science B.V. All rights reserved.