This work presents an investigation of the potential of artificial
neural networks for classification of registered magnetic resonance
and X-ray computer tomography images of the human brain. First, topological
and learning parameters are established experimentally. Second, the
learning and generalization properties of the neural networks are
compared to those of a classical maximum likelihood classifier and
the superiority of the neural network approach is demonstrated when
small training sets are utilized. Third, the generalization properties
of the neural networks are utilized to develop an adaptive learning
scheme able to overcome interslice intensity variations typical of
MR images. This approach permits the segmentation of image volumes
based on training sets selected on a single slice. Finally, the segmentation
results obtained both with the artificial neural network and the
maximum likelihood classifiers are compared to contours drawn manually
%0 Journal Article
%1 Ozkan1993
%A Ozkan, M.
%A Dawant, B.M.
%A Maciunas, R.J.
%D 1993
%J Medical Imaging, IEEE Transactions on
%K NMR, X-ray adaptive artificial biomedical brain, classical classification, classifier, computer computerised diagnostic generalization human image images images, imaging, intensity interslice learning likelihood magnetic maximum medical multimodal nets, networks, neural parameters parameters, processing, properties, resonance scheme, segmentation, tomography, topological variations,
%N 3
%P 534--544
%T Neural-network-based segmentation of multi-modal medical images:
a comparative and prospective study
%V 12
%X This work presents an investigation of the potential of artificial
neural networks for classification of registered magnetic resonance
and X-ray computer tomography images of the human brain. First, topological
and learning parameters are established experimentally. Second, the
learning and generalization properties of the neural networks are
compared to those of a classical maximum likelihood classifier and
the superiority of the neural network approach is demonstrated when
small training sets are utilized. Third, the generalization properties
of the neural networks are utilized to develop an adaptive learning
scheme able to overcome interslice intensity variations typical of
MR images. This approach permits the segmentation of image volumes
based on training sets selected on a single slice. Finally, the segmentation
results obtained both with the artificial neural network and the
maximum likelihood classifiers are compared to contours drawn manually
@article{Ozkan1993,
abstract = {This work presents an investigation of the potential of artificial
neural networks for classification of registered magnetic resonance
and X-ray computer tomography images of the human brain. First, topological
and learning parameters are established experimentally. Second, the
learning and generalization properties of the neural networks are
compared to those of a classical maximum likelihood classifier and
the superiority of the neural network approach is demonstrated when
small training sets are utilized. Third, the generalization properties
of the neural networks are utilized to develop an adaptive learning
scheme able to overcome interslice intensity variations typical of
MR images. This approach permits the segmentation of image volumes
based on training sets selected on a single slice. Finally, the segmentation
results obtained both with the artificial neural network and the
maximum likelihood classifiers are compared to contours drawn manually},
added-at = {2009-09-12T19:19:34.000+0200},
author = {Ozkan, M. and Dawant, B.M. and Maciunas, R.J.},
biburl = {https://www.bibsonomy.org/bibtex/2ec56309c0325302e948d78872ba68435/mozaher},
file = {00241881.pdf:Ozkan1993.pdf:PDF},
interhash = {e8417b9698bb7e46a3a077f7b6749e85},
intrahash = {ec56309c0325302e948d78872ba68435},
issn = {0278-0062},
journal = {Medical Imaging, IEEE Transactions on},
keywords = {NMR, X-ray adaptive artificial biomedical brain, classical classification, classifier, computer computerised diagnostic generalization human image images images, imaging, intensity interslice learning likelihood magnetic maximum medical multimodal nets, networks, neural parameters parameters, processing, properties, resonance scheme, segmentation, tomography, topological variations,},
number = 3,
owner = {Mozaher},
pages = {534--544},
timestamp = {2009-09-12T19:19:41.000+0200},
title = {Neural-network-based segmentation of multi-modal medical images:
a comparative and prospective study},
volume = 12,
year = 1993
}