Title :
Optimization neural networks for the segmentation of magnetic resonance images
Author :
Amartur, S.C. ; Piraino, D. ; Takefuji, Y.
Author_Institution :
Dept. of Radiol., Case Western Reserve Univ., Cleveland, OH, USA
fDate :
6/1/1992 12:00:00 AM
Abstract :
The application of the Hopfield neural network for the multispectral unsupervised classification of MR images is reported. Winner-take-all neurons were used to obtain a crisp classification map using proton density-weighted and T2-weighted images in the head. The preliminary studies indicate that the number of iterations needed to reach `good´ solutions was nearly constant with the number of clusters chosen for the problem
Keywords :
biomedical NMR; neural nets; optimisation; picture processing; Hopfield neural network; MR images; T2-weighted images; clusters; crisp classification map; head; iterations; magnetic resonance images segmentation; multispectral unsupervised classification; optimization neural networks; proton density-weighted images; winner-take-all neurons; Artificial neural networks; Clustering algorithms; Computer architecture; Humans; Image segmentation; Magnetic resonance; Magnetic resonance imaging; Neural networks; Radiology; Visualization;
Journal_Title :
Medical Imaging, IEEE Transactions on
