Treelike neural network for brain magnetic resonance image segmentation
Iren Valova, Yukio Kosugi · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 1997
In this paper we report the application of neural trees for image segmentation of magnetic resonance (MR) images. The network, built up during training, effectively partitions the feature space into subregions and each final subregion is assigned a class label according to the data routed to it. As the tree grows, the number of training data for each node decreases, which results in less weight update epochs and decreases the time consumption. The growing algorithm is based on depth-first search, which is guaranteed to find deep solutions, i.e. linearly non-separable classes. We also introduce a measure for estimation of the best-fit neuron to split the feature space at each tree node. This eliminates the necessity for postponing perturbance of the hyperplanes and proves essential for solving linearly non-separable difficult tasks. The network performance is compared to the multilayered perceptron (MLP) over the white/gray matter MRI segmentation problem.