Two-stage neural network for volume segmentation of medical images

A new system to segment and label CTrMRI brain slices using feature extraction and unsupervised clustering is Ž . presented. Each volume element voxel is assigned a feature pattern consisting of a scaled family of differential geometrical invariant features. The invariant feature pattern is then assigned to a specific region using a two-stage neural network Ž . system. The first stage is a self-organizing principal components analysis SOPCA network that is used to project the feature vector onto its leading principal axes found by using principal components analysis. This step provides an effective Ž . basis for feature extraction. The second stage consists of a self-organizing feature map SOFM which automatically clusters the input vector into different regions. A 3D connected component labeling algorithm is then applied to ensure region connectivity. We demonstrate the power of this approach to volume segmentation of medical images. q 1997 Elsevier Science B.V.