Multi-Linear Subspace Learning Methods for Statistical Texture Atlases of the Liver

摘要

Digital atlases of the human anatomy are a new and hot topic in medical image analysis. The basic idea of the digital atlas is to capture the variability of an organ's location, shape and voxel intensity (texture) from a training set. In this paper, we present current progress toward constructing digital atlases of the liver and a new mathematic framework based on multi-linear subspace learning methods for medical volume analysis.