Towards automated image understanding

In this paper, we discuss an approach to the pattern synthesis and analysis of biological shapes which show a lot of variability, but at the same time also show a characteristic structure. This structure is captured by way of 'shape classes' which are constructed via a deformable template. Examples of pattern analysis in both two and three dimensions are presented. We define what we call intrinsic and extrinsic understanding of images and apply this to the detection of abnormalities.