Using multidimensional scaling to assess shape differences of human corpus callosum

The corpus callosum (CC) is the largest interhemispheric commissure in the human brain and of great interest in child psychiatry imaging studies. The article presents a fully automated method of revealing shape difference of CC using multidimensional scaling technique. When applied to image analysis, multidimensional scaling (MDS) classifies objects based on a shape similarity metric. It is a process that simplifies a similarity measurement of high dimensional objects into a configuration of points in a lower Euclidean space, typically of two or three-dimension for easy visualization. We use the simplest form of MDS to demonstrate the feasibility of the technique in a fully automated process.