The objective of the project is to develop computational shape models of brain structures for shape analysis. We develop a deformable template model framework based on “progressive surface deformation” method (Kim et al, IEEE-TMI: 34(6), 2015) to reconstruct target shapes with point-wise correspondence to template models robustly by minimizing the distortion of template models. Our toolbox contains a set of tools for constructing mean images, reconstructing individual surface models, computing average surface and finally computing local deformity.

This toolbox was developed by KAIST and University of Edinburgh (http://cgv.kaist.ac.kr/brain)