Scalable sparse shape composition and its application to liver surgical planning

Guotai Wang,Shaoting Zhang,Lixu Gu

DOI: https://doi.org/10.1109/ISBI.2014.6867887

2014-01-01

Abstract:The recently proposed Sparse Shape Composition (SSC) models shape prior as a sparse linear combination of existing shapes. It is effective to represent complex shape variations, with its ability to capture gross errors and preserve local details. However, SSC has low efficiency when dealing with large-scale training data, which adversely affects its more widespread clinical use. In this paper, we investigate efficient and scalable convex optimization methods and propose a nearly real-time SSC for large dataset. The new method solves the convex optimization problem in SSC by continuously transforming it into a series of simplified problems whose solution is fast to compute, without sacrificing the accuracy. It significantly speeds up the shape modeling process. When the repository's capacity is 10000, with 2000 vertices on each shape, the optimization can be solved by the new method in less than 10 seconds, nearly 2000 times faster than traditional method in SSC. Thus, it is more applicable in real-time clinical applications.