Atlas to Image-with-Tumor Registration based on Demons and Deformation Inpainting
Abstract
This paper presents a method for nonlinear registration of images, where there exists no one-to-one correspondence in parts of the image. Such a situation occurs for instance in the case where an atlas of normal anatomy shall be matched to pathological data, such as tumors, resections or lesions. Our idea is to use local con dence weights and to model pathological regions with zero con dence. We integrate this concept into the e cient and publicly available di eomorphic demons registration framework. Finally, we show that this process better captures deformations in high-con dence regions than without using the proposed modi cation. Furthermore, it is easy to implement and runs faster than previous approaches.
Origin : Files produced by the author(s)
Loading...