Theme: Registration


 

Registration

Image registration or alignment is the process of spatially mapping images of different subjects at different time points. We develop novel registration methods for analyzing large population data by using the spatial mappings to determine changes in MR properties and spatial deformations. Our registration methodologies also allow us to develop atlases which describe the likely images and structure that appear in a given population through the computation of diffeomorphisms between subject images and a central image.

 

 theme-registration

Registration results of neonates mapped to two-year olds. From left to right: (a) neonatal T1 image after ane registration, (b) reference T1 image at 2-years, followed by (c) neonatal T1 after deformable mutual information registration using B-splines, and (d) after combined probabilistic and geometric registration. From top to bottom: subject 0012, 0102, 0106.

 

Publications:

  • L.Ha, M. Prastawa, G. Gerig, J.H. Gilmore, C.T. Silva, S. Joshi. “Image Registration Driven by Combined Probabilistic and Geometric Descriptors,” In Med Image Comput Comput Assist Interv., Vol. 13, No. 2, pp. 602--609. 2010. PubMed ID: 20879365
  • S. Durrleman, P. Fillard, X. Pennec, A. Trouvé, N. Ayache. “Registration, Atlas Estimation and Variability Analysis of White Matter Fiber Bundles Modeled as Currents,” In NeuroImage, pp. (in press). 2010.
  • L. Ha, M. Prastawa, G. Gerig, J.H. Gilmore, C.T. Silva, S. Joshi. “Image Registration Driven by Combined Probabilistic and Geometric Descriptors,” In Med Image Comput Comput Assist Interv., Vol. 13, No. 2, pp. 602--609. 2010. PubMed ID: 20879365

 

Quick Links

 

Research Projects

 

PIs