pFIRE

pFIRE is a parallel framework for the elastic registration of large images by the method of Barber and Hose. 1 2 It is implemented in C++ using the PETSc scientific toolkit to provide parallelised mathematical routines.

There are two ways to use pFIRE:

  1. Use the provided executable tools to interact with pFIRE from the command line. These tools expose the majority of pFIRE’s functionality through a simple interface. If you just have a pair of images to register, start here.

  2. Embed pFIRE functionality within your own code using the API. This is primarily intended for more experienced users and developers to embed elastic registration within larger workflows or to extend pFIRE’s functionality.

What is Elastic Registration?

Image registration is a process by which an image is transformed to match a second image as closely as possible. The image which is transformed is known as the moved image, and the target image to which is is matched the fixed image.

There are two types of image registration:

  1. Rigid registration uses global transformations which affect the whole image in the same way. Translation, scaling, linear shearing and rotation are all examples of rigid registration. These are relatively simple to compute but cannot describe all changes to an image, such as non-linear shearing or local deformations.

  2. Elastic registration creates a displacement map or field which describes how individual image pixels are moved to transform between the moved and fixed images. This map can have an arbitrary resolution and can therefore, in principle, describe any transformation of one image to another.

Elastic registration is far more computationally intensive than rigid registration, however, the ability to measure and describe nonlinear transformations is potentially extremely useful for e.g determination of stress inside a deformed mechanical structure.

1

Barber D, Hose D. Automatic segmentation of medical images using image registration: diagnostic and simulation applications. Journal of medical engineering & technology, 29(2), pp. 53-63, (2005), DOI:10.1080/03091900412331289889.

2

Barber DC, Oubel E, Frangi AF, Hose D. Efficient computational fluid dynamics mesh generation by image registration. Medical image analysis, 11(6), pp. 648–662, (2007), DOI:10.1016/j.media.2007.06.011.