Yav++ is a generic platform for the visualization and processing of volumetric medical images.
It is being currently developped at the INRIA research group Epidaure, in Sophia-Antipolis, France.
It structurs the code being developed inside the research group. It
avoids to have several implementations within the same research team,
creates collaboration rather than competition and also allows isolated
development by creation of new libraries.
CVS ( Concurrent Version System ) is used internally for sharing source code.
The primary targeted application of this platform was the segmentation of medical images based on 3D and 4D deformable models (,) where user interaction is required to guide the surface deformation during the deformation process. We have then extended its scope to solve different problems related to medical image registration and analysis.
It is a set of C++ libraries, linked with Tcl/Tk and OpenGL to provide scripting ability, graphical interface and visual interactivity. This was motivated by the belief that for many medical imaging applications, user interaction must be closely tighted with medical image processing.
This platform has an architecture which is somewhat similar to the Visualization Toolkit. Nevertheless, because it is based on the data-flow paradigm, VTK is not well suited for highly interactive application where computation and user interface are closely related. Furthermore, the yav++ platform is not targeted to be a general visualization software but dedicated to the processing of volumetric images and to the deformation of contours, surfaces or volumes.
This platform combines three characteristics that are required for developing large interactive applications : modularity, portability and scripting ability.