The last unsolved problem, which was listed in the previous section, is being addressed at present. It has also led us to realisation of a fundamental problem with the way in which we describe models and, more importantly, our definition of model cost.
Having solved some of the problems that we had encountered, we plan
to extend our method to deal with 2-D data and then preferably apply
it to bio-medical images. This will make a comparison between our
results and results produced by other methods possible. We will then
be able to argue that our method indeed does what it was set to do
and that the one unique solution we get is most preferable.