Introduction to deformable registration: transformations, metrics & inference - MISS 2014 (04)

Prof. Nikos Paragios (Ecole Centrale Paris, France) was one of the speakers of Medical Imaging Summer School (Sicily, July 28-August 1, 2014), here you can find his lecture 1: "Introduction to deformable registration: transformations, metrics & inference".

Abstract. Deformable registration/fusion is a pillar of biomedical image analysis. It consists of finding a nonlinear transformation that will map one multi-dimensional signal to another through a geometric transformation. The task is often expressed as an inverse modeling one where given a family of transformations and similarity criterion we seek the instance of this family that creates an optimal alignment between the two signals.

In this lecture we introduce the theoretical foundations of deformable registration. In particular we provide a generic mathematical framework to express existing methods and discuss stated of the art techniques in terms of (i) type of deformations, (ii) similarity criterion and (iii) optimization methods.

Shooting: Lorenzo Di Silvestro, Mauro Sodano
Editing and post-production: Lorenzo Di Silvestro, Agata Ventura