Medical Image Analysis Seminar
-------------------------------------------------------------
Friday April 2nd 9:00 - 12:00, room 053 building 321



Virtual Liver Surgery Planning System using Augmented Reality
-------------------------------------------------------------
Milan Sonka, Iowa University

A system for virtual planning of liver tumor resections will be
discussed that consists of the environment for diaphragm, liver,
liver tumor, and liver vasculature segmentation. The second part
of the environment allows augmented reality planning of liver
tumor resections. The system is under development and many
individual modules will be discussed and their functionality
presented.



Detection and Quantification of Motion and Growth
using Non-rigid Registration
-------------------------------------------------------------
Daniel Rueckert, Imperial College

Three-dimensional (3D) and four-dimensional (4D) imaging of
dynamic structures like the heart is a rapidly developing area
of research in medical imaging. Recent advances in development
of MR imaging for fast spatio-temporal cardiac imaging have led
to an increased interest in the use of MR imaging for functional
analysis of the cardiovascular system. Segmentation of left and
right ventricle and modelling of myocardial motion provide
important information for quantitative functional analysis. In
this talk we show how non-rigid registration techniques can be
used to solve these problems in cardiac MR images. We will also
demonstrate how similar techniques can be used for the modelling
of temporal changes such as growth or atrophy in the brain.



Statistical Models of Appearance for Image Analysis
-------------------------------------------------------------
Tim Cootes, University of Manchester

Statistical models of shape and appearance have been shown to be
powerful tools for image interpretation, as they can explicitly
deal with the natural variation in objects of interest. Such
models can be built from suitably labelled training sets. Given
a model of appearance we can match it to a new image using the
efficient optimisation algorithms, which seek to minimise the
difference between a synthesized model image and the target
image. This talk will describe the approach and recent
developments, including new work on automatically finding
correspondences across training sets of unlabelled images.
Examples will be included from the domains of face
interpretation and medical image analysis.