Comparing and combining complementary information from different breast imaging devices is of high interest for the development of new imaging modalities respectively for a precise diagnosis of breast cancer. Unfortunately, different devices typically image the breast in a considerably different configurations (e.g. 2D vs. 3D, compression vs. prone).

To overcome these differences we develope model-based registration algorithms, which are able to predict the spatial relationship between the images, e.g. between 2D mammograms and 3D MRI images (Figure 1). The registration is based on a biomechanical model of the breast which enables Finite Element simulation of the breast deformation. Furthermore image segmentation and optimization of registration and model parameters are within the scope of our work.

Basic Processing steps of the model-based registration. 3D data (here: MRI volume) ist used to automatically generate a FEM model of the breast, which is afterwards used to carry out a deformation simulation of the mammographic compression. The resulting configuration of the breast can then be compared directly to the according mammogram

Figure 1: Basic Processing steps of the model-based registration. 3D data (here: MRI volume) ist used to automatically generate a FEM model of the breast, which is afterwards used to carry out a deformation simulation of the mammographic compression. The resulting configuration of the breast can then be compared directly to the according mammogram

 

Intuitive visualization of the registration results is the key to gain maximal profit from multimodal information for radiologists. Therefore we emphasize on the developement of adequate visualization techniques (e.g. image fusion, Figure 2 left) and interactive viewer software for multimodal diagnosis (e.g. including linked cursors, Figure 2 right). 

 Intuitive visualization of the image registration by image fusion (left) and linked cursors (right)

Figure 2: Intuitive visualization of the image registration by image fusion (left) and linked cursors (right)

 

Articles about the project

 

Publications

see USCT publication list

Contacts

Copyright by SWM, KIT – Universität des Landes Baden-Württemberg und nationales Forschungszentrum in der Helmholtz-Gemeinschaft
Templates Joomla 1.7 by Wordpress themes free