Faculty of Informatics Vienna University of Technology Institute of Computer Aided Automation PRIP Home PRIP Home
Personal tools
You are here: Home Teaching PRIP-Talks Tanacs Attila: Registration methods in medical image processing
Navigation
 

Tanacs Attila: Registration methods in medical image processing

What
  • Presentation
When May 26, 2009
from 04:00 pm to 04:45 pm
Where Sem 183/2
Add event to calendar vCal
iCal
There is an increasing number of applications that require accurate aligning of one image with another taken from different viewpoints, by different imaging devices, and/or at different times. When the image is taken from an object with exactly known geometry, the mapping between the model and the image can also be established. Thus, the task of registration in a general sense is to establish the geometrical correspondence between image and/or geometrical information contents.

Registration is an important task in medical image processing. By aligning different images it is possible e.g., to monitor changes in size, shape, or image intensity over time, to combine information from multiple imaging modalities e.g., when relating functional information from nuclear medicine images to anatomy delineated in high-resolution MR or CT images. By relating preoperative images and surgical plans to the physical reality of the patient in the operating room, the surgical intervention can be controlled and monitored, the model of the operating tool can be displayed in the coordinate system of the image.

In the talk, I briefly introduce our previous results in point-based registration, automatic voxel similarity-based registration, and planning and executing percutaneous therapies using localization frames.

I also introduce our current research topic related to shape matching. Domokos et.al. proposed an extension to the parametric estimation method of Francos et.al. to deal with affine matching of binary shapes. Parametric estimation methods have the advantage of providing accurate and computationally simple solution, avoiding both the correspondence problem as well as the need for optimization. We extended this binary approach by investigating the case when the segmentation method is capable of producing fuzzy object descriptions instead of a binary result. In a study of 2000 pairs of synthetic images we observed the effect of the number of quantization levels of the fuzzy membership function to the precision of image registration and we compared the results with the binary case.

In the beginning of the talk I briefly introduce our department, and some of the research and industrial projects our group was involved in.
Document Actions