Our group is developing a simulation tool for radio frequency ablation and a computer-aided-diagnosis tool for the assessment of atherosclerosis in coronary arteries. In both tools, knowledge about the vascular structure of the coronary arteries is needed for further processing:
As blood-flow is the most relevant factor in the performance of radio frequency ablation, knowledge about the exact location of the major neighboring vessels significantly improves the reliability of simulations. For the assessment of atherosclerosis in coronary arteries, segmentation of the vessels provides the fundamental basis for a detailed analysis of the vessel′s composition.
The goal of this project is the development of a segmentation algorithm for coronary arteries in MSCT datasets that simultaneously extracts the coronary artery tree and provides a rough estimate of the vessel′s centerline.
MeVisLab − a fast prototyping development environment for medical image processing and visualization − is an excellent framework for the implementation of this work. Much functionality (e.g., loading and visualization of data) is already provided in the form of modules, which can graphically be linked to your own modules tol be implemented in C++. Due to our close collaboration with the Unispital Zurich, we have access to a wide range of datasets on which the algorithms can be developed and evaluated.
|Supervisor:||Dr. Esra Neufeld, ITIS Foundation, Tel.: +41 44 245 96 98, email@example.com|
|Type of work:||40% theory, 60% implementation (MeVisLab/C++)|
|Professor:||Gábor Székely, Computer Vision Lab, ETHZ