Magnetic resonance imaging is of great interest for interventional radiology due to its excellent soft tissue characterization, lack of ionizing radiation, multiplanar capabilities and ability to derive functional as well as anatomic information. However, performing real time MR guided procedures with the patient in the MR scanner poses additional challenges. This work addresses some of the key issues, including thermography in the presence of breathing and other sources of motion, tracking devices in the MR scanner and providing intuitive feedback for guidance to physicians, and fusing different types of MR information at different stages of a procedure to optimize the combined use of both real time and tissue characterization images.

The overall goal of this research project is to develop new methods and prototypes for:

• Quality control/ motion estimation for MR guided thermal therapy

• Automated image based device tracking for MR guided percutaneous procedures

• Intra-modality fusion for guiding MR procedures