Minimally invasive surgeries (MIS) are continuously evolving due to their potential high impact on improving patient management and overall cost effectiveness. The spearhead in current research and development on MIS is the incorporation of image-guidance and robot-assistance (IGRA) for improved visualization, high precision, and dexterity. With those features, MIS can offer us the needed methodologies to perform surgeries in beating heart, thereby substantially improving the provided health care and long term quality of life for the patients. Despite the potential benefits, MIS in a beating heart face unique challenges posed by the continuous motion of the heart and respiration. Therefore, it is necessary to track certain cardiac landmarks within this dynamic environment, while controlling a robotic device, in real-time. The most efficient way to remedy this challenge is to use imaging for both preoperative planning and intraoperative guidance.

The objective of this dissertation is to introduce a novel computational system for magnetic resonance imaging (MRI)-guided and robot-assisted interventions in a beating heart. In particular, we implemented our methodology for the clinical paradigm of transapical aortic valve implantation in a beating heart. This prototype system includes dedicated software modules that operate synergistically to control a semi-autonomous robotic device and to adjust on-the-fly the image acquisition parameters on the MR scanner to better suit the particular conditions of the interventions as they evolve. The approach is based on generating a dynamic "access corridor" from the apex of the heart to the aortic annulus. This corridor is updated on-the-fly as the procedure evolves to reflect the morphologic changes in the heart.

Our recent studies presented herein, illustrate that by combining the appropriate computational tools we can indeed develop such an integrated computational system. We also believe that the methodology introduced in this dissertation can be the basis and stepping stone for guiding surgeries in other moving anatomies with IGRA approaches. Moreover, the work in this project provides us with additional understanding and assessment of the feasibility and the true capabilities of IGRA surgeries and the potential of computational methods to alleviate the challenges associated with them.