Enabling technologies for ultrasound imaging in computer-assisted intervention

There has been an increasing interest in minimally invasive US-guided interventions that require precise placement of a surgical tool (needle, tissue ablator, etc.) to the anatomical target. To address this problem, I have developed a collection of novel technologies and integrated those in prototype systems.; I introduce multiple system embodiments that involve robotics, tracking, anatomical modeling, ultrasound image processing, and elasticity monitoring. The underlying themes in these systems are (1) simultaneous tracking of surgical tool with respect to the US images and (2) monitoring physiological changes, specifically tissue coagulation, throughout the procedure.; My main contribution to the first theme is inventing a robust method for 2D and 3D ultrasound probe calibration with a closed-form solution. As a result, one can discern the unknown spatial transformation between image pixels and tracker coordinates in real-time, in-vivo while the patient is being scanned. I also introduced a novel methodology for in-vivo quality control of tracked US systems, by capturing system errors that manifest in changes of calibration parameters. The concept, mathematical formulation, and experimental evaluation are presented and demonstrated in-vitro experiments.; With respect to the second theme, I presented a rapid US-based approach to monitor ablative therapy by optimizing shape parameters. My method involves the integration of a biomechanical computational model of the tissue, a correlation approach to estimate and track tissue deformation, and an optimization method to solve the inverse problem of recovering the shape parameters in the volume of interest. I demonstrate convergence and reliability on simulated data and present successful monitoring of tissue ablation of ex-vivo bovine liver samples.