Toward Practical and Accurate Touch-Based Image Guidance for Robotic Partial Nephrectomy

Partial nephrectomy involves removing a tumor while sparing surrounding healthy kidney tissue. Compared to total kidney removal, partial nephrectomy improves outcomes for patients but is underutilized because it is challenging to accomplish minimally invasively, requiring accurate spatial awareness of unseen subsurface anatomy. Image guidance can enhance spatial awareness by displaying a 3D model of anatomical relationships derived from medical imaging information. It has been qualitatively suggested that the da Vinci robot is well suited to facilitate image guidance through touch-based registration. In this paper we validate and advance this concept toward real-world use in several important ways. First, we contribute the first quantitative accuracy evaluation of touch-based registration with the da Vinci. Next, we demonstrate real-time touch-based registration and display of medical images for the first time. Lastly, we perform the first experiments validating use of touch-based image guidance to improve a surgeon's ability to localize subsurface anatomical features in a geometrically realistic phantom.