What technical barriers exist for real‐time fluoroscopic and video image overlay in robotic surgery?

Abstract

BACKGROUND: A hypothetical advantage of the da Vinci console is its ability to integrate multiple visual data sources. Current platforms for augmented reality surgery fuse pre-operative radiographic studies but are limited with their ability to update with intra-operative imaging. The aim of our study was to evaluate the feasibility of real-time radiographic image overlay with current technology. METHODS: S-video composite output from a fluoroscopic C-arm was superimposed onto the video output of the da Vinci device. Image superimposition disparity measurements were evaluated in a bench model. The feasibility of robotic dissection assisted by real-time cholangiogram and intravenous pyelogram was evaluated. RESULTS: Image alignment resulted in a radiographic blind spot and image disparity with severely limited application in an in vivo model. CONCLUSIONS: External collisions of the robotic device and visual disparity in multiple planes negate the current implementation of fluoroscopic overlay and will require more elegant methods of computer-assisted registration.