An image-navigation system for vascular interventional surgery robot

Abstract

A new image-navigation system for vascular interventional surgery robot (VISR) is proposed. This system can provide robot with navigation information based on projection images without other external tracking devices. Firstly, projection images acquired from digital subtraction angiography (DSA) machine are calibrated by iterative optimization process. Secondly, 3D vascular model is reconstructed from two calibrated projection images so that surgeons observe pathological vasculature and make surgical planning easily. Thirdly, roadmap view is provided by superimposing 3D vascular model and projection images, which allows surgeon to quantify robotic motion accurately. Finally, physical phantom experiments are performed to validate the 3D reconstruction accuracy by comparing the distance between reconstruction results with actual values. Animal experiments show that image-navigation system can satisfy the demanding of VISR.