Image-based Intraluminal Contact Force Monitoring in Robotic Vascular Navigation
Embolization, stroke, ischaemic lesion, and perforation remain significant concerns in endovascular interventions. Intravascular sensing of tool interaction with the arteries is advantageous to minimize such complications and enhance navigation safety. Intraluminal information is currently limited due to the lack of intravascular contact sensing technologies. We present monitoring of the intraluminal tool interaction with the arterial wall using an image-based estimation approach within vascular robotic navigation. The proposed image-based method employs continuous finite element simulation of the tool using imaging data to estimate multi-point forces along tool-vessel wall interaction. We implemented imaging algorithms to detect and track contacts, and compute pose measurements. The model is constructed based on the nonlinear beam element and flexural rigidity profile over the tool length. During remote cannulation of aortic arteries, intraluminal monitoring achieved tracking local contact forces, building a contour map of force on the arterial wall and estimating tool structural stress. Results suggest that high risk intraluminal forces may happen even with low insertion force. The presented online monitoring system delivers insight into the intraluminal behavior of endovascular tools and is well suited for intraoperative visual guidance for the clinician, robotic control of vascular procedures and research on interventional device design.