Using augmented reality to assist laparoscopic surgery

By using small incisions and a camera, laparoscopic surgery substantially reduces trauma compared to open surgery. However, in spite of the numerous recent technical achievements such as the high definition cameras, the laparoscopic approach remains challenging in many cases and is yet far from forming the dominant surgical approach. As they cannot directly touch the organs, a major difficulty for the surgeons is to localise the hidden structures, in particular the tumours to resect and the vessels to respect. Augmented reality from preoperative image data is a possible approach to aid localisation, by overlaying information extracted from the preoperative MRI or CT onto the intraoperative laparoscopic images during surgery. This raises a major computer vision problem: finding the non-rigid transformation from the preoperative 3D MRI or CT volume to the intraoperative 2D or 3D laparoscopic images. I will present our efforts in proposing concrete models and approaches to develop computer-assisted laparoscopic surgery systems and their clinical validation.


Adrien Bartoli is a Professor of Computer Science at Université Clermont Auvergne and a member of Institut Universitaire de France, currently on leave as research scientist at the University Hospital of Clermont-Ferrand and as Chief Scientific Officer at SurgAR. He is leading the EnCoV research group jointly with gynecologist surgeon Prof. Michel Canis. His main research interests are deformable 3D reconstruction in computer vision and computer-aided diagnosis and surgery. He is an Associate Editor for the International Journal of Computer Vision and for the Journal of Artificial Intelligence Research. He is or has been an Area Chair for the major conferences of computer vision, medical image analysis and augmented reality.