1444 IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 59, NO. 5, MAY 2012 US–Fluoroscopy Registration for Transcatheter Aortic Valve Implantation Pencilla Lang ∗ , Student Member, IEEE, Petar Seslija, Michael W. A.Chu, Daniel Bainbridge, Gerard M. Guiraudon, Douglas L. Jones, and Terry M. Peters, Fellow, IEEE Abstract—Transcatheter aortic valve implantation is a mini- mally invasive alternative to open-heart surgery for aortic stenosis in which a stent-based bioprosthetic valve is delivered into the heart on a catheter. Limited visualization during this procedure can lead to severe complications. Improved visualization can be provided by live registration of transesophageal echo (TEE) and fluoroscopy images intraoperatively. Since the TEE probe is al- ways visible in the fluoroscopy image, it is possible to track it us- ing fiducial-based single-perspective pose estimation. In this study, inherent probe tracking performance was assessed, and TEE to fluoroscopy registration accuracy and robustness were evaluated. Results demonstrated probe tracking errors of below 0.6 mm and 0.2 ◦ , a 2-D RMS registration error of 1.5 mm, and a tracking fail- ure rate of below 1%. In addition to providing live registration and better accuracy and robustness compared to existing TEE probe tracking methods, this system is designed to be suitable for clinical use. It is fully automatic, requires no additional operating room hardware, does not require intraoperative calibration, maintains existing procedure and imaging workflow without modification, and can be implemented in all cardiac centers at extremely low cost. I. INTRODUCTION A ORTIC stenosis, a condition in which the aortic valve calcifies and obstructs flow from the heart, is the most common acquired heart valve disease in the western world [1]. Manuscript received October 22, 2011; revised January 13, 2012; accepted February 15, 2012. Date of publication February 28, 2012; date of current ver- sion April 20, 2012. This work was supported by the Canadian Foundation for Innovation under Grant CFI#20994, the Canadian Institute of Health Research under Grant MOP 179298 and M.D./Ph.D. Graduate Training Award, and the Ontario Research Fund under Grant RE-02-038. Asterisk indicates correspond- ing author. ∗ P. Lang is with the Image Research Laboratories, Robarts Research Insti- tute, University of Western Ontario, London, ON N6A 5K8, Canada (e-mail: plang@imaging.robarts.ca). P. Seslija and T. M. Peters are with the Image Research Laboratories, Ro- barts Research Institute, University of Western Ontario, London, ON N6A 5K8, Canada (e-mail: pseslija@imaging.robarts.ca; tpeters@robarts.ca). M. W. A. Chu is with the Division of Cardiac Surgery, University of Western Ontario, London, ON N6A 5K8, Canada (e-mail: michael.chu@lhsc.on.ca). D. Bainbridge is with the Department of Anesthesia and Perioperative Medicine, University of Western Ontario, London, ON N6A 5K8, Canada (e- mail: daniel.bainbridge@lhsc.on.ca). G. M. Guiraudon is with the Canadian Surgical Technologies and Advanced Robotics, London Health Sciences Centre, London, ON N5Z 2C1, Canada (e-mail: gerard.guiraudon@lhsc.on.ca). D. L. Jones is with the Department of Pharmacology and Physiology, University of Western Ontario, London, ON N6A 5K8, Canada (e-mail: doug.jones@schulich.uwo.ca). Digital Object Identifier 10.1109/TBME.2012.2189392 Fig. 1. (a) Edwards SAPIEN Transcatheter aortic valve—xenogenic cusps mounted on an expandable stent. (b) Representative example of a fluoroscopy image used in TAVI. Surgical valve replacement is required to treat symptomatic, severe aortic stenosis [2]. In patients with advanced age or sig- nificant comorbidities, the operative risk can be excessive [3]. Transcatheter aortic valve implantation (TAVI) is a less invasive alternative to open heart surgery in which a stented biopros- thetic valve is delivered inside a catheter through the femoral artery (transfemoral) or the left ventricular apex (transapical), displacing and functionally replacing the native valve. The stent relies upon radial forces to seat the prosthesis within the native aortic annulus [see Fig. 1(a)]. TAVI is a minimally invasive pro- cedure that does not require cardiopulmonary bypass or large incisions, making it a growing alternative for patients deemed inoperable [4]. Since Cribier et al. [5] described the first TAVI implantation in a human patient in 2002, the number of TAVI cases performed has grown exponentially, with more than 20 000 cases to date. However, the risk of periprocedural death remains significant, at 5.0% after 30 days [6]. The majority of TAVI complica- tions, including stent malpositioning and migration, coronary obstruction, paravalvular leak, atrioventricular block and aor- tic root rupture, are related to poor positioning of the valve at the time of deployment. Positioning of a transcatheter valve re- quires both the anatomical structures of the aorta and the valved stent to be visualized intraoperatively. The majority of centers rely primarily on single-plane fluoroscopy or root aortography (contrast-enhanced fluoroscopy) to visualize placement of the stent within the aortic valve annulus prior to deployment [7], [8] [see Fig. 1(b)]. Fluoroscopy provides only gross imaging of the aortic valve without 3-D context, leaving the surgeon blind to surrounding structures. Furthermore, contrast-induced nephro- toxicity is a frequent problem, and has been reported in 11–24% of patients [9]–[11]. Better intraoperative image guidance is required to reduce TAVI complications. 0018-9294/$31.00 © 2012 IEEE