Peter Maria Johannes Rongen

Detection of electrophysiology catheters in noisy fluoroscopy images

Cardiac catheter ablation is a minimally invasive medical procedure to treat patients with heart rhythm disorders. It is useful to know the positions of the catheters and electrodes during the intervention, e.g. for the automatization of cardiac mapping. Our goal is therefore to develop a robust image analysis method that can detect the catheters in X-ray fluoroscopy images. Our method uses steerable tensor voting in combination with a catheter-specific multi-step extraction algorithm. The evaluation on clinical fluoroscopy images shows that especially the extraction of the catheter tip is successful and that the use of tensor voting accounts for a large increase in performance.

Spatiotemporal multiscale vessel enhancement for coronary angiograms

In coronary angiography, coronaries are imaged filled by a contrast medium which is injected through a catheter. To increase vessel visibility relative to surrounding structures, a background-less, subtraction-like appearance of the angiograms may be desired. This paper describes algorithms to increase vessel contrast and to attenuate background. Due to the strong motion in coronary angiograms, direct subtraction of a mask image acquired initially without contrast agent cannot be applied. We therefore distinguish vessels from background by their contrast, their size and their motion. These criteria are evaluated on a multiscale structure. Enhancement is then applied at locations which are likely to contain vessels. To avoid unacceptable noise boosting, we integrate a multiscale noise reduction filter into this concept. Both performance and computational simplicity make our algorithms attractive.

3D catheter reconstruction using non-rigid structure-from-motion and robotics modeling

Surgical guidance during minimally invasive intervention could be greatly enhanced if the 3D location and orientation of instruments, especially catheters, is available. In this paper, we present a new method for the 3D reconstruction of deforming curvilinear objects such as catheters, using the framework of Non-Rigid Structurefrom- Motion (NRSfM). We combine NRSfM with a kinematics model from the field of Robotics, which provides a low-dimensional parametrization of the object deformation. This is used in the context of an X-ray imaging system where multiple views are acquired with a small view separation. We show that using such a kinematics model, a non-linear optimization scheme succeeds in retrieving the deformable 3D pose from the 2D projections. Experiments on synthetic and real X-ray data show promising results of the proposed method as compared to state-of-the-art NRSfM.