Model-based fusion of CT and non-contrasted 3D C-arm CT: Application to transcatheter valve therapies

In recent years transcatheter valve therapies are beginning to replace invasive surgical procedures. As there is no direct view and access to the affected anatomy advanced imaging techniques such as 3D rotational angiography (C-arm CT) and real-time fluoroscopy are used for intra-operative guidance. However, intra-operative modalities have limited image quality of the soft tissue and a reliable assessment of the cardiac anatomy can only be made by injecting contrast agent, which is harmful to the patient and requires complex acquisition protocols. We propose a novel method to align pre-operative and intra-operative data by using surrogate anatomical structures which are visible in both modalities without adding contrast agent. The trachea bifurcation model is used as a surrogate structure and the model parameters are estimated using robust machine learning algorithms. High quality patient specific models can be extracted from the pre-operative CT and mapped to the intra-operative 3D C-arm CT for guidance. In addition we learn a weighted mapping function for the trachea bifurcation model extracted from the pre-operative and intra-operative images which minimizes the mapping error in respect to the anatomy of interest. Experiments performed on 28 patient pairs of CT and contrasted 3D C-arm CT data sets assure an accuracy of the mapped aortic valve model of 9.08 ± 7.31 deg and 7.57 ± 3.22mm.