Project Description
The aim of this project is to develop a technique to correlate different 3D medical images for scar localization and characterization which is robust in clinical practice, leading to high precision cardiac stereotactic body radiation therapy (cSBRT).
Different diagnostic modalities for cardiac SBRT are available today, but the extraction of all required information from a single imaging modality is currently impossible. Each modality, primarily invasive 3D electroanatomic maps (EAMs), computed tomography (CT) and magnetic resonance imaging (MRI), requires a specific workflow to use them to define the radiation target. Each of these modalities focuses on a particular type of information, and no single modality contains all the relevant data. While EAMs emphasize anatomical (size, location) and functional (electrical) information about the myocardial region of interest, namely the arrhythmogenic substrate, CT or MRI contain precise information about the transmurality of scar tissue, that cannot be assessed with EAM mapping alone. The correlation of EAM information with CT/MRI images obtained as treatment planning or during the cSBRT remains challenging and is poorly standardized today. Therefore, the goal of the project is the development of a novel workflow to allow for the integration of all noteworthy information from these diverse modalities. This will have immediate consequences on cSBRT efficacy and safety and strengthen clinical accuracy significantly.
In a step-by-step approach, our interdisciplinary project aims to implement different algorithms to develop a generic and reproducible model for cSBRT target definition and treatment planning, based on invasive EAMs and on scar detection using late iodine enhancement CT (LIE-CT) or late gadolinium enhancement MRI (LGE-MRI). In a first step, printed 3D objects will be used to acquire and process 3D point cloud data to develop an optimized reconstruction and fusion approach of purely anatomical information. Second, we will perform a state-of-the-art substrate based cSBRT with precision image integration including functional cardiac information (EA mapping and LIE-CT imaging) for a porcine ischemic heart failure (HF) model. Third, we will perform a retrospective and last a prospective data analysis and ultimately treatment planning of cSBRT in humans using our developed algorithm.
