Coronary artery tree tracking with robust junction detection in 3D CT Angiography

Computed Tomography Angiography (CTA) of the heart is a non-invasive procedure to rule out coronary artery disease or measure its extent and plan treatments and interventions. The need for coronary tree tracking methods that require minimum human interaction and produce accurate and robust measurements is therefore of great clinical importance. In this work we present a probabilistic coronary artery tree tracking method incorporating efficient statistical vessel junction detection. The method uses a single initial seed point at the top of the tree and recursively tracks every vessel branch based on a Sequential Monte Carlo framework. All junctions along the vessel tree are detected automatically. Performance of the proposed method is evaluated using the Rotterdam public coronary tracking evaluation framework that includes 24 cardiac CTA datasets. Our method yields accurate and robust tracking results with 92.7% average overlap with expert-defined centerlines.