An ASP approach for arteries classification in CT scans

Automated segmentation of computed tomography (CT) scans is the first step in the pipeline for the interpretation and identification of potential pathologies in human organs. Several methods based on machine learning (ML) are currently available, even if their precision is still outperformed by medical doctors. In this field there are some intrinsic limitations to ML approaches, such as the following: cost and time to acquire high-quality annotated scans for training; and a remarkable high variability of organ morphology due to age, conditions, genetics and acquisition. This paper outlines a new methodology based on Answer Set Programming, which returns reliable, easy-to-program and explainable interpretations. In particular, we focus on the CT scan analysis and retrieval of tree-like structure, corresponding to main blood vessels (arteries) arrangement. The structure is compared to the knowledge base of vessels contained in anatomy textbooks. The mapping of vessel names is computed by an Answer Set Programming program. This preliminary step produces a robust input to a reasoner for the multi-organ labelling and localization problem.