Reasoning-based patient classification for enhanced medical image annotation

Abstract

Medical imaging plays an important role in today's clinical daily tasks, such as patient screening, diagnosis, treatment planning and follow up. But still a generic and flexible image understanding is missing. Although, there exist several approaches for semantic image annotation, those approaches do not make use of practical clinical knowledge, such as best practice solutions or clinical guidelines. We introduce a knowledge engineering approach aiming for reasoning-based enhancement of medical images annotation by integrating practical clinical knowledge. We will exemplify the reasoning steps of the methodology along a use case for automatic lymphoma patient staging. © 2010 Springer-Verlag.