Tutorial 1: The Hitchhiker's Guide to Biomedical Imaging

Daniel Tenbrinck, École Nationale Supérieure d'Ingénieurs de Caen, France

Biomedical image analysis is an important field in life sciences which strongly depends on the advances in image processing and computer vision. Due to the non-standard imaging techniques used for data acquisition there are a variety of challenging problems to face when working with biomedical data. In order to master typical image analysis tasks, e.g., segmentation or motion estimation, one has to investigate innovative ways to describe the given data and subsequently propose feasible solutions applicable in basic research and daily clinical routine.

This tutorial aims to provide an overview of the challenges of biomedical imaging for computer vision and simultaneously give insightful ideas to deal with problems such as physical noise perturbations, structural artifacts, inhomogeneity and fuzzy edges. We provide a tour through the universe of biomedical imaging and head on this way for popular imaging modalities and their respective characteristics, e.g., ultrasound imaging, positron emission tomography, or fluorescence microscopy. Generally applicable techniques are illustrated on real application data. This tutorial is meant as a useful guide for both researchers already working in this expanding field of computer vision and those who dare to explore this fascinating topic for the first time.