Machine learning for imaging problems with limited dataMS7

In recent years, machine learning has proved to be successful in several imaging fields. In this talk, different approaches for applying machine learning to challenging imaging problems involving limited data will be discussed. For example, in low-dose tomography problems, neural networks can be used to improve reconstruction quality, enabling analysis of new types of samples. Results will be shown for various problems, and important practical considerations, e.g. computational requirements, will be discussed.

This presentation is part of Minisymposium “MS7 - Limited data problems in imaging (2 parts)
organized by: Bernadette Hahn (University of Würzburg) , Gaël Rigaud (Saarland University) , Jürgen Frikel (OTH Regensburg) .

Daniel Pelt (CWI)
Allard Hendriksen (CWI)
computed tomography, deep learning, image enhancement, image reconstruction, image segmentation, machine learning