In this talk, we present our latest work on learning reconstructions of low-dose computed tomography data. We focus on two methods to decrease the radiation dose: x-ray tube current reduction, reducing the signal-to-noise ratio, and x-ray beam interruption, which undersamples data and results in images with aliasing artifacts. Our reconstruction results using trainable variational networks enable higher radiation dose reductions and/or increase the image quality for a given dose.

This presentation is part of Minisymposium “MS50 - Analysis, Optimization, and Applications of Machine Learning in Imaging (3 parts)”
organized by: Michael Moeller (University of Siegen) , Gitta Kutyniok (Technische Universität Berlin) .

Authors:

Erich Kobler (Graz University of Technology)

Thomas Pock (Graz University of Technology)

Kerstin Hammernik (Graz University of Technology)

Teresa Klatzer (Graz University of Technology)

Florian Knoll (New York University)

Matthew Muckley (New York University)

Daniel Sodickson (New York University)

Ricardo Otazo (New York University)

Baiyu Chen (New York University)

Keywords:

computed tomography, deep learning, image reconstruction, machine learning