Deep Learning for Photoacoustic Tomography from Sparse DataMS65

We improve image reconstruction from sparse data in photoacoustic tomography using a deep convolutional network. The weights of the convolutional network are adjusted prior to the actual image reconstruction through training with pairs of reconstructions including artifacts and the corresponding artifact-free images. We demonstrate with simulated and experimental data that the proposed approach provides reconstructed images with a quality comparable to state of the art approaches and reduced numerical costs.

This presentation is part of Minisymposium “MS65 - Machine learning techniques for image reconstruction (2 parts)
organized by: Markus Haltmeier (University Innsbruck) , Linh Nguyen (University of Idaho) .

Johannes Schwab (Universität Innsbruck)
Stephan Antholzer (University of Innsbruck)
Markus Haltmeier (University Innsbruck)
Robert Nuster (Karl-Franzens-Universität Graz )
deep learning, image reconstruction, machine learning