Learnable Slice-to-volume Reconstruction for Motion Compensation in Fetal Magnetic Resonance Imaging
Jehn C, Müller J, Kainz B (2023)
Publication Type: Conference contribution
Publication year: 2023
Journal
Publisher: Springer Science and Business Media Deutschland GmbH
Pages Range: 25-31
Conference Proceedings Title: Informatik aktuell
Event location: Braunschweig
ISBN: 9783658416560
DOI: 10.1007/978-3-658-41657-7_10
Abstract
Reconstructing motion-free 3D magnetic resonance imaging (MRI) volumes of fetal organs comes with the challenge of motion artefacts due to fetal motion and maternal respiration. Current methods rely on iterative procedures of outlier removal, super-resolution (SR) and slice-to-volume registration (SVR). Long runtimes and missing volume preservation over multiple iterations are still challenges for widespread clinical implementation. We envision an end-to-end learnable reconstruction framework that enables faster inference times and that can be steered by downstream tasks like segmentation. Therefore, we propose a new hybrid architecture for fetal brain reconstruction, consisting of a fully differentiable pre-registration module and a CycleGAN model for 3D image-toimage translation that is pretrained on our custom-generated dataset of 209 pairs of low-resolution (LoRes) and high-resolution (HiRes) fetal brain volumes. Our results are evaluated quantitatively with respect to five different similarity metrics. We incorporate the learned perceptual image patch similarity (LPIPS) metric and apply it to quantify volumetric image similarity for the first time in literature. Furthermore, we evaluate the model outputs qualitatively and conduct an expert survey to compare our method’s reconstruction quality to an established approach.
Authors with CRIS profile
Constantin Jehn
Professur für Sensorische Neurotechnologie
Johanna Müller
Professur für Image Data Exploration and Analysis
Bernhard Kainz
Professur für Image Data Exploration and Analysis
How to cite
APA:
Jehn, C., Müller, J., & Kainz, B. (2023). Learnable Slice-to-volume Reconstruction for Motion Compensation in Fetal Magnetic Resonance Imaging. In Thomas M. Deserno, Heinz Handels, Andreas Maier, Klaus Maier-Hein, Christoph Palm, Thomas Tolxdorff (Eds.), Informatik aktuell (pp. 25-31). Braunschweig, DE: Springer Science and Business Media Deutschland GmbH.
MLA:
Jehn, Constantin, Johanna Müller, and Bernhard Kainz. "Learnable Slice-to-volume Reconstruction for Motion Compensation in Fetal Magnetic Resonance Imaging." Proceedings of the Bildverarbeitung für die Medizin Workshop, BVM 2023, Braunschweig Ed. Thomas M. Deserno, Heinz Handels, Andreas Maier, Klaus Maier-Hein, Christoph Palm, Thomas Tolxdorff, Springer Science and Business Media Deutschland GmbH, 2023. 25-31.
BibTeX: Download