Deep Learning Computed Tomography: Learning Projection-Domain Weights From Image Domain in Limited Angle Problems

In this paper, we present a new deep learning framework for 3-D tomographic reconstruction. To this end, we map filtered back-projection-type algorithms to neural networks. However, the back-projection cannot be implemented as a fully connected layer due to its memory requirements. To overcome this...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on medical imaging Jg. 37; H. 6; S. 1454 - 1463
Hauptverfasser: Wurfl, Tobias, Hoffmann, Mathis, Christlein, Vincent, Breininger, Katharina, Huang, Yixin, Unberath, Mathias, Maier, Andreas K.
Format: Journal Article
Sprache:Englisch
Veröffentlicht: United States IEEE 01.06.2018
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:
Algorithms
Apodization
Computed tomography
Computer applications
Deep Learning
Geometry
Humans
Image reconstruction
Iterative methods
Machine learning
Neural networks
Optimization
Projection
Radiographic Image Interpretation, Computer-Assisted - methods
Reconstruction algorithms
Tomography, X-Ray Computed - methods
ISSN:0278-0062, 1558-254X, 1558-254X
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In this paper, we present a new deep learning framework for 3-D tomographic reconstruction. To this end, we map filtered back-projection-type algorithms to neural networks. However, the back-projection cannot be implemented as a fully connected layer due to its memory requirements. To overcome this problem, we propose a new type of cone-beam back-projection layer, efficiently calculating the forward pass. We derive this layer's backward pass as a projection operation. Unlike most deep learning approaches for reconstruction, our new layer permits joint optimization of correction steps in volume and projection domain. Evaluation is performed numerically on a public data set in a limited angle setting showing a consistent improvement over analytical algorithms while keeping the same computational test-time complexity by design. In the region of interest, the peak signal-to-noise ratio has increased by 23%. In addition, we show that the learned algorithm can be interpreted using known concepts from cone beam reconstruction: the network is able to automatically learn strategies such as compensation weights and apodization windows.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:0278-0062
1558-254X
1558-254X
DOI:10.1109/TMI.2018.2833499