Show simple item record

dc.contributor.author
Bernhardt, Melanie
dc.contributor.author
Vishnevskiy, Valery
dc.contributor.author
Rau, Richard
dc.contributor.author
Goksel, Orcun
dc.date.accessioned
2020-12-15T07:45:17Z
dc.date.available
2020-12-13T04:13:53Z
dc.date.available
2020-12-15T07:45:17Z
dc.date.issued
2020-12
dc.identifier.issn
0885-3010
dc.identifier.issn
1525-8955
dc.identifier.other
10.1109/TUFFC.2020.3010186
en_US
dc.identifier.uri
http://hdl.handle.net/20.500.11850/456117
dc.description.abstract
Speed-of-sound (SoS) has been shown as a potential biomarker for breast cancer imaging, successfully differentiating malignant tumors from benign ones. SoS images can be reconstructed from time-of-flight measurements from ultrasound images acquired using conventional handheld ultrasound transducers. Variational networks (VNs) have recently been shown to be a potential learning-based approach for optimizing inverse problems in image reconstruction. Despite earlier promising results, these methods, however, do not generalize well from simulated to acquired data, due to the domain shift. In this work, we present for the first time a VN solution for a pulse-echo SoS image reconstruction problem using diverging waves with conventional transducers and single-sided tissue access. This is made possible by incorporating simulations with varying complexity into training. We use loop unrolling of gradient descent with momentum, with an exponentially weighted loss of outputs at each unrolled iteration in order to regularize the training. We learn norms as activation functions regularized to have smooth forms for robustness to input distribution variations. We evaluate reconstruction quality on the ray-based and full-wave simulations as well as on the tissue-mimicking phantom data, in comparison with a classical iterative [limited-memory Broyden–Fletcher–Goldfarb–Shanno (L-BFGS)] optimization of this image reconstruction problem. We show that the proposed regularization techniques combined with multisource domain training yield substantial improvements in the domain adaptation capabilities of VN, reducing the median root mean squared error (RMSE) by 54% on a wave-based simulation data set compared to the baseline VN. We also show that on data acquired from a tissue-mimicking breast phantom, the proposed VN provides improved reconstruction in 12 ms.
en_US
dc.language.iso
en
en_US
dc.publisher
IEEE
en_US
dc.subject
Image reconstruction
en_US
dc.subject
neural networks
en_US
dc.subject
ultrasonography
en_US
dc.title
Training Variational Networks with Multidomain Simulations: Speed-of-Sound Image Reconstruction
en_US
dc.type
Journal Article
dc.date.published
2020-07-20
ethz.journal.title
IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control
ethz.journal.volume
67
en_US
ethz.journal.issue
12
en_US
ethz.journal.abbreviated
IEEE trans. ultrason. ferroelectr. freq. control
ethz.pages.start
2584
en_US
ethz.pages.end
2594
en_US
ethz.identifier.wos
ethz.identifier.scopus
ethz.publication.place
New York, NY
en_US
ethz.publication.status
published
en_US
ethz.leitzahl
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02140 - Dep. Inf.technologie und Elektrotechnik / Dep. of Inform.Technol. Electrical Eng.::02652 - Institut für Bildverarbeitung / Computer Vision Laboratory::09528 - Göksel, Orçun (ehemalig) / Göksel, Orçun (former)
ethz.leitzahl.certified
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02140 - Dep. Inf.technologie und Elektrotechnik / Dep. of Inform.Technol. Electrical Eng.::02652 - Institut für Bildverarbeitung / Computer Vision Laboratory::09528 - Göksel, Orçun (ehemalig) / Göksel, Orçun (former)
ethz.date.deposited
2020-12-13T04:13:57Z
ethz.source
SCOPUS
ethz.eth
yes
en_US
ethz.availability
Metadata only
en_US
ethz.rosetta.installDate
2020-12-15T07:45:28Z
ethz.rosetta.lastUpdated
2022-03-29T04:35:13Z
ethz.rosetta.exportRequired
true
ethz.rosetta.versionExported
true
ethz.COinS
ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.atitle=Training%20Variational%20Networks%20with%20Multidomain%20Simulations:%20Speed-of-Sound%20Image%20Reconstruction&rft.jtitle=IEEE%20Transactions%20on%20Ultrasonics,%20Ferroelectrics,%20and%20Frequency%20Control&rft.date=2020-12&rft.volume=67&rft.issue=12&rft.spage=2584&rft.epage=2594&rft.issn=0885-3010&1525-8955&rft.au=Bernhardt,%20Melanie&Vishnevskiy,%20Valery&Rau,%20Richard&Goksel,%20Orcun&rft.genre=article&rft_id=info:doi/10.1109/TUFFC.2020.3010186&
 Search print copy at ETH Library

Files in this item

FilesSizeFormatOpen in viewer

There are no files associated with this item.

Publication type

Show simple item record