Learning-based enhancement of limited-view optoacoustic tomography based on image- and time-domain data

Neda Davoudi, Berkan Lafci, Ali Özbek, Xosé Luís Deán-Ben, Daniel Razansky

Publikation: Beitrag in Buch/Bericht/KonferenzbandKonferenzbeitragBegutachtung

1 Zitat (Scopus)

Abstract

Optoacoustic images are often afflicted with distortions and artifacts corresponding to system limitations, including limited-view tomographic data. We developed a convolutional neural network (CNN) approach for optoacoustic image quality enhancement combining training on both time-resolved signals and tomographic reconstructions. Reference human finger data for training the CNN were recorded using a full-ring array system with optimal tomographic coverage. The reconstructions were further refined with a dedicated algorithm that minimizes acoustic reflection artifacts induced by acoustically mismatch structures, such as bones. The combined methodology is shown to outperform other CNN-based methods solely operating on image-domain data.

OriginalspracheEnglisch
TitelPhotons Plus Ultrasound
UntertitelImaging and Sensing 2022
Redakteure/-innenAlexander A. Oraevsky, Lihong V. Wang
Herausgeber (Verlag)SPIE
ISBN (elektronisch)9781510647916
DOIs
PublikationsstatusVeröffentlicht - 2022
Extern publiziertJa
VeranstaltungPhotons Plus Ultrasound: Imaging and Sensing 2022 - Virtual, Online
Dauer: 20 Feb. 202224 Feb. 2022

Publikationsreihe

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Band11960
ISSN (Print)1605-7422

Konferenz

KonferenzPhotons Plus Ultrasound: Imaging and Sensing 2022
OrtVirtual, Online
Zeitraum20/02/2224/02/22

Fingerprint

Untersuchen Sie die Forschungsthemen von „Learning-based enhancement of limited-view optoacoustic tomography based on image- and time-domain data“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren