Deep model-based optoacoustic image reconstruction (DeepMB)

Christoph Dehner, Vasilis Ntziachristos, Dominik Jüstel, Guillaume Zahnd

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Multispectral optoacoustic tomography requires real-time image feedback during clinical use. Herein, we present DeepMB, a deep learning framework to express the model-based reconstruction operator with a deep neural network and reconstruct high-quality optoacoustic images from arbitrary experimental input data at speeds that enable live imaging (31ms per image).

Original languageEnglish
Title of host publicationPhotons Plus Ultrasound
Subtitle of host publicationImaging and Sensing 2024
EditorsAlexander A. Oraevsky, Lihong V. Wang
PublisherSPIE
ISBN (Electronic)9781510669437
DOIs
StatePublished - 2024
EventPhotons Plus Ultrasound: Imaging and Sensing 2024 - San Francisco, United States
Duration: 28 Jan 202431 Jan 2024

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume12842
ISSN (Print)1605-7422

Conference

ConferencePhotons Plus Ultrasound: Imaging and Sensing 2024
Country/TerritoryUnited States
CitySan Francisco
Period28/01/2431/01/24

Keywords

  • Inverse problems
  • Model-based reconstruction
  • Multispectral Optoacoustic Tomography (MSOT)
  • Real-time imaging
  • Synthesized training data

Fingerprint

Dive into the research topics of 'Deep model-based optoacoustic image reconstruction (DeepMB)'. Together they form a unique fingerprint.

Cite this