Image restoration via learning on a digital twin for multi-core fiber endoscopy

Tijue Wang; Jakob Dremel; Sven Richter; Witold Polanski; Ortrud Uckermann; Ilker Eyüpoglu; Jürgen W. Czarske; Robert Kuschmierz

Image restoration via learning on a digital twin for multi-core fiber endoscopy

Research output: Contribution to conferences › Paper › Contributed › peer-review

Contributors

Tijue Wang - , Chair of Measurement and Sensor System Technique, Else Kröner Fresenius Center for Digital Health (Author)
Jakob Dremel - , Chair of Measurement and Sensor System Technique, Else Kröner Fresenius Center for Digital Health (Author)
Sven Richter - , Department of Neurosurgery, Else Kröner Fresenius Center for Digital Health, University Hospital Carl Gustav Carus Dresden (Author)
Witold Polanski - , Department of Neurosurgery, Else Kröner Fresenius Center for Digital Health, University Hospital Carl Gustav Carus Dresden (Author)
Ortrud Uckermann - , Department of Neurosurgery, Department of Child and Adolescent Psychiatry and Psychotherapy, Else Kröner Fresenius Center for Digital Health, University Hospital Carl Gustav Carus Dresden (Author)
Ilker Eyüpoglu - , Department of Neurosurgery, Else Kröner Fresenius Center for Digital Health, University Hospital Carl Gustav Carus Dresden (Author)
Jürgen W. Czarske - , Chair of Measurement and Sensor System Technique, Clusters of Excellence PoL: Physics of Life, Faculty of Physics, Else Kröner Fresenius Center for Digital Health (Author)
Robert Kuschmierz - , Chair of Measurement and Sensor System Technique, Else Kröner Fresenius Center for Digital Health (Author)

Abstract

Fiber endoscopy has great potential in all-optical biopsies but suffers fiber-specific distortions. We propose a method based on deep learning and digital twins for image reconstruction with a few training samples to enable clinical transfer.

Details

Original language	English
Publication status	Published - 2024
Peer-reviewed	Yes

Conference

Title	2024 Imaging Systems and Applications, ISA 2024 - Part of Optica Imaging Congress
Duration	15 - 19 July 2024
City	Toulouse
Country	France

External IDs

ORCID	/0000-0002-6603-5375/work/169643481

Research Portal of the TU Dresden

Image restoration via learning on a digital twin for multi-core fiber endoscopy

Contributors

Abstract

Details

Conference

External IDs

Keywords

ASJC Scopus subject areas