Microscopy image reconstruction with physics-informed denoising diffusion probabilistic model

Abstract Light microscopy is a practical tool for advancing biomedical research and diagnostics, offering invaluable insights into the cellular and subcellular structures of living organisms. However, diffraction and optical imperfections actively hinder the attainment of high-quality images. In rec...

Full description

Saved in:
Bibliographic Details
Main Authors: Rui Li, Gabriel della Maggiora, Vardan Andriasyan, Anthony Petkidis, Artsemi Yushkevich, Nikita Deshpande, Mikhail Kudryashev, Artur Yakimovich
Format: Article
Language:English
Published: Nature Portfolio 2024-12-01
Series:Communications Engineering
Online Access:https://doi.org/10.1038/s44172-024-00331-z
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846101144899682304
author Rui Li
Gabriel della Maggiora
Vardan Andriasyan
Anthony Petkidis
Artsemi Yushkevich
Nikita Deshpande
Mikhail Kudryashev
Artur Yakimovich
author_facet Rui Li
Gabriel della Maggiora
Vardan Andriasyan
Anthony Petkidis
Artsemi Yushkevich
Nikita Deshpande
Mikhail Kudryashev
Artur Yakimovich
author_sort Rui Li
collection DOAJ
description Abstract Light microscopy is a practical tool for advancing biomedical research and diagnostics, offering invaluable insights into the cellular and subcellular structures of living organisms. However, diffraction and optical imperfections actively hinder the attainment of high-quality images. In recent years, there has been a growing interest in applying deep learning techniques to overcome these challenges in light microscopy imaging. Nonetheless, the resulting reconstructions often suffer from undesirable artefacts and hallucinations. Here, we introduce a deep learning-based approach that incorporates the fundamental physics of light propagation in microscopy into the loss function. This model employs a conditioned diffusion model in a physics-informed architecture. To mitigate the issue of limited available data, we utilise synthetic datasets for training purposes. Our results demonstrate consistent enhancements in image quality and substantial reductions in artefacts when compared to state-of-the-art methods. The presented technique is intuitively accessible and allows obtaining higher quality microscopy images for biomedical studies.
format Article
id doaj-art-c02ff148b03d4bedb5342e41dfcc3be5
institution Kabale University
issn 2731-3395
language English
publishDate 2024-12-01
publisher Nature Portfolio
record_format Article
series Communications Engineering
spelling doaj-art-c02ff148b03d4bedb5342e41dfcc3be52024-12-29T12:33:36ZengNature PortfolioCommunications Engineering2731-33952024-12-013111210.1038/s44172-024-00331-zMicroscopy image reconstruction with physics-informed denoising diffusion probabilistic modelRui Li0Gabriel della Maggiora1Vardan Andriasyan2Anthony Petkidis3Artsemi Yushkevich4Nikita Deshpande5Mikhail Kudryashev6Artur Yakimovich7Center for Advanced Systems Understanding (CASUS)Center for Advanced Systems Understanding (CASUS)Department of Molecular Life Sciences, University of ZurichDepartment of Molecular Life Sciences, University of ZurichIn situ Structural Biology, Max Delbrück Center for Molecular Medicine in the Helmholtz AssociationCenter for Advanced Systems Understanding (CASUS)In situ Structural Biology, Max Delbrück Center for Molecular Medicine in the Helmholtz AssociationCenter for Advanced Systems Understanding (CASUS)Abstract Light microscopy is a practical tool for advancing biomedical research and diagnostics, offering invaluable insights into the cellular and subcellular structures of living organisms. However, diffraction and optical imperfections actively hinder the attainment of high-quality images. In recent years, there has been a growing interest in applying deep learning techniques to overcome these challenges in light microscopy imaging. Nonetheless, the resulting reconstructions often suffer from undesirable artefacts and hallucinations. Here, we introduce a deep learning-based approach that incorporates the fundamental physics of light propagation in microscopy into the loss function. This model employs a conditioned diffusion model in a physics-informed architecture. To mitigate the issue of limited available data, we utilise synthetic datasets for training purposes. Our results demonstrate consistent enhancements in image quality and substantial reductions in artefacts when compared to state-of-the-art methods. The presented technique is intuitively accessible and allows obtaining higher quality microscopy images for biomedical studies.https://doi.org/10.1038/s44172-024-00331-z
spellingShingle Rui Li
Gabriel della Maggiora
Vardan Andriasyan
Anthony Petkidis
Artsemi Yushkevich
Nikita Deshpande
Mikhail Kudryashev
Artur Yakimovich
Microscopy image reconstruction with physics-informed denoising diffusion probabilistic model
Communications Engineering
title Microscopy image reconstruction with physics-informed denoising diffusion probabilistic model
title_full Microscopy image reconstruction with physics-informed denoising diffusion probabilistic model
title_fullStr Microscopy image reconstruction with physics-informed denoising diffusion probabilistic model
title_full_unstemmed Microscopy image reconstruction with physics-informed denoising diffusion probabilistic model
title_short Microscopy image reconstruction with physics-informed denoising diffusion probabilistic model
title_sort microscopy image reconstruction with physics informed denoising diffusion probabilistic model
url https://doi.org/10.1038/s44172-024-00331-z
work_keys_str_mv AT ruili microscopyimagereconstructionwithphysicsinformeddenoisingdiffusionprobabilisticmodel
AT gabrieldellamaggiora microscopyimagereconstructionwithphysicsinformeddenoisingdiffusionprobabilisticmodel
AT vardanandriasyan microscopyimagereconstructionwithphysicsinformeddenoisingdiffusionprobabilisticmodel
AT anthonypetkidis microscopyimagereconstructionwithphysicsinformeddenoisingdiffusionprobabilisticmodel
AT artsemiyushkevich microscopyimagereconstructionwithphysicsinformeddenoisingdiffusionprobabilisticmodel
AT nikitadeshpande microscopyimagereconstructionwithphysicsinformeddenoisingdiffusionprobabilisticmodel
AT mikhailkudryashev microscopyimagereconstructionwithphysicsinformeddenoisingdiffusionprobabilisticmodel
AT arturyakimovich microscopyimagereconstructionwithphysicsinformeddenoisingdiffusionprobabilisticmodel