Cargando…

Computational tissue staining of non-linear multimodal imaging using supervised and unsupervised deep learning

Hematoxylin and Eosin (H&E) staining is the ’gold-standard’ method in histopathology. However, standard H&E staining of high-quality tissue sections requires long sample preparation times including sample embedding, which restricts its application for ’real-time’ disease diagnosis. Due to th...

Descripción completa

Detalles Bibliográficos
Autores principales: Pradhan, Pranita, Meyer, Tobias, Vieth, Michael, Stallmach, Andreas, Waldner, Maximilian, Schmitt, Michael, Popp, Juergen, Bocklitz, Thomas
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Optical Society of America 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8086483/
https://www.ncbi.nlm.nih.gov/pubmed/33996229
http://dx.doi.org/10.1364/BOE.415962
_version_ 1783686521502564352
author Pradhan, Pranita
Meyer, Tobias
Vieth, Michael
Stallmach, Andreas
Waldner, Maximilian
Schmitt, Michael
Popp, Juergen
Bocklitz, Thomas
author_facet Pradhan, Pranita
Meyer, Tobias
Vieth, Michael
Stallmach, Andreas
Waldner, Maximilian
Schmitt, Michael
Popp, Juergen
Bocklitz, Thomas
author_sort Pradhan, Pranita
collection PubMed
description Hematoxylin and Eosin (H&E) staining is the ’gold-standard’ method in histopathology. However, standard H&E staining of high-quality tissue sections requires long sample preparation times including sample embedding, which restricts its application for ’real-time’ disease diagnosis. Due to this reason, a label-free alternative technique like non-linear multimodal (NLM) imaging, which is the combination of three non-linear optical modalities including coherent anti-Stokes Raman scattering, two-photon excitation fluorescence and second-harmonic generation, is proposed in this work. To correlate the information of the NLM images with H&E images, this work proposes computational staining of NLM images using deep learning models in a supervised and an unsupervised approach. In the supervised and the unsupervised approach, conditional generative adversarial networks (CGANs) and cycle conditional generative adversarial networks (cycle CGANs) are used, respectively. Both CGAN and cycle CGAN models generate pseudo H&E images, which are quantitatively analyzed based on mean squared error, structure similarity index and color shading similarity index. The mean of the three metrics calculated for the computationally generated H&E images indicate significant performance. Thus, utilizing CGAN and cycle CGAN models for computational staining is beneficial for diagnostic applications without performing a laboratory-based staining procedure. To the author’s best knowledge, it is the first time that NLM images are computationally stained to H&E images using GANs in an unsupervised manner.
format Online
Article
Text
id pubmed-8086483
institution National Center for Biotechnology Information
language English
publishDate 2021
publisher Optical Society of America
record_format MEDLINE/PubMed
spelling pubmed-80864832021-05-13 Computational tissue staining of non-linear multimodal imaging using supervised and unsupervised deep learning Pradhan, Pranita Meyer, Tobias Vieth, Michael Stallmach, Andreas Waldner, Maximilian Schmitt, Michael Popp, Juergen Bocklitz, Thomas Biomed Opt Express Article Hematoxylin and Eosin (H&E) staining is the ’gold-standard’ method in histopathology. However, standard H&E staining of high-quality tissue sections requires long sample preparation times including sample embedding, which restricts its application for ’real-time’ disease diagnosis. Due to this reason, a label-free alternative technique like non-linear multimodal (NLM) imaging, which is the combination of three non-linear optical modalities including coherent anti-Stokes Raman scattering, two-photon excitation fluorescence and second-harmonic generation, is proposed in this work. To correlate the information of the NLM images with H&E images, this work proposes computational staining of NLM images using deep learning models in a supervised and an unsupervised approach. In the supervised and the unsupervised approach, conditional generative adversarial networks (CGANs) and cycle conditional generative adversarial networks (cycle CGANs) are used, respectively. Both CGAN and cycle CGAN models generate pseudo H&E images, which are quantitatively analyzed based on mean squared error, structure similarity index and color shading similarity index. The mean of the three metrics calculated for the computationally generated H&E images indicate significant performance. Thus, utilizing CGAN and cycle CGAN models for computational staining is beneficial for diagnostic applications without performing a laboratory-based staining procedure. To the author’s best knowledge, it is the first time that NLM images are computationally stained to H&E images using GANs in an unsupervised manner. Optical Society of America 2021-03-23 /pmc/articles/PMC8086483/ /pubmed/33996229 http://dx.doi.org/10.1364/BOE.415962 Text en Published by The Optical Society under the terms of the Creative Commons Attribution 4.0 License. Further distribution of this work must maintain attribution to the author(s) and the published article’s title, journal citation, and DOI. https://creativecommons.org/licenses/by/4.0/https://creativecommons.org/licenses/by/4.0/
spellingShingle Article
Pradhan, Pranita
Meyer, Tobias
Vieth, Michael
Stallmach, Andreas
Waldner, Maximilian
Schmitt, Michael
Popp, Juergen
Bocklitz, Thomas
Computational tissue staining of non-linear multimodal imaging using supervised and unsupervised deep learning
title Computational tissue staining of non-linear multimodal imaging using supervised and unsupervised deep learning
title_full Computational tissue staining of non-linear multimodal imaging using supervised and unsupervised deep learning
title_fullStr Computational tissue staining of non-linear multimodal imaging using supervised and unsupervised deep learning
title_full_unstemmed Computational tissue staining of non-linear multimodal imaging using supervised and unsupervised deep learning
title_short Computational tissue staining of non-linear multimodal imaging using supervised and unsupervised deep learning
title_sort computational tissue staining of non-linear multimodal imaging using supervised and unsupervised deep learning
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8086483/
https://www.ncbi.nlm.nih.gov/pubmed/33996229
http://dx.doi.org/10.1364/BOE.415962
work_keys_str_mv AT pradhanpranita computationaltissuestainingofnonlinearmultimodalimagingusingsupervisedandunsuperviseddeeplearning
AT meyertobias computationaltissuestainingofnonlinearmultimodalimagingusingsupervisedandunsuperviseddeeplearning
AT viethmichael computationaltissuestainingofnonlinearmultimodalimagingusingsupervisedandunsuperviseddeeplearning
AT stallmachandreas computationaltissuestainingofnonlinearmultimodalimagingusingsupervisedandunsuperviseddeeplearning
AT waldnermaximilian computationaltissuestainingofnonlinearmultimodalimagingusingsupervisedandunsuperviseddeeplearning
AT schmittmichael computationaltissuestainingofnonlinearmultimodalimagingusingsupervisedandunsuperviseddeeplearning
AT poppjuergen computationaltissuestainingofnonlinearmultimodalimagingusingsupervisedandunsuperviseddeeplearning
AT bocklitzthomas computationaltissuestainingofnonlinearmultimodalimagingusingsupervisedandunsuperviseddeeplearning