Cargando…

On the acceptance of “fake” histopathology: A study on frozen sections optimized with deep learning

BACKGROUND: The fast acquisition process of frozen sections allows surgeons to wait for histological findings during the interventions to base intrasurgical decisions on the outcome of the histology. Compared with paraffin sections, however, the quality of frozen sections is often strongly reduced,...

Descripción completa

Detalles Bibliográficos
Autores principales: Siller, Mario, Stangassinger, Lea Maria, Kreutzer, Christina, Boor, Peter, Bulow, Roman D., Kraus, Theo J.F., von Stillfried, Saskia, Wolfl, Soraya, Couillard-Despres, Sebastien, Oostingh, Gertie Janneke, Hittmair, Anton, Gadermayr, Michael
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Elsevier 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8794030/
https://www.ncbi.nlm.nih.gov/pubmed/35136673
http://dx.doi.org/10.4103/jpi.jpi_53_21
_version_ 1784640737027031040
author Siller, Mario
Stangassinger, Lea Maria
Kreutzer, Christina
Boor, Peter
Bulow, Roman D.
Kraus, Theo J.F.
von Stillfried, Saskia
Wolfl, Soraya
Couillard-Despres, Sebastien
Oostingh, Gertie Janneke
Hittmair, Anton
Gadermayr, Michael
author_facet Siller, Mario
Stangassinger, Lea Maria
Kreutzer, Christina
Boor, Peter
Bulow, Roman D.
Kraus, Theo J.F.
von Stillfried, Saskia
Wolfl, Soraya
Couillard-Despres, Sebastien
Oostingh, Gertie Janneke
Hittmair, Anton
Gadermayr, Michael
author_sort Siller, Mario
collection PubMed
description BACKGROUND: The fast acquisition process of frozen sections allows surgeons to wait for histological findings during the interventions to base intrasurgical decisions on the outcome of the histology. Compared with paraffin sections, however, the quality of frozen sections is often strongly reduced, leading to a lower diagnostic accuracy. Deep neural networks are capable of modifying specific characteristics of digital histological images. Particularly, generative adversarial networks proved to be effective tools to learn about translation between two modalities, based on two unconnected data sets only. The positive effects of such deep learning-based image optimization on computer-aided diagnosis have already been shown. However, since fully automated diagnosis is controversial, the application of enhanced images for visual clinical assessment is currently probably of even higher relevance. METHODS: Three different deep learning-based generative adversarial networks were investigated. The methods were used to translate frozen sections into virtual paraffin sections. Overall, 40 frozen sections were processed. For training, 40 further paraffin sections were available. We investigated how pathologists assess the quality of the different image translation approaches and whether experts are able to distinguish between virtual and real digital pathology. RESULTS: Pathologists' detection accuracy of virtual paraffin sections (from pairs consisting of a frozen and a paraffin section) was between 0.62 and 0.97. Overall, in 59% of images, the virtual section was assessed as more appropriate for a diagnosis. In 53% of images, the deep learning approach was preferred to conventional stain normalization (SN). CONCLUSION: Overall, expert assessment indicated slightly improved visual properties of converted images and a high similarity to real paraffin sections. The observed high variability showed clear differences in personal preferences.
format Online
Article
Text
id pubmed-8794030
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher Elsevier
record_format MEDLINE/PubMed
spelling pubmed-87940302022-02-07 On the acceptance of “fake” histopathology: A study on frozen sections optimized with deep learning Siller, Mario Stangassinger, Lea Maria Kreutzer, Christina Boor, Peter Bulow, Roman D. Kraus, Theo J.F. von Stillfried, Saskia Wolfl, Soraya Couillard-Despres, Sebastien Oostingh, Gertie Janneke Hittmair, Anton Gadermayr, Michael J Pathol Inform Technical Note BACKGROUND: The fast acquisition process of frozen sections allows surgeons to wait for histological findings during the interventions to base intrasurgical decisions on the outcome of the histology. Compared with paraffin sections, however, the quality of frozen sections is often strongly reduced, leading to a lower diagnostic accuracy. Deep neural networks are capable of modifying specific characteristics of digital histological images. Particularly, generative adversarial networks proved to be effective tools to learn about translation between two modalities, based on two unconnected data sets only. The positive effects of such deep learning-based image optimization on computer-aided diagnosis have already been shown. However, since fully automated diagnosis is controversial, the application of enhanced images for visual clinical assessment is currently probably of even higher relevance. METHODS: Three different deep learning-based generative adversarial networks were investigated. The methods were used to translate frozen sections into virtual paraffin sections. Overall, 40 frozen sections were processed. For training, 40 further paraffin sections were available. We investigated how pathologists assess the quality of the different image translation approaches and whether experts are able to distinguish between virtual and real digital pathology. RESULTS: Pathologists' detection accuracy of virtual paraffin sections (from pairs consisting of a frozen and a paraffin section) was between 0.62 and 0.97. Overall, in 59% of images, the virtual section was assessed as more appropriate for a diagnosis. In 53% of images, the deep learning approach was preferred to conventional stain normalization (SN). CONCLUSION: Overall, expert assessment indicated slightly improved visual properties of converted images and a high similarity to real paraffin sections. The observed high variability showed clear differences in personal preferences. Elsevier 2022-12-20 /pmc/articles/PMC8794030/ /pubmed/35136673 http://dx.doi.org/10.4103/jpi.jpi_53_21 Text en © 2022 Published by Elsevier Inc. on behalf of Association for Pathology Informatics. https://creativecommons.org/licenses/by-nc-nd/4.0/This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
spellingShingle Technical Note
Siller, Mario
Stangassinger, Lea Maria
Kreutzer, Christina
Boor, Peter
Bulow, Roman D.
Kraus, Theo J.F.
von Stillfried, Saskia
Wolfl, Soraya
Couillard-Despres, Sebastien
Oostingh, Gertie Janneke
Hittmair, Anton
Gadermayr, Michael
On the acceptance of “fake” histopathology: A study on frozen sections optimized with deep learning
title On the acceptance of “fake” histopathology: A study on frozen sections optimized with deep learning
title_full On the acceptance of “fake” histopathology: A study on frozen sections optimized with deep learning
title_fullStr On the acceptance of “fake” histopathology: A study on frozen sections optimized with deep learning
title_full_unstemmed On the acceptance of “fake” histopathology: A study on frozen sections optimized with deep learning
title_short On the acceptance of “fake” histopathology: A study on frozen sections optimized with deep learning
title_sort on the acceptance of “fake” histopathology: a study on frozen sections optimized with deep learning
topic Technical Note
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8794030/
https://www.ncbi.nlm.nih.gov/pubmed/35136673
http://dx.doi.org/10.4103/jpi.jpi_53_21
work_keys_str_mv AT sillermario ontheacceptanceoffakehistopathologyastudyonfrozensectionsoptimizedwithdeeplearning
AT stangassingerleamaria ontheacceptanceoffakehistopathologyastudyonfrozensectionsoptimizedwithdeeplearning
AT kreutzerchristina ontheacceptanceoffakehistopathologyastudyonfrozensectionsoptimizedwithdeeplearning
AT boorpeter ontheacceptanceoffakehistopathologyastudyonfrozensectionsoptimizedwithdeeplearning
AT bulowromand ontheacceptanceoffakehistopathologyastudyonfrozensectionsoptimizedwithdeeplearning
AT kraustheojf ontheacceptanceoffakehistopathologyastudyonfrozensectionsoptimizedwithdeeplearning
AT vonstillfriedsaskia ontheacceptanceoffakehistopathologyastudyonfrozensectionsoptimizedwithdeeplearning
AT wolflsoraya ontheacceptanceoffakehistopathologyastudyonfrozensectionsoptimizedwithdeeplearning
AT couillarddespressebastien ontheacceptanceoffakehistopathologyastudyonfrozensectionsoptimizedwithdeeplearning
AT oostinghgertiejanneke ontheacceptanceoffakehistopathologyastudyonfrozensectionsoptimizedwithdeeplearning
AT hittmairanton ontheacceptanceoffakehistopathologyastudyonfrozensectionsoptimizedwithdeeplearning
AT gadermayrmichael ontheacceptanceoffakehistopathologyastudyonfrozensectionsoptimizedwithdeeplearning