Quantifying the effects of data augmentation and stain color normalization in convolutional neural networks for computational pathology

↓ Direkt till sidans innehåll
↓ Direkt till sidans sekundära innehåll (sidomenyn)

Search: onr:"swepub:oai:DiVA.org:liu-162492" > Quantifying the eff...

1 of 1
Previous record
Next record
To hitlist

Details
MARC

Tellez, DavidRadboud Univ Nijmegen, Netherlands (author)

Quantifying the effects of data augmentation and stain color normalization in convolutional neural networks for computational pathology

Article/chapterEnglish2019

Publisher, publication year, extent ...

ELSEVIER,2019
printrdacarrier

Numbers

LIBRIS-ID:oai:DiVA.org:liu-162492
https://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-162492URI
https://doi.org/10.1016/j.media.2019.101544DOI

Supplementary language notes

Language:English
Summary in:English

Part of subdatabase

SwePubSwePub

Classification

Subject category:ref swepub-contenttype
Subject category:art swepub-publicationtype

Notes

Funding Agencies|Radboud Institute of Health Sciences (RIHS), Nijmegen, The NetherlandsNetherlands Government; Dutch Cancer SocietyKWF Kankerbestrijding [KUN 2015-7970]; Alpe dHuZes fund [KUN 2014-7032]; European Unions Horizon 2020 research and innovation programme [825292]
Stain variation is a phenomenon observed when distinct pathology laboratories stain tissue slides that exhibit similar but not identical color appearance. Due to this color shift between laboratories, convolutional neural networks (CNNs) trained with images from one lab often underperform on unseen images from the other lab. Several techniques have been proposed to reduce the generalization error, mainly grouped into two categories: stain color augmentation and stain color normalization. The former simulates a wide variety of realistic stain variations during training, producing stain-invariant CNNs. The latter aims to match training and test color distributions in order to reduce stain variation. For the first time, we compared some of these techniques and quantified their effect on CNN classification performance using a heterogeneous dataset of hematoxylin and eosin histopathology images from 4 organs and 9 pathology laboratories. Additionally, we propose a novel unsupervised method to perform stain color normalization using a neural network. Based on our experimental results, we provide practical guidelines on how to use stain color augmentation and stain color normalization in future computational pathology applications. (C) 2019 Elsevier B.V. All rights reserved.

Subject headings and genre

TEKNIK OCH TEKNOLOGIER Medicinteknik Medicinsk bildbehandling hsv//swe
ENGINEERING AND TECHNOLOGY Medical Engineering Medical Image Processing hsv//eng
Deep learning; Convolutional neural network; Computational pathology

Added entries (persons, corporate bodies, meetings, titles ...)

Litjens, GeertRadboud Univ Nijmegen, Netherlands (author)
Bandi, PeterRadboud Univ Nijmegen, Netherlands (author)
Bulten, WouterRadboud Univ Nijmegen, Netherlands (author)
Bokhorst, John-MelleRadboud Univ Nijmegen, Netherlands (author)
Ciompi, FrancescoRadboud Univ Nijmegen, Netherlands (author)
van der Laak, JeroenLinköpings universitet,Avdelningen för radiologiska vetenskaper,Medicinska fakulteten,Centrum för medicinsk bildvetenskap och visualisering, CMIV,Region Östergötland, Klinisk patologi,Radboud Univ Nijmegen, Netherlands(Swepub:liu)jerva26 (author)
Radboud Univ Nijmegen, NetherlandsAvdelningen för radiologiska vetenskaper (creator_code:org_t)

Related titles

In:Medical Image Analysis: ELSEVIER581361-84151361-8423

Internet link

Find in a library

Medical Image Analysis (Search for host publication in LIBRIS)

To the university's database

1 of 1
Previous record
Next record
To hitlist

Find more in SwePub

By the author/editor: Tellez, David; Litjens, Geert; Bandi, Peter; Bulten, Wouter; Bokhorst, John-M ...; Ciompi, Francesc ...; show more...; van der Laak, Je ...; show less...

About the subject

ENGINEERING AND TECHNOLOGY: ENGINEERING AND ...; and Medical Engineer ...; and Medical Image Pr ...

Articles in the publication: Medical Image An ...

By the university: Linköping University

Search outside SwePub

Extend your search to:: Google; Google Book Search; Google Scholar

Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.

LIBRIS.kb.se