Cargando…

A self-supervised contrastive learning approach for whole slide image representation in digital pathology

Image analysis in digital pathology has proven to be one of the most challenging fields in medical imaging for AI-driven classification and search tasks. Due to their gigapixel dimensions, whole slide images (WSIs) are difficult to represent for computational pathology. Self-supervised learning (SSL...

Descripción completa

Detalles Bibliográficos
Autores principales: Fashi, Parsa Ashrafi, Hemati, Sobhan, Babaie, Morteza, Gonzalez, Ricardo, Tizhoosh, H.R.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Elsevier 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9808093/
https://www.ncbi.nlm.nih.gov/pubmed/36605114
http://dx.doi.org/10.1016/j.jpi.2022.100133
_version_ 1784862860636061696
author Fashi, Parsa Ashrafi
Hemati, Sobhan
Babaie, Morteza
Gonzalez, Ricardo
Tizhoosh, H.R.
author_facet Fashi, Parsa Ashrafi
Hemati, Sobhan
Babaie, Morteza
Gonzalez, Ricardo
Tizhoosh, H.R.
author_sort Fashi, Parsa Ashrafi
collection PubMed
description Image analysis in digital pathology has proven to be one of the most challenging fields in medical imaging for AI-driven classification and search tasks. Due to their gigapixel dimensions, whole slide images (WSIs) are difficult to represent for computational pathology. Self-supervised learning (SSL) has recently demonstrated excellent performance in learning effective representations on pretext objectives, which may improve the generalizations of downstream tasks. Previous self-supervised representation methods rely on patch selection and classification such that the effect of SSL on end-to-end WSI representation is not investigated. In contrast to existing augmentation-based SSL methods, this paper proposes a novel self-supervised learning scheme based on the available primary site information. We also design a fully supervised contrastive learning setup to increase the robustness of the representations for WSI classification and search for both pretext and downstream tasks. We trained and evaluated the model on more than 6000 WSIs from The Cancer Genome Atlas (TCGA) repository provided by the National Cancer Institute. The proposed architecture achieved excellent results on most primary sites and cancer subtypes. We also achieved the best result on validation on a lung cancer classification task.
format Online
Article
Text
id pubmed-9808093
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher Elsevier
record_format MEDLINE/PubMed
spelling pubmed-98080932023-01-04 A self-supervised contrastive learning approach for whole slide image representation in digital pathology Fashi, Parsa Ashrafi Hemati, Sobhan Babaie, Morteza Gonzalez, Ricardo Tizhoosh, H.R. J Pathol Inform Original Research Article Image analysis in digital pathology has proven to be one of the most challenging fields in medical imaging for AI-driven classification and search tasks. Due to their gigapixel dimensions, whole slide images (WSIs) are difficult to represent for computational pathology. Self-supervised learning (SSL) has recently demonstrated excellent performance in learning effective representations on pretext objectives, which may improve the generalizations of downstream tasks. Previous self-supervised representation methods rely on patch selection and classification such that the effect of SSL on end-to-end WSI representation is not investigated. In contrast to existing augmentation-based SSL methods, this paper proposes a novel self-supervised learning scheme based on the available primary site information. We also design a fully supervised contrastive learning setup to increase the robustness of the representations for WSI classification and search for both pretext and downstream tasks. We trained and evaluated the model on more than 6000 WSIs from The Cancer Genome Atlas (TCGA) repository provided by the National Cancer Institute. The proposed architecture achieved excellent results on most primary sites and cancer subtypes. We also achieved the best result on validation on a lung cancer classification task. Elsevier 2022-08-28 /pmc/articles/PMC9808093/ /pubmed/36605114 http://dx.doi.org/10.1016/j.jpi.2022.100133 Text en © 2022 The Author(s) 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 Original Research Article
Fashi, Parsa Ashrafi
Hemati, Sobhan
Babaie, Morteza
Gonzalez, Ricardo
Tizhoosh, H.R.
A self-supervised contrastive learning approach for whole slide image representation in digital pathology
title A self-supervised contrastive learning approach for whole slide image representation in digital pathology
title_full A self-supervised contrastive learning approach for whole slide image representation in digital pathology
title_fullStr A self-supervised contrastive learning approach for whole slide image representation in digital pathology
title_full_unstemmed A self-supervised contrastive learning approach for whole slide image representation in digital pathology
title_short A self-supervised contrastive learning approach for whole slide image representation in digital pathology
title_sort self-supervised contrastive learning approach for whole slide image representation in digital pathology
topic Original Research Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9808093/
https://www.ncbi.nlm.nih.gov/pubmed/36605114
http://dx.doi.org/10.1016/j.jpi.2022.100133
work_keys_str_mv AT fashiparsaashrafi aselfsupervisedcontrastivelearningapproachforwholeslideimagerepresentationindigitalpathology
AT hematisobhan aselfsupervisedcontrastivelearningapproachforwholeslideimagerepresentationindigitalpathology
AT babaiemorteza aselfsupervisedcontrastivelearningapproachforwholeslideimagerepresentationindigitalpathology
AT gonzalezricardo aselfsupervisedcontrastivelearningapproachforwholeslideimagerepresentationindigitalpathology
AT tizhooshhr aselfsupervisedcontrastivelearningapproachforwholeslideimagerepresentationindigitalpathology
AT fashiparsaashrafi selfsupervisedcontrastivelearningapproachforwholeslideimagerepresentationindigitalpathology
AT hematisobhan selfsupervisedcontrastivelearningapproachforwholeslideimagerepresentationindigitalpathology
AT babaiemorteza selfsupervisedcontrastivelearningapproachforwholeslideimagerepresentationindigitalpathology
AT gonzalezricardo selfsupervisedcontrastivelearningapproachforwholeslideimagerepresentationindigitalpathology
AT tizhooshhr selfsupervisedcontrastivelearningapproachforwholeslideimagerepresentationindigitalpathology