Self-supervised deep learning for highly efficient spatial immunophenotyping.
Cell classification
Deep learning
Imaging mass cytometry
Multiplex imaging
Multiplex immunohistochemistry
Self-supervised learning
Journal
EBioMedicine
ISSN: 2352-3964
Titre abrégé: EBioMedicine
Pays: Netherlands
ID NLM: 101647039
Informations de publication
Date de publication:
Sep 2023
Sep 2023
Historique:
received:
14
02
2023
revised:
07
08
2023
accepted:
08
08
2023
medline:
18
9
2023
pubmed:
7
9
2023
entrez:
6
9
2023
Statut:
ppublish
Résumé
Efficient biomarker discovery and clinical translation depend on the fast and accurate analytical output from crucial technologies such as multiplex imaging. However, reliable cell classification often requires extensive annotations. Label-efficient strategies are urgently needed to reveal diverse cell distribution and spatial interactions in large-scale multiplex datasets. This study proposed Self-supervised Learning for Antigen Detection (SANDI) for accurate cell phenotyping while mitigating the annotation burden. The model first learns intrinsic pairwise similarities in unlabelled cell images, followed by a classification step to map learnt features to cell labels using a small set of annotated references. We acquired four multiplex immunohistochemistry datasets and one imaging mass cytometry dataset, comprising 2825 to 15,258 single-cell images to train and test the model. With 1% annotations (18-114 cells), SANDI achieved weighted F1-scores ranging from 0.82 to 0.98 across the five datasets, which was comparable to the fully supervised classifier trained on 1828-11,459 annotated cells (-0.002 to -0.053 of averaged weighted F1-score, Wilcoxon rank-sum test, P = 0.31). Leveraging the immune checkpoint markers stained in ovarian cancer slides, SANDI-based cell identification reveals spatial expulsion between PD1-expressing T helper cells and T regulatory cells, suggesting an interplay between PD1 expression and T regulatory cell-mediated immunosuppression. By striking a fine balance between minimal expert guidance and the power of deep learning to learn similarity within abundant data, SANDI presents new opportunities for efficient, large-scale learning for histology multiplex imaging data. This study was funded by the Royal Marsden/ICR National Institute of Health Research Biomedical Research Centre.
Sections du résumé
BACKGROUND
BACKGROUND
Efficient biomarker discovery and clinical translation depend on the fast and accurate analytical output from crucial technologies such as multiplex imaging. However, reliable cell classification often requires extensive annotations. Label-efficient strategies are urgently needed to reveal diverse cell distribution and spatial interactions in large-scale multiplex datasets.
METHODS
METHODS
This study proposed Self-supervised Learning for Antigen Detection (SANDI) for accurate cell phenotyping while mitigating the annotation burden. The model first learns intrinsic pairwise similarities in unlabelled cell images, followed by a classification step to map learnt features to cell labels using a small set of annotated references. We acquired four multiplex immunohistochemistry datasets and one imaging mass cytometry dataset, comprising 2825 to 15,258 single-cell images to train and test the model.
FINDINGS
RESULTS
With 1% annotations (18-114 cells), SANDI achieved weighted F1-scores ranging from 0.82 to 0.98 across the five datasets, which was comparable to the fully supervised classifier trained on 1828-11,459 annotated cells (-0.002 to -0.053 of averaged weighted F1-score, Wilcoxon rank-sum test, P = 0.31). Leveraging the immune checkpoint markers stained in ovarian cancer slides, SANDI-based cell identification reveals spatial expulsion between PD1-expressing T helper cells and T regulatory cells, suggesting an interplay between PD1 expression and T regulatory cell-mediated immunosuppression.
INTERPRETATION
CONCLUSIONS
By striking a fine balance between minimal expert guidance and the power of deep learning to learn similarity within abundant data, SANDI presents new opportunities for efficient, large-scale learning for histology multiplex imaging data.
FUNDING
BACKGROUND
This study was funded by the Royal Marsden/ICR National Institute of Health Research Biomedical Research Centre.
Identifiants
pubmed: 37672979
pii: S2352-3964(23)00335-3
doi: 10.1016/j.ebiom.2023.104769
pmc: PMC10493897
pii:
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
104769Informations de copyright
Copyright © 2023 The Authors. Published by Elsevier B.V. All rights reserved.
Déclaration de conflit d'intérêts
Declaration of interests The authors declare the following competing financial interests: a patent has been filed for the methodology reported in the paper (applicant, the Institute of Cancer Research; inventors, H.Z and Y.Y; application number, UK patent GB 2106397.9 and PCT/EP2022/061941). Other authors have no conflict of interest to disclose.