Bridging Domain Gaps in Computational Pathology: A Comparative Study of Adaptation Strategies
Sensors · 2025
DOI: 10.3390/s25092856 ↗Kurz gefasst
Vergleichsstudie in der Zeitschrift Sensors, 2025. Verfasst von João D Nunes und 6 weiteren Autor:innen. Beteiligte Einrichtung: Institute for Systems and Computer Engineering, Technology and Science, 4200-465 Porto, Portugal. Hauptthemen laut MEDLINE: Image Processing, Computer-Assisted, Pathology, Computational Biology. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Fundação para a Ciência e Tecnologia, National Funds through the Portuguese funding agency, FCT—Fundação para a Ciência e a Tecnologia, IMP Diagnostics.
Abstract (Originalsprache)
Due to the high variability in Hematoxylin and Eosin (H&E)-stained Whole Slide Images (WSIs), hidden stratification, and batch effects, generalizing beyond the training distribution is one of the main challenges in Deep Learning (DL) for Computational Pathology (CPath). But although DL depends on large volumes of diverse and annotated data, it is common to have a significant number of annotated samples from one or multiple source distributions, and another partially annotated or unlabeled dataset representing a target distribution for which we want to generalize, the so-called Domain Adaptation (DA). In this work, we focus on the task of generalizing from a single source distribution to a target domain. As it is still not clear which domain adaptation strategy is best suited for CPath, we evaluate three different DA strategies, namely FixMatch, CycleGAN, and a self-supervised feature extractor, and show that DA is still a challenge in CPath.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Institute for Systems and Computer Engineering, Technology and Science, 4200-465 Porto, Portugal.
- Faculty of Engineering, University of Porto, 4200-465 Porto, Portugal.
- IMP Diagnostics, 4150-146 Porto, Portugal.
- Cancer Biology and Epigenetics Group, Research Center of Portuguese Oncology Institute of Porto/RISE@Research Center of Portuguese Oncology Institute of Porto (Health Research Network), Portuguese Oncology Institute of…
- FCTUC-Faculty of Sciences and Technology, University of Coimbra, 3004-516 Coimbra, Portugal.
- Institute of Tissue Medicine and Pathology, University of Bern, 3008 Bern, Switzerland.
ORCID-Profile: João D Nunes, Diana Montezuma, Domingos Oliveira, Tania Pereira, Inti Zlobec, Jaime S Cardoso
Förderung
- Fundação para a Ciência e Tecnologia
- National Funds through the Portuguese funding agency, FCT—Fundação para a Ciência e a Tecnologia
- IMP Diagnostics
Zitieren
APA
Nunes, J. D., Montezuma, D., Oliveira, D., Pereira, T., Zlobec, I., Pinto, I. M., & Cardoso, J. S. (2025). Bridging Domain Gaps in Computational Pathology: A Comparative Study of Adaptation Strategies. Sensors. https://doi.org/10.3390/s25092856
Vancouver
Nunes JD, Montezuma D, Oliveira D, Pereira T, Zlobec I, Pinto IM, et al. Bridging Domain Gaps in Computational Pathology: A Comparative Study of Adaptation Strategies. Sensors. 2025. doi:10.3390/s25092856
BibTeX
@article{Nunes2025,
author = {Nunes, João D and Montezuma, Diana and Oliveira, Domingos and Pereira, Tania and Zlobec, Inti and Pinto, Isabel Macedo and Cardoso, Jaime S},
title = {Bridging Domain Gaps in Computational Pathology: A Comparative Study of Adaptation Strategies},
journal = {Sensors},
year = {2025},
doi = {10.3390/s25092856}
}
EUVIMED