PuSH - Publication Server of Helmholtz Zentrum München

Chossegros, M.* ; Wagner, S. ; Matek, C. ; Stockholm, D.* ; Tannier, X.* ; Marr, C.

Hierarchical supervision in DINOv2 training improves generalizability on white blood cell images.

Expert Syst. Appl. 327:132710 (2026)
Publ. Version/Full Text DOI
Open Access Hybrid
Creative Commons Lizenzvertrag
The microscopic observation of blood cells is a crucial step in diagnosing pathologies such as leukemia. DINOv2 models have been employed to extract features from blood cell images, but they do not include biological knowledge, nor do they allow multi-granular labels. To enhance the representation of these cells, we propose leveraging a biologically informed hierarchy of white blood cell types. We train a DINOv2-based foundation model with a semi-supervised framework that uses hierarchical supervision. It enables using datasets with varying levels of label precision within a structure that represents the process of cell differentiation. To support multi-level label precision, we modify the original hierarchical loss function, allowing any hierarchy level to serve as a ground truth class. We evaluate our model on three external datasets, including an out-of-domain set of cervical cells. Our approach improves generalization of the model to new datasets, improving by 1 percentage point the balanced accuracy on the two blood cell external datasets, and by 2.5 percentage point the balanced accuracy on the out-of-domain dataset. In addition the proposed strategy better aligns the model’s latent space with biological properties, leading to more acceptable misclassifications.
Altmetric
Additional Metrics?
[➜Log in]
Edit extra informations Login
Publication type Article: Journal article
Document type Scientific Article
Keywords Dinov2 ; Foundation Model ; Hierarchical Supervision ; Latent Space ; Semi-supervised Learning ; White Blood Cells
ISSN (print) / ISBN 0957-4174
e-ISSN 0957-4174
Journal Expert Systems with Applications
Quellenangaben Volume: 327, Issue: , Pages: , Article Number: 132710 Supplement: ,
Publisher Elsevier
Publishing Place The Boulevard, Langford Lane, Kidlington, Oxford Ox5 1gb, England
Reviewing status Peer reviewed
Institute(s) Institute of AI for Health (AIH)