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

Chossegros M, Wagner S, Matek C, Stockholm D, Tannier X, Marr C (2026)

Publication Type: Journal article

Publication year: 2026

Journal

Expert Systems With Applications Elsevier

Book Volume: 327

Article Number: 132710

DOI: 10.1016/j.eswa.2026.132710

Abstract

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.

Authors with CRIS profile

Christian Matek

Involved external institutions

Sorbonne Université FR France (FR) Harvard University US United States (USA) (US) Helmholtz Zentrum München - Deutsches Forschungszentrum für Gesundheit und Umwelt (HMGU) / Helmholtz Munich DE Germany (DE) PSL Research University / Université de recherche Paris Sciences et Lettres FR France (FR)

How to cite

APA:

Chossegros, M., Wagner, S., Matek, C., Stockholm, D., Tannier, X., & Marr, C. (2026). Hierarchical supervision in DINOv2 training improves generalizability on white blood cell images. Expert Systems With Applications, 327. https://doi.org/10.1016/j.eswa.2026.132710

MLA:

Chossegros, Manon, et al. "Hierarchical supervision in DINOv2 training improves generalizability on white blood cell images." Expert Systems With Applications 327 (2026).

BibTeX: Download