Self-supervision enhances instance-based multiple instance learning methods in digital pathology: a benchmark study
As simple instance-based MIL methods are naturally more interpretable and explainable to clinicians, our results suggest that more effort should be put into well-adapted SSL methods for WSI rather than into complex embedding-based MIL methods.
Authors
- Julien Adam (ORCID: https://orcid.org/0000-0001-7556-9101)
- A. Buronfosse (ORCID: https://orcid.org/0000-0002-6593-8935)
- А. И. Маммадов (ORCID: https://orcid.org/0000-0003-1178-9257)
- Gilles Hayem (ORCID: https://orcid.org/0000-0001-9595-8446)
- Loïc Le Folgoc (ORCID: https://orcid.org/0000-0002-5156-6616)
- Pietro Gori (ORCID: https://orcid.org/0000-0003-3456-8529)
- Guillaume Hocquet
Institutions
- Télécom Paris (FR)
- Hôpital Paris Saint-Joseph (FR)
Publication Details
- Journal
- Journal of Medical Imaging
- Published
- 2025-06-03
- DOI
- https://doi.org/10.1117/1.jmi.12.6.061404
- Citations
- 4
- Primary Topic
- AI in cancer detection
- Type
- article
- Field-Weighted Citation Impact
- 9.92
Funders
- Agence Nationale de la Recherche