EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

BioFuse: an embedding fusion framework for biomedical foundation models

Mirza Nasir Hossain iD, David Harris-Birtill

PLOS One · 18.03.2026 · 21 (3) · e0320989

DOI: 10.1371/journal.pone.0320989 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 18.03.2026. Verfasst von Mirza Nasir Hossain und 1 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Science, University of St Andrews, St Andrews, Fife, United Kingdom. Hauptthemen laut MEDLINE: Software. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 85 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The biomedical field has witnessed a surge in pre-trained foundation models excelling in specific subdomains such as radiology and histopathology. While integrating these models promises a more comprehensive understanding of biomedical data, it poses challenges in model compatibility and feature fusion. We present BioFuse, a novel open-source framework designed to generate optimised biomedical embeddings. BioFuse utilises a pool of 9 state-of-the-art (SOTA) foundation models to create task-specific embeddings. It employs grid search to automatically identify the optimal combination of models, fusing their embeddings through vector concatenation. On the MedMNIST+ benchmark, using XGBoost as the downstream classifier, BioFuse outperforms several existing methods, achieving SOTA AUC in 5/12 datasets, while maintaining near-SOTA performance across most remaining datasets. Remarkably, our experiments reveal unexpected cross-modal capabilities, with histopathology and radiology models showing strong performance when applied to other imaging modalities. BioFuse features a high-level API for immediate deployment and an extensible architecture to incorporate future models and fusion techniques. We anticipate BioFuse will not only enhance the utility of foundation models in biomedicine but also open new avenues for uncovering cross-modal relationships in biomedical data.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

SoftwareHumansBoosting Machine Learning Algorithms

Autor:innen und Institutionen

  • School of Computer Science, University of St Andrews, St Andrews, Fife, United Kingdom.

ORCID-Profile: Mirza Nasir Hossain

Zitieren

APA

Hossain, M. N., & Harris-Birtill, D. (2026). BioFuse: an embedding fusion framework for biomedical foundation models. PLOS One, 21(3), e0320989. https://doi.org/10.1371/journal.pone.0320989

Vancouver

Hossain MN, Harris-Birtill D. BioFuse: an embedding fusion framework for biomedical foundation models. PLOS One. 2026;21(3):e0320989. doi:10.1371/journal.pone.0320989

BibTeX
@article{Hossain2026,
  author = {Hossain, Mirza Nasir and Harris-Birtill, David},
  title = {BioFuse: an embedding fusion framework for biomedical foundation models},
  journal = {PLOS One},
  year = {2026},
  volume = {21},
  number = {3},
  pages = {e0320989},
  doi = {10.1371/journal.pone.0320989}
}
Kostenlos starten →