SP
BravenNow
Enabling clinical use of foundation models in histopathology
| USA | technology | ✓ Verified - arxiv.org

Enabling clinical use of foundation models in histopathology

#Foundation Models #Histopathology #Computational Pathology #Deep Learning #Medical Imaging #AI Clinical Application #Model Robustness #Technical Variability

📌 Key Takeaways

  • Researchers developed a novel method to make foundation models in histopathology more clinically applicable
  • The approach introduces robustness losses during training to reduce sensitivity to technical variability
  • Validation used 27,042 whole slide images from 6,155 patients across eight foundation models
  • Method improves both robustness and accuracy without requiring retraining of foundation models
  • This breakthrough enables development of reliable AI tools for routine clinical pathology practice

📖 Full Retelling

An international research team led by Audun L. Henriksen and 33 co-authors developed a new method to enable clinical use of foundation models in histopathology, addressing issues with technical variability in medical imaging analysis, in a study submitted on February 25, 2026. The research addresses a significant challenge in computational pathology where foundation models, while promising for developing high-performing deep learning systems, currently capture both biologically relevant features and pre-analytic variations that bias predictions. By introducing novel robustness losses during training of downstream task-specific models, the researchers successfully reduced sensitivity to technical variability without compromising biological relevance. The team validated their approach using an unprecedented dataset of 27,042 whole slide images from 6,155 patients to train thousands of models derived from eight popular foundation models in computational pathology. Their results demonstrated substantial improvements in both robustness and prediction accuracy when focusing on biologically relevant features, marking a significant advancement toward making artificial intelligence tools more reliable for medical diagnostics.

🏷️ Themes

Artificial Intelligence in Medicine, Computational Pathology, Model Robustness

📚 Related People & Topics

Medical imaging

Medical imaging

Technique and process of creating visual representations of the interior of a body

Medical imaging is the technique and process of imaging the interior of a body for clinical analysis and medical intervention, as well as visual representation of the function of some organs or tissues (physiology). Medical imaging seeks to reveal internal structures hidden by the skin and bones, as...

View Profile → Wikipedia ↗
Deep learning

Deep learning

Branch of machine learning

In machine learning, deep learning focuses on utilizing multilayered neural networks to perform tasks such as classification, regression, and representation learning. The field takes inspiration from biological neuroscience and revolves around stacking artificial neurons into layers and "training" t...

View Profile → Wikipedia ↗
Histopathology

Histopathology

Microscopic examination of tissue in order to study and diagnose disease

Histopathology (compound of three Greek words: ἱστός histos 'tissue', πάθος pathos 'suffering', and -λογία -logia 'study of') is the microscopic examination of tissue in order to study the manifestations of disease. Specifically, in clinical medicine, histopathology refers to the examination of a bi...

View Profile → Wikipedia ↗

Entity Intersection Graph

Connections for Medical imaging:

🌐 Deep learning 1 shared
🌐 Explainable artificial intelligence 1 shared
View full profile
Original Source
--> Computer Science > Computer Vision and Pattern Recognition arXiv:2602.22347 [Submitted on 25 Feb 2026] Title: Enabling clinical use of foundation models in histopathology Authors: Audun L. Henriksen , Ole-Johan Skrede , Lisa van der Schee , Enric Domingo , Sepp De Raedt , Ilyá Kostolomov , Jennifer Hay , Karolina Cyll , Wanja Kildal , Joakim Kalsnes , Robert W. Williams , Manohar Pradhan , John Arne Nesheim , Hanne A. Askautrud , Maria X. Isaksen , Karmele Saez de Gordoa , Miriam Cuatrecasas , Joanne Edwards , TransSCOT group , Arild Nesbakken , Neil A. Shepherd , Ian Tomlinson , Daniel-Christoph Wagner , Rachel S. Kerr , Tarjei Sveinsgjerd Hveem , Knut Liestøl , Yoshiaki Nakamura , Marco Novelli , Masaaki Miyo , Sebastian Foersch , David N. Church , Miangela M. Lacle , David J. Kerr , Andreas Kleppe View a PDF of the paper titled Enabling clinical use of foundation models in histopathology, by Audun L. Henriksen and 33 other authors View PDF HTML Abstract: Foundation models in histopathology are expected to facilitate the development of high-performing and generalisable deep learning systems. However, current models capture not only biologically relevant features, but also pre-analytic and scanner-specific variation that bias the predictions of task-specific models trained from the foundation model features. Here we show that introducing novel robustness losses during training of downstream task-specific models reduces sensitivity to technical variability. A purpose-designed comprehensive experimentation setup with 27,042 WSIs from 6155 patients is used to train thousands of models from the features of eight popular foundation models for computational pathology. In addition to a substantial improvement in robustness, we observe that prediction accuracy improves by focusing on biologically relevant features. Our approach successfully mitigates robustness issues of foundation models for computational pathology without retraining the foundation models themselves, ...
Read full article at source

Source

arxiv.org

More from USA

News from Other Countries

🇬🇧 United Kingdom

🇺🇦 Ukraine