PuSH - Publication Server of Helmholtz Zentrum München: Evaluating normative representation learning in generative AI for robust anomaly detection in brain imaging.

Navigation

Home

Deutsch

Research

Advanced Search

Browse by ...

... Journal

... Publication Type

... Research Data

... Publication Year

Publication overview

Support & Contact

Contact persons

Help

Data protection

Bercea, C.-I. ; Wiestler, B.* ; Rueckert, D.* ; Schnabel, J.A.

Evaluating normative representation learning in generative AI for robust anomaly detection in brain imaging.

Nat. Commun. 16:1624 (2025)

Publ. Version/Full Text

Research data

DOI

PMC

	Open Access Gold

Abstract
Metrics
Extra information

Normative representation learning focuses on understanding the typical anatomical distributions from large datasets of medical scans from healthy individuals. Generative Artificial Intelligence (AI) leverages this attribute to synthesize images that accurately reflect these normative patterns. This capability enables the AI allowing them to effectively detect and correct anomalies in new, unseen pathological data without the need for expert labeling. Traditional anomaly detection methods often evaluate the anomaly detection performance, overlooking the crucial role of normative learning. In our analysis, we introduce novel metrics, specifically designed to evaluate this facet in AI models. We apply these metrics across various generative AI frameworks, including advanced diffusion models, and rigorously test them against complex and diverse brain pathologies. In addition, we conduct a large multi-reader study to compare these metrics to experts' evaluations. Our analysis demonstrates that models proficient in normative learning exhibit exceptional versatility, adeptly detecting a wide range of unseen medical conditions. Our code is available at https://github.com/compai-lab/2024-ncomms-bercea.git .

Altmetric

Additional Metrics?

[➜Log in]

Edit extra informations Login

Publication type Article: Journal article

Document type Scientific Article

Corresponding Author

ISSN (print) / ISBN 2041-1723

e-ISSN 2041-1723

Journal Nature Communications

Quellenangaben Volume: 16, Issue: 1, Article Number: 1624

Publisher Nature Publishing Group

Publishing Place London

Non-patent literature Publications

Reviewing status Peer reviewed

Institute(s) Institute for Machine Learning in Biomed Imaging (IML)

Grants Berdelle-Stiftung
Helmholtz Association under the joint research school 'Munich School for Data Science'
Free State of Bavaria
C.I.B. is funded via the EVUK program ("Next-generation Al for Integrated Diagnostics") of the Free State of Bavaria