Health AI Central
Peer-Reviewed Publication
Nat Commun2025;16(1):1624.February 13, 2025Journal Article

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

Cosmin I Bercea1,2, Benedikt Wiestler3,4, Daniel Rueckert4,5,6, Julia A Schnabel7,8,4,9
1Chair of Computational Imaging and AI in Medicine, Technical University of Munich (TUM), Munich, Germany. cosmin.bercea@tum.de.
2Helmholtz AI and Helmholtz Center Munich, Munich, Germany. cosmin.bercea@tum.de.
3Chair of AI for Image-Guided Diagnosis and Therapy, TUM School of Medicine and Health, Munich, Germany.
4Munich Center for Machine Learning (MCML), Munich, Germany.
5Chair of AI in Healthcare and Medicine, Technical University of Munich (TUM) and TUM University Hospital, Munich, Germany.
6Department of Computing, Imperial College London, London, UK.
7Chair of Computational Imaging and AI in Medicine, Technical University of Munich (TUM), Munich, Germany.
8Helmholtz AI and Helmholtz Center Munich, Munich, Germany.
9School of Biomedical Engineering and Imaging Sciences, King's College London, London, UK.
PubMed PMC DOI PDF

Abstract

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 p…

Create a free account to keep reading

Free members get 10 full research views every month across publications, clinical trials, FDA clearances, adverse events, and NIH grants. No credit card required.

Create Free AccountSign In

Want unlimited research access? See Pro plans

Data Accuracy Notice: Research intelligence on Health AI Central is aggregated from public sources (PubMed, ClinicalTrials.gov, FDA, NIH, CMS, and others) and refreshed nightly. Classifications and derived metrics are produced by automated methods described in our Methodology. We recommend verifying critical data points against the primary sources before making decisions.