EUVIMEDEUVIMED entdecken
Open Access

Cancer Detection in Breast MRI Screening via Explainable AI Anomaly Detection

Felipe Oviedo iD, Anum S Kazerouni iD, Philipp Liznerski iD, Yixi Xu, Michael Hirano iD, Robert A Vandermeulen, Marius Kloft iD, Elyse Blum, Adam M Alessio iD, Christopher I Li iD, William B Weeks iD, Rahul Dodhia iD und 3 weitere

Radiology · 2025

DOI: 10.1148/radiol.241629 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Radiology, 2025. Verfasst von Felipe Oviedo und 14 weiteren Autor:innen. Beteiligte Einrichtung: AI for Good Lab, Microsoft, 1 Microsoft Way, 4330 150th Ave NE, Redmond, WA 98052. Hauptthemen laut MEDLINE: Breast Neoplasms, Magnetic Resonance Imaging, Artificial Intelligence, Image Interpretation, Computer-Assisted. Die Arbeit wurde bisher 19-mal zitiert (Crossref) und nennt 24 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NCI NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

An anomaly detection model for breast cancer on MRI scans achieved high performance, surpassing binary classification in high- and low-cancer-prevalence scenarios and producing explanation heat maps not significantly different from radiologist annotations.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Breast NeoplasmsMagnetic Resonance ImagingArtificial IntelligenceImage Interpretation, Computer-AssistedHumansFemaleRetrospective StudiesMiddle AgedBreastAdultAgedEarly Detection of CancerSensitivity and Specificity

Autor:innen und Institutionen

  • AI for Good Lab, Microsoft, 1 Microsoft Way, 4330 150th Ave NE, Redmond, WA 98052.
  • Department of Radiology, University of Washington School of Medicine, Seattle, Wash.
  • University of Kaiserslautern-Landau, Kaiserslautern, Germany.
  • Berlin Institute for the Foundations of Learning and Data, Berlin, Germany.
  • Machine Learning Group, Technical University of Berlin, Berlin, Germany.
  • Department of Biomedical Engineering, Michigan State University, East Lansing, Mich.

ORCID-Profile: Felipe Oviedo, Anum S Kazerouni, Philipp Liznerski, Michael Hirano, Marius Kloft, Adam M Alessio, Christopher I Li, William B Weeks, Rahul Dodhia, Habib Rahbar

Förderung

  • NCI NIH HHS

Zitieren

APA

Oviedo, F., Kazerouni, A. S., Liznerski, P., Xu, Y., Hirano, M., Vandermeulen, R. A., Kloft, M., Blum, E., Alessio, A. M., Li, C. I., Weeks, W. B., Dodhia, R., Lavista Ferres, J. M., Rahbar, H., & Partridge, S. C. (2025). Cancer Detection in Breast MRI Screening via Explainable AI Anomaly Detection. Radiology. https://doi.org/10.1148/radiol.241629

Vancouver

Oviedo F, Kazerouni AS, Liznerski P, Xu Y, Hirano M, Vandermeulen RA, et al. Cancer Detection in Breast MRI Screening via Explainable AI Anomaly Detection. Radiology. 2025. doi:10.1148/radiol.241629

BibTeX
@article{Oviedo2025,
  author = {Oviedo, Felipe and Kazerouni, Anum S and Liznerski, Philipp and Xu, Yixi and Hirano, Michael and Vandermeulen, Robert A and Kloft, Marius and Blum, Elyse and Alessio, Adam M and Li, Christopher I and Weeks, William B and Dodhia, Rahul and Lavista Ferres, Juan M and Rahbar, Habib and Partridge, Savannah C},
  title = {Cancer Detection in Breast MRI Screening via Explainable AI Anomaly Detection},
  journal = {Radiology},
  year = {2025},
  doi = {10.1148/radiol.241629}
}
Kostenlos starten →