Retrospective multi-reader evaluation of a deep learning-based cancer detection model in contrast-enhanced mammography
British Journal of Radiology · 19.10.2026
DOI: 10.1093/bjr/tqag252 ↗Kurz gefasst
Fachartikel in der Zeitschrift British Journal of Radiology, online erschienen am 19.10.2026. Verfasst von Amr Farouk Ibrahim Moustafa und 8 weiteren Autor:innen. Beteiligte Einrichtung: Baheya Foundation for Early Detection and Treatment of Breast Cancer, El Haram, Giza, Egypt. Die Arbeit nennt 30 Literaturangaben. Gefördert von: Baheya Foundation.
Abstract (Originalsprache)
Objectives
To evaluate the clinical impact of a deep learning (DL)-based computer-aided diagnostic tool applied to contrast-enhanced mammography on radiologists’ diagnostic accuracy and reading efficiency.
Methods
In this retrospective single-center study, 150 contrast-enhanced mammography examinations were interpreted by 5 radiologists. Each case was read and annotated twice in 2 sequential sessions: first without assistance and then with the computer-aided diagnostic tool, with a minimum 4-week washout period. Diagnostic performance was assessed at lesion and breast levels using free-response receiver operating characteristic analysis, area under the curve, and false-positive rate. Reading time for 1 radiologist was recorded as an exploratory measure.
Results
The computer-aided diagnostic tool was associated with improved lesion-level performance for most radiologists, with a mean increase in the area under the free-response curve of 0.030 ± 0.018. At 89% sensitivity, false-positive findings decreased by an average of 11.4% ± 24% across all radiologists. No improvement was observed at the breast level, possibly related to the tool’s single-view analysis and dataset characteristics. An exploratory analysis showed a 31% reduction in reading time for 1 radiologist.
Conclusions
Integration of a DL-based computer-aided diagnostic tool into contrast-enhanced mammography was associated with improved lesion-level diagnostic performance and reduced false-positive findings, without measurable improvement at the breast level. Advances in knowledge To our knowledge, this study provides one of the early clinical evaluations of a DL-based computer-aided diagnostic tool for contrast-enhanced mammography, suggesting potential improvements in lesion-level diagnostic performance and workflow efficiency while highlighting current limitations at the breast level.
Autor:innen und Institutionen
- Baheya Foundation for Early Detection and Treatment of Breast Cancer, El Haram, Giza, Egypt.
- Radiology Department, National Cancer Institute, Cairo University, Cairo, Egypt.
- Women's Health, GE HealthCare, Buc, 78530, France.
- Laboratoire de Mécanique Paris-Saclay, CentraleSupélec, ENS Paris-Saclay, Université Paris-Saclay, CNRS, Gif-sur-Yvette, 91190, France.
- Radiology Department, Faculty of Medicine, Cairo University, Cairo, Egypt.
Förderung
- Baheya Foundation
Zitieren
APA
Moustafa, A. F. I., Jailin, C., Vancamberg, L., Mohamed, S., Milioni De Carvalho, P., Iordache, R., Kamal, E. F., Gomaa, M. M. M., & Kamal, R. M. (2026). Retrospective multi-reader evaluation of a deep learning-based cancer detection model in contrast-enhanced mammography. British Journal of Radiology. https://doi.org/10.1093/bjr/tqag252
Vancouver
Moustafa AFI, Jailin C, Vancamberg L, Mohamed S, Milioni De Carvalho P, Iordache R, et al. Retrospective multi-reader evaluation of a deep learning-based cancer detection model in contrast-enhanced mammography. British Journal of Radiology. 2026. doi:10.1093/bjr/tqag252
BibTeX
@article{Moustafa2026,
author = {Moustafa, Amr Farouk Ibrahim and Jailin, Clement and Vancamberg, Laurence and Mohamed, Sara and Milioni De Carvalho, Pablo and Iordache, Razvan and Kamal, Eman Faker and Gomaa, Mohammed Mohammed Mohammed and Kamal, Rasha Mohammed},
title = {Retrospective multi-reader evaluation of a deep learning-based cancer detection model in contrast-enhanced mammography},
journal = {British Journal of Radiology},
year = {2026},
doi = {10.1093/bjr/tqag252}
}
EUVIMED