Related Experiment Video
Updated: Nov 6, 2025
05:33
Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System
Published on: July 11, 2025
487
Impact of artificial intelligence support on accuracy and reading time in breast tomosynthesis image interpretation:
Suzanne L van Winkel1, Alejandro Rodríguez-Ruiz2, Linda Appelman3
1Department of Medical Imaging, Radboud University Medical Center, PO Box 9101, 6500 HB Nijmegen, Geert Grooteplein 10, 6525 GA, Post 766, Nijmegen, The Netherlands. Suzanne.vanWinkel@radboudumc.nl.
European Radiology
|May 5, 2021
Summary
Artificial intelligence (AI) support enhances radiologists' accuracy in detecting breast cancer with digital breast tomosynthesis (DBT), while also reducing reading time. The AI system
Area of Science:
- Radiology and Medical Imaging
- Artificial Intelligence in Healthcare
- Breast Cancer Screening
Background:
- Digital breast tomosynthesis (DBT) improves mammography sensitivity for breast cancer detection.
- Increased image volume in DBT can lead to longer reading times and potential errors.
- AI is being explored to assist radiologists in interpreting complex medical imaging.
Purpose of the Study:
- To evaluate if an AI support system improves breast radiologists' accuracy in reading wide-angle DBT exams.
- To assess the impact of AI support on the reading time for DBT examinations.
- To compare the standalone performance of an AI system against average radiologist performance in detecting malignancies.
Main Methods:
- A multi-reader, multi-case study involving 240 DBT exams and 18 radiologists.
- Radiologists interpreted exams with and without AI assistance, providing cancer suspicion scores.
- Area under the ROC curve (AUC) and reading time were compared between reading conditions.
Main Results:
- AI support significantly increased the average AUC (0.863 vs 0.833, p=0.0025).
- Reading time per DBT exam decreased with AI support (41s to 36s, p<0.001).
- The standalone AI system's AUC was non-inferior to the average radiologist's performance (p=0.8115).
Conclusions:
- AI support improves cancer detection accuracy for radiologists reading DBT.
- AI assistance reduces the time required for interpreting DBT exams.
- AI systems show comparable performance to average radiologists, potentially increasing accessibility and cost-effectiveness of screening programs.
Keywords:
Artificial intelligence (AI)Breast cancerDigital breast tomosynthesis (DBT)MammographyMass screening
