Related Experiment Video
Updated: Jul 4, 2025
13:44
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
42.9K
Diagnostic performance with and without artificial intelligence assistance in real-world screening mammography
Si Eun Lee1, Hanpyo Hong1, Eun-Kyung Kim1
1Department of Radiology, Yongin Severance Hospital, Yonsei University College of Medicine, Yongin, Korea.
European Journal of Radiology Open
|January 31, 2024
Summary
Artificial intelligence-based computer-aided diagnosis (AI-CAD) did not significantly change radiologist performance in screening mammography. However, AI-CAD assistance improved radiologist specificity and accuracy while lowering recall rates compared to standalone AI-CAD.
Area of Science:
- Radiology
- Medical Imaging
- Artificial Intelligence in Healthcare
Background:
- Screening mammography is crucial for early breast cancer detection.
- Artificial intelligence-based computer-aided diagnosis (AI-CAD) tools are emerging to assist radiologists.
- Evaluating the real-world impact of AI-CAD on radiologist performance is essential.
Purpose of the Study:
- To assess the impact of AI-CAD on radiologist diagnostic performance during screening mammography.
- To compare the performance of radiologists with and without AI-CAD assistance.
- To compare radiologist performance with standalone AI-CAD capabilities.
Main Methods:
- Retrospective analysis of screening mammography and ultrasound data from 1819 women (August 2020 - May 2022).
- Radiologists interpreted mammography with AI-CAD results provided or withheld alternatively each month.
- Diagnostic performance metrics including cancer detection rate, recall rate, sensitivity, specificity, accuracy, and AUC were analyzed.
Main Results:
- Radiologist diagnostic performance showed no significant difference with or without AI-CAD assistance.
- Radiologists with AI-CAD assistance had similar sensitivity (76.5%) and specificity (92.3%) compared to standalone AI-CAD (93.8%).
- Without AI-CAD, radiologists exhibited lower specificity (91.9%) and accuracy (91.5%) and higher recall rates (8.6%) versus standalone AI-CAD.
Conclusions:
- AI-CAD assistance did not significantly alter radiologist diagnostic performance in screening mammography when combined with ultrasound.
- Radiologists without AI-CAD assistance demonstrated reduced specificity and accuracy and increased recall rates compared to standalone AI-CAD.
- The study highlights the potential of AI-CAD to support radiologists, particularly in improving specificity and reducing unnecessary recalls.
More Related Videos
Related Concept Videos
01:24Magnetic Resonance Imaging
5.2K
Magnetic resonance imaging (MRI) is a noninvasive medical imaging technique based on a phenomenon of nuclear physics discovered in the 1930s, in which matter exposed to magnetic fields and radio waves was found to emit radio signals. In 1970, a physician and researcher named Raymond Damadian noticed that malignant (cancerous) tissue gave off different signals than normal body tissue. He applied for a patent for the first MRI scanning device in clinical use by the early 1980s. The early MRI...
5.2K
01:15Receiver Operating Characteristic Plot
179
A ROC (Receiver Operating Characteristic) plot is a graphical tool used to assess the performance of a binary classification model by illustrating the trade-off between sensitivity (true positive rate) and specificity (false positive rate). By plotting sensitivity against 1 - specificity across various threshold settings, the ROC curve shows how well the model distinguishes between classes, with a curve closer to the top-left corner indicating a more accurate model. The area under the ROC curve...
179