Related Experiment Video
Updated: Jul 29, 2026
13:44
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
Soft-copy mammographic readings with different computer-assisted detection cuing environments: preliminary findings
B Zheng1, M A Ganott, C A Britton
1Division of Imaging Research, Department of Radiology, University of Pittsburgh, 300 Halket St, Suite 4200, Pittsburgh, PA 15213, USA. bzheng@radserv.arad.upmc.edu
Radiology
|November 24, 2001
Summary
Computer-assisted detection (CAD) significantly improves radiologist performance in mammography when the system is accurate. Poorly performing CAD systems can decrease accuracy in detecting masses and microcalcifications.
Area of Science:
- Radiology
- Medical Imaging
- Computer-Aided Diagnosis
Background:
- Mammography is crucial for early breast cancer detection.
- Computer-assisted detection (CAD) systems aim to enhance radiologist performance.
- The impact of CAD system accuracy on diagnostic performance requires careful evaluation.
Purpose of the Study:
- To evaluate radiologist performance in detecting mammographic abnormalities using different computer-assisted detection (CAD) cuing environments.
- To assess how varying levels of CAD sensitivity and specificity influence the detection of masses and microcalcification clusters.
Main Methods:
- Seven radiologists interpreted 209 digitized mammograms with varying CAD outputs.
- Five display modes were used, including no CAD and CAD with different sensitivity/false-positive rates.
- Receiver operating characteristic (ROC) analysis was employed for performance assessment.
Main Results:
- High-sensitivity, low-false-positive CAD cuing significantly improved observer performance (P < .01).
- Decreased CAD accuracy led to reduced radiologist performance (P < .01).
- CAD specificity primarily impacted mass detection, while sensitivity affected microcalcification cluster detection.
Conclusions:
- Accurate CAD systems can substantially enhance diagnostic performance in mammography.
- Inaccurate CAD systems may negatively impact radiologist accuracy in both cued and non-cued areas.
- Optimizing CAD system performance is critical for maximizing its benefits in mammography.
