Interpretation Time Using a Concurrent-Read Computer-Aided Detection System for Automated Breast Ultrasound in Breast...

Area of Science:

• Radiology
• Medical Imaging
• Breast Cancer Screening

Background:

• Dense breast tissue can obscure abnormalities on mammograms.
• Automated breast ultrasound (ABUS) is an emerging tool for screening women with dense breasts.
• Computer-aided detection (CAD) systems aim to improve diagnostic performance in medical imaging.

Purpose of the Study:

• To compare the diagnostic accuracy and interpretation time of screening automated breast ultrasound (ABUS) with and without a computer-aided detection (CAD) system.
• To evaluate the effectiveness of a U.S. Food and Drug Administration-approved CAD system for concurrent read in ABUS interpretation.

Main Methods:

• A retrospective observer performance study involving 18 radiologists interpreting 185 screening ABUS studies (cancer-enriched set).
• Each radiologist interpreted studies twice: once without CAD and once with CAD, separated by 4 weeks.
• Diagnostic accuracy was assessed using ROC analysis (Dorfman-Berbaum-Metz method), and interpretation times were compared using ANOVA.

Main Results:

• The area under the receiver operating characteristic curve (AUC) was noninferior with CAD (0.848) compared to without CAD (0.828).
• Mean interpretation time was significantly reduced with CAD, saving an average of 1 minute and 9 seconds per case (67% of the time without CAD).

Conclusions:

• The concurrent-read CAD system for screening ABUS in women with dense breasts offers noninferior diagnostic accuracy.
• The CAD system significantly decreases interpretation time, making it a more efficient screening tool.