[Understanding CAD (computer-aided diagnosis) in mammography]

Area of Science:

• Radiology and Medical Imaging
• Artificial Intelligence in Healthcare
• Breast Cancer Screening

Background:

• Efficient double reading of mammograms is crucial for reducing false negatives in breast screening programs but presents logistical challenges.
• Computed Aided Detection (CAD) systems are advancing, aiding in the identification of suspicious mammographic lesions like microcalcifications, masses, and architectural distortions.
• The improving characterization capabilities of CAD suggest its potential to augment or replace human double reading.

Purpose of the Study:

• To review commercially available Computed Aided Detection (CAD) systems for mammography.
• To explain the underlying principles of CAD technology in mammographic interpretation.
• To discuss the efficacy and results of CAD mammography, particularly its role in breast screening programs based on recent prospective studies.

Main Methods:

• Review of current commercial Computed Aided Detection (CAD) systems.
• Analysis of the principles and technological advancements in CAD for mammography.
• Synthesis of findings from recent prospective studies evaluating CAD's performance in breast screening.

Main Results:

• CAD systems demonstrate increasing effectiveness in detecting various suspicious mammographic lesions.
• The potential for CAD to either complement or substitute human double reading is supported by emerging evidence.
• Prospective studies are providing data on CAD's specific impact within the context of organized breast screening programs.

Conclusions:

• Computed Aided Detection (CAD) shows significant promise for enhancing the efficiency and accuracy of mammogram interpretation.
• CAD technology may offer a viable solution to the challenges of implementing widespread double reading in breast screening.
• Further evaluation of CAD's role, informed by prospective studies, is essential for its integration into standard breast screening protocols.