Related Experiment Video
Updated: Aug 5, 2026
06:03
Integrating Augmented Reality Tools in Breast Cancer Related Lymphedema Prognostication and Diagnosis
Published on: February 6, 2020
Concordance Between Radiologist and AI-based Volumetric Breast Density Assessments: Clinical and Economic
Jacob R Devine1, Alexander D M Withrow1, Sabina Choudhry2,1
1University of South Dakota Sanford School of Medicine.
Summary
AI and radiologist breast density assessments show high agreement, but discrepancies at critical thresholds can impact patient care and costs. Ongoing oversight and updates are vital for AI accuracy in mammography.
Area of Science:
- Radiology
- Artificial Intelligence
- Medical Imaging
Background:
- Mammography Quality Standards Act (MQSA) mandates breast density disclosure.
- FDA-approved AI offers objective breast density assessment for workflow and screening.
- This study evaluates AI-radiologist concordance in breast density classification.
Purpose of the Study:
- To assess agreement between AI (Volpara/Lunit) and radiologist breast density assessments.
- To explore clinical and economic implications of AI-radiologist concordance.
- To identify impacts on patient risk scores and healthcare costs.
Main Methods:
- Retrospective cohort study of 54,819 mammograms (Feb 2024 - June 2025).
- Radiologists reviewed and could override AI-generated density assessments.
- Comparison of AI vs. radiologist classifications and analysis of effects on risk scores and costs.
Main Results:
- High agreement (92-97%) for BI-RADS B/C, 85-90% for A/D.
- Discrepancies primarily in BI-RADS B/C, leading to potential misclassification.
- Overestimation caused unnecessary procedures/costs; underestimation risked delayed cancer detection.
Conclusions:
- Current AI shows concordance with radiologists but has variability at critical thresholds.
- Standardized protocols, AI updates, and radiologist oversight are essential.
- Minimizing misclassification improves efficiency, outcomes, and trust in AI-assisted imaging.
