Related Experiment Video
Updated: Jun 6, 2025
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
Added Value of a Previous Exam Comparison in the Final Breast Imaging and Reporting Data System (BI-RADS®) Assessment
Eduardo De Faria Castro Fleury1, Veronica J Ayres2
1Radiology, MD duFLE Diagnósticos, São Paulo, BRA.
Purpose:
This study aims to determine the impact of prior examination comparison in the final Breast Imaging and Reporting Data System (BI-RADS®) assessment in screening mammography. We compared the incidence of BI-RADS® categories 0, 1, 2, 3, 4, and 5 before and after comparing to the previous exam, with a focus on category 0. We also compared the radiologists' final classification according to their clinical practice experience.
Materials And Methods:
This is a prospective observational study conducted from August 2020 to February 2021. Two radiologists independently interpreted 3,896 consecutive mammograms. The final BI-RADS® category was given for the mammography blinded to the previous examinations and after comparing the previous examination. We also compared the radiologists' final classification according to their clinical practice experience.
Results:
A total of 3,656 patients referred for breast mammography were evaluated, and 3,531 screening mammograms were included. The final BI-RADS® classification for mammograms blinded to previous examinations was as follows: 389 (11.02%), 3,017 (85.44%), 35 (0.99%), 87 (2.46%), and 3 (0.08%) for categories 0, 1, 2, 3, 4, and 5, respectively. After comparing the previous examinations, the results were as follows: 103 (2.92%), 3,311 (93.77%), 39 (1.10%), 75 (2.12%), and 3 (0.08%) for categories 0, 1, 2, 3, 4, and 5, respectively. Reader 1 read 1,439 mammograms, while reader 2 read 2,093 mammograms. When comparing the final category of each reader independently, reader 1 classified the mammograms as 91 (7.10%), 1,181 (92.19%), two (0.16%), and seven (0.55%) for blinded mammograms and 0, 1,261 (98.44%), six (0.47%), and 14 (1.09%) for mammograms after prior examination comparison for categories 0, 1 and 2, 3 and 4, respectively. Reader 2 classified the mammograms as 194 (9.27%), 1,181 (87.52%), 27 (1.29%), and 40 (1.91%) for blinded mammograms and one (0.05%), 2,044 (97.71%), 26 (1.24%), and 21 (1.00%) for mammograms after previous exams comparison for categories 0, 1 and 2, 3, and 4, respectively.
Conclusion:
Comparing previous mammograms of participants in breast cancer screening programs may reduce the number of BI-RADS® category 0 final classifications by 73.8% and the number of positive findings in the final classification, especially for the less experienced readers.
