Resumen

GONZALEZ-ULLOA, Beatriz E.; CORONA-GONZALEZ, Claudia B.  y  MOLINA-GUTIERREZ, Rosa L.. Comparison of the diagnostic performance of Quantra artificial intelligence software with an experienced radiologist in the mammographic breast density assessment in women with and without breast implants. J. Mex. Fed. Radiol. Imaging [online]. 2025, vol.4, n.3, pp.198-204.  Epub 25-Nov-2025. ISSN 2696-8444.  https://doi.org/10.24875/jmexfri.m25000108.

Mammographic sensitivity decreases when mammographic breast density (MBD) is assessed in women with dense breasts and/or breast implants. This study compared the diagnostic performance of Quantra artificial intelligence (AI) software with an experienced radio logist with 32 years of experience interpreting breast images as the gold standard in MBD assessment of dense categories (c+d) in women with and without breast implants. In this prospective cohort study, an experienced radiologist and AI Quantra (v 2.2.2) assessed 2D mammograms and tomosyntheses of women over 35 years of age with and without breast implants in dense categories (c+d) based on the Breast Imaging Reporting and Data System (BI-RADS) 5th Edition. Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were recorded. AI Quantra sensitivity was low (30.6%, 95% CI 18.2-45.4) in dense categories (c+d) in women with breast implants (n = 130). In contrast, sensitivity was high (95.2%, 95% CI 92.1-97.3) in women without breast implants (n = 548). Accuracy was 73.1% (95% CI 64.6-80.5%) and 81.0% (95% CI 77.5-84.2), respectively. The diagnostic performance of the current version of Quantra AI in assessing MBD in dense categories (c+d) was unacceptably low in women with breast implants.

Palabras llave : Mammographic breast density; Artificial intelligence; Experienced radiologist; BI-RADS; Quantra.

        · texto en Inglés     · Inglés ( pdf )