Assessing Standalone AI Performance for Breast Imaging for DBT, Screening Mammography

Background: There is considerable interest in the potential use of artificial intelligence (AI) systems in mammography including digital mammography and tomosynthesis for screening. It is critical to evaluate the performance of AI before it can become a modality used for independent mammographic interpretation. Objective: To determine the standalone performances of AI programs reported in the literature for interpretation of digital mammography and digital breast tomosynthesis (DBT). Methods: A medical database search was conducted (PubMed, Google Scholar, EMBASE [Ovid], and Web of Science) for studies published during a 5-year period (2017 to 2022). Sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) values were reviewed. Study quality was also assessed. Results: In total, 16 studies that include >1 million examinations in just Conclusions: Standalone AI for breast imaging outperformed radiologists in reader studies for screening digital mammog more...