BREAST IMAGINGB reast cancer screening using mammography has been implemented in many countries, which has resulted in reduced breast cancer mortality rates (1,2). Despite all efforts, a stable rate of approximately 30% of breast cancers still manifest between screening rounds. Such interval cancers (ICs) often have a worse prognosis than do screeningdetected cancers (3).Nevertheless, prior studies have shown that approximately half of ICs could be retrospectively identified by visual inspection of the last screening images as obvious false-negative findings or with minimal signs (4). The sensitivity of mammography in breast cancer detection decreases with increasing density of the breast tissue due to masking by the increasing amount of fibroglandular tissue within the breast (5). Furthermore, women with dense breasts have a higher risk of developing breast cancer (6). Consequently, the incidence of IC also increases with higher breast density (BD) (7). Studies have shown that both automated and clinical Breast Imaging Reporting and Data System density similarly enable prediction of interval and screen-detected cancer risk (8), which is now integrated in several existing breast cancer prediction models among other classic risk factors, such as age, ethnicity, family history of breast cancer, and history of breast biopsy (9-11).Artificial intelligence (AI) models for breast cancer detection based on deep learning technology are proposed as a clinical tool that could potentially increase both quality and efficiency of breast cancer screening. Several reader studies have demonstrated improved performance of the radiologist when AI was used for support during evaluation of both mammography and tomosynthesis images (12-17). The suggested applications of AI include replacing one reader in a double reading screening program (18,19) or for performing study triage to distinguish low-from high-risk findings, aiming to exclude low-risk studies from double human reading (20-22). Although the clinical evidence Background: Inclusion of mammographic breast density (BD) in breast cancer risk models improves accuracy, but accuracy remains modest. Interval cancer (IC) risk prediction may be improved by combining assessments of BD and an artificial intelligence (AI) cancer detection system.Purpose: To evaluate the performance of a neural network (NN)-based model that combines the assessments of BD and an AI system in the prediction of risk of developing IC among women with negative screening mammography results. Materials and Methods:This retrospective nested case-control study performed with screening examinations included women who developed IC and women with normal follow-up findings (from January 2011 to January 2015). An AI cancer detection system analyzed all studies yielding a score of 1-10, representing increasing likelihood of malignancy. BD was automatically computed using publicly available software. An NN model was trained by combining the AI score and BD using 10-fold cross-validation. Bootstrap analysis was used to c...