Artificial Intelligence Integration in Breast Cancer Screening: The ANIHTA Study

The goal of this observational study is to evaluate how artificial intelligence (AI) can be integrated into an organized breast cancer screening program for women aged 50-74 years undergoing mammography screening. The main questions it aims to answer are: * Can an AI-based workflow that identifies selected mammograms for review by a third radiologist improve or maintain the cancer detection rate compared with the current AI-supported double-reading workflow? * How would this AI-based workflow affect recall rates, biopsy referrals, false-positive results, interval cancers, radiologists' workload, and the number of cases requiring third-reader review? Researchers will compare the currently used screening strategy, in which mammograms are independently read by two radiologists with AI support, with an alternative AI-based strategy reconstructed using a target trial emulation approach. In the alternative strategy, predefined AI score thresholds will be used to identify selected cases for an offline review by an independent third radiologist. Participants will not be asked to undergo additional examinations or changes to their clinical care. Researchers will use routinely collected mammography screening data, AI scores, radiologists' readings, screening outcomes, and follow-up information. The alternative AI-based screening workflow will be evaluated retrospectively and will not affect participants' screening results or clinical management.