A new study published in Nature Medicine has discovered an exciting new development in breast cancer screening. According to the research findings, using artificial intelligence tools during screenings increases the chance of detection, leading to diagnoses that may have been previously missed by clinicians.
The study authors examined data from more than 463,000 women in Germany between the ages of 50 and 69 who were screened for breast cancer between July 2021 and February 2023. Each woman’s scan was reviewed by two independent clinicians, and 260,739 of these scans were reviewed by a provider who utilized an AI tool during their analysis.
In total, 2,881 women participants were diagnosed with breast cancer. After controlling for various factors, the study authors found the rate of breast cancer detection was 17.6 percent higher for the group of women whose scans were reviewed using AI. Furthermore, the AI tool led to 204 additional diagnoses that were initially dismissed as normal by the reviewing clinician. In contrast, the tool missed 20 diagnoses that were identified by the human clinician.
The authors conclude that their “findings substantially add to the growing body of evidence suggesting that AI-supported mammography screening is feasible and safe and can reduce workload.” They believe “urgent efforts should be made to integrate AI-supported mammography into screening guidelines and to promote the widespread adoption of AI in mammography screening programs.”