An article just published in the August 2023 issue of The Lancet Oncology can be expected to create considerable discussion. The article reports the preliminary findings of a large prospective study in Sweden that has essentially found equivalency of use of AI by a single radiologist in diagnosing screening mammography studies when compared to double readings of mammograms by two diagnostic radiologists, reducing the manpower required by almost half.

This is the first randomized trial ever conducted where the clinical safety of an AI-supported mammography screen-reading protocol used by a single radiologist was compared with interpretations of mammography studies performed by two radiologists. The results are significant.

Between April 2021 and July 2022, more than 80,000 women were randomly assigned to AI-supported interpretations performed by a single radiologist versus two radiologists' double reading without AI. In the study, the AI-supported screening model and the traditional double read radiologist approach each performed interpretations of studies for approximately 40,000 participants. According to the The Guardian newspaper, "[i]n total, 244 women (28%) recalled from AI-supported screening were found to have cancer compared with 203 women (25%) recalled from standard screening. This resulted in 41 more cancers being detected with the support of AI, of which 19 were invasive and 22 were in situ cancers." The false positive rate was only 1.5% in both groups. Additionally, there were 36,886 fewer screen readings by radiologists in the AI group compared with the group receiving standard care, resulting in a 44% reduction in the screen-reading workload of radiologists.

According to the lead author, Dr. Kristina Lång from Lund University in Sweden, "[w]hile our AI-supported screening system requires at least one radiologist in charge of detection, it could potentially do away with the need for double reading of the majority of mammograms, easing the pressure on workloads and enabling radiologists to focus on more advanced diagnostics while shortening waiting times for patients."

This trial is ongoing. The authors report that the endpoint of interval cancer rate will be assessed in 100,000 enrolled participants after two years of follow up.

We will want to follow this study's ultimate findings.

AI-supported mammography screening resulted in a similar cancer detection rate compared with standard double reading, with a substantially lower screen-reading workload, indicating that the use of AI in mammography screening is safe.

This article is presented for informational purposes only and is not intended to constitute legal advice.