The use of artificial intelligence in Sweden’s national breast cancer screening program has revealed that AI-assisted mammography screening is more effective than standard mammography, according to the full results of a 2023 trial published in the Lancet journal. Results showed that AI-assisted breast cancer screening identified more women with clinically relevant cancers without increasing the false-positive rate. Researchers from Lund University and other institutions in Sweden, Norway, Denmark and the Netherlands also found that women who underwent AI-assisted screening were less likely to be diagnosed with more aggressive breast cancer over the next two years. In August 2023, interim results from Mammography Screening with Artificial Intelligence (MASAI) showed that the use of AI detects 20% more cancers compared to standard screening. The research team also found that radiologists’ screen reading workload decreased by 44%. The full results show that AI-assisted mammography reduces cancer diagnoses by 12 percent in the years after a breast cancer screening appointment, an important test of the effectiveness of screening programs, the researchers said. “Our study is the first randomized controlled trial to investigate the use of AI in breast cancer screening, and the largest to date to examine the use of AI in cancer screening in general,” said lead author Dr. Kristina Lang, a breast radiologist and clinical researcher at Lund University in Sweden. He added: “AI-assisted screening has improved early detection of clinically relevant breast cancer, leading to fewer aggressive and advanced cancers diagnosed during screening.” From April 2021 to December 2022, more than 1,05,900 women were randomly assigned to either AI-assisted mammography screening or standard double reading by a radiologist without using AI. The AI system was trained, validated, and tested with over 200,000 inspections from multiple institutions in more than 10 countries. During the 2-year follow-up period, 1.55 interval cancers per 1,000 women (82 of 53,043) were detected in the AI-assisted mammography group, compared with 1.76 interval cancers per 1,000 women (93 of 52,872) in the standard duplication group. This was a 12% reduction in interval cancer diagnoses in the AI group. Interval cancers are malignant tumors that are detected between scheduled screening tests, after a previous negative result, and before the next routine test. Additionally, 81 percent (338 of 420) of cancer cases in the AI-assisted mammography group were detected by screening, compared to 74 percent (262 of 355) in the standard reading group, an increase of 9 percent. The false-positive rate was similar in both groups: 1.5% in the AI-assisted mammography interpretation group and 1.4% in the standard interpretation group. “Wide deployment of AI-assisted mammography in breast cancer screening programs will not only help reduce workload pressure on radiologists, but could also help detect more cancers, including more aggressive subtypes, at an earlier stage,” Lang said.
(This story has not been edited by Devdiscourse staff and is auto-generated from a syndicated feed.)
