Using AI, radiologists can see signs of breast cancer up to ten years earlier in the mammograms of patients later diagnosed with the disease, according to new research published this week in Radiology. These findings could help greatly accelerate diagnosis and improve decisions about who to screen sooner and with more regularity—questions that are increasingly important as more younger women are diagnosed with the disease.
The researchers tested three commercially available AI-based computer-assisted detection (AI-CAD) systems on mammogram data from a large screening population. Cancer prediction scores issued by AI-CAD were elevated for a significant number of patients later diagnosed with breast cancer compared to those who remained cancer-free. The team was led by researchers at the Karolinska University Hospital in Stockholm.
“Approximately 20% of breast cancer cases demonstrate mammographic signs that are already visible to AI around six years before diagnosis,” said senior coauthor Fredrik Strand, MD, PhD, of Karolinska. “Our study confirms the potential of AI to, in some cases, find signs of cancer in the mammograms much earlier than when radiologists detected it.”
Earlier diagnosis is crucial in breast cancer, as it is for most malignancies. Although annual breast screening, using mammography, is recommended for women in the U.S. aged 40 to 74, it is known that approximately 20-25% of women show early signs of breast cancer on their scans before they are diagnosed.
Strand’s team investigated AI’s potential to flag mammographic signs that were present up to 10 years in retrospect. The study included a total of 88,963 mammograms performed on 31,394 patients over 10 years—between January 2008 and April 2019. Of these participants, 12,072, or 38.5%, were diagnosed with cancer.
The data came from the Validation of Artificial Intelligence for Breast Imaging (VAI-B) database, which collects breast imaging data from volunteers across four regions of Sweden. The Swedish national breast screening program invites women between the ages of 40 and 74 to participate in screening examinations every two years, and each mammogram has traditionally been read by two radiologists. The average age of screening was 57.6 years.
The AI-CAD systems found many of those cancers at earlier screening points, achieving 90% specificity—distinguishing between a true positive and a true negative result—in up to 17% of patients at 10 years before diagnosis, up to 19.7% six years before diagnosis, up to 25.2% four years before diagnosis, and up to 39.3% two years before diagnosis.
“This study aims to add to the growing literature regarding the application of AI in breast cancer screening and how it can help play a role in earlier detection of breast cancer,” said Strand. “Analyzing the AI scores of screened individuals over time could provide insight into how early detectable changes arise, potentially allowing for earlier intervention.”
Breast cancer is the most commonly diagnosed cancer worldwide. It accounts for roughly 30% of all new female cancer diagnoses. In the U.S., AI-based systems have shown promise for predicting five-year risk of breast cancer and identifying women at risk of cancers that arise between regular screening mammograms.
The post AI Detects Breast Cancer Much Earlier than Radiologists Alone appeared first on Inside Precision Medicine.