Summary: Lunit’s AI mammography tool, Lunit Insight MMG, demonstrated the ability to predict breast cancer risk years in advance and reduce radiologist workload while maintaining detection accuracy, based on studies from Norway and Denmark.
Key Takeaways
- Lunit Insight MMG was able to predict breast cancer risk 4 to 6 years before detection, as shown in the Norwegian study.
- The Danish study found that using AI in double reading scenarios could reduce radiologist workload by nearly 50% without compromising detection accuracy.
- Both studies highlight the potential for AI to assist in improving breast cancer screening efficiency and accuracy.
—————————————————————————————————————————————————————————————
In response to increasing demand for radiologists and a global shortage, Lunit has announced findings from two large studies evaluating its artificial intelligence (AI) mammography technology, Lunit Insight MMG. The studies, conducted by the Cancer Registry of Norway and Odense University Hospital in Denmark, highlight the potential of the AI tool to improve breast cancer screening programs.
Lunit Insight MMG was shown to predict breast cancer risk and assist in streamlining radiologist workflows. The Norwegian study, published in JAMA Network Open and led by Professor Solveig Hofvind, analyzed data from 116,495 women aged 50-69. Results indicated that Lunit’s AI could estimate future breast cancer risk 4 to 6 years before detection, with higher AI scores correlating with cancer development across multiple screening rounds.
Reducing Radiologist Workload
In the Danish study, led by Mohammad T. Elhakim, MD, and published in Radiology: Artificial Intelligence, researchers analyzed 249,402 mammograms. The study found that replacing one or both radiologists with AI in double reading scenarios reduced workload by nearly 50%, while maintaining or improving detection accuracy.
“These studies provide compelling evidence of our AI’s potential to transform breast cancer screening,” says Brandon Suh, CEO of Lunit. “In Norway, our AI identified future breast cancer risk up to six years ahead. The Danish study showed a 50% reduction in radiologist workload without sacrificing detection quality.”