Researchers from Rochester, Minn.-based Mayo Clinic have invented a new class of healthcare artificial intelligence they say could help advance cancer care.
Hypotheses-driven AI differs from traditional AI in that it incorporates up-to-date knowledge about a given disease, such as genetic variants or interactions between genes, the researchers say. Conventional AI learns only from large datasets that often lack new scientific information.
"This fosters a new era in designing targeted and informed AI algorithms to solve scientific questions, better understand diseases, and guide individualized medicine," said senior author and co-inventor Hu Li, PhD, a Mayo Clinic biology and AI researcher, in a March 11 news release. "It has the potential to uncover insights missed by conventional AI."
The study was published in February in Cancers.
Mayo Clinic is also testing several new generative AI applications.