A new, ready-to-use AI system published in Nature Cancer identifies eighteen different types of cancers with high accuracy from a reduced number of histological slides.
A research team has developed a "plug-and-play" artificial intelligence system capable of recognizing eighteen distinct types of cancer from a limited number of tissue slides, without the need for specific retraining for each new type of cancer. The findings, published in the journal Nature Cancer, are part of a series of recent advances in AI-assisted medical imaging.
Other studies published between 2024 and 2026 report systems detecting cancerous lesions on average 40% earlier than radiologists working alone, a fundamental trend that is pushing health authorities to accelerate the approval of AI-assisted diagnostic algorithms in radiology, dermatology and ophthalmology.
Read the full article on the original source
Go to the source →