AI trained to diagnose cancer of unknown primary with 83% accuracy

Written by Louis Gautier, Future Science Group

cancer of unknown primary

A recent publication in Nature has described the development of an artificial intelligence (AI) model that uses routinely collected histology slides to accurately diagnose the origin of metastasis in cancer of unknown primary. In around 2% of cancer cases, the primary site of tumor growth cannot be determined and a diagnosis of cancer is only possible after identifying a secondary site of metastasis. Standard diagnostic workups may never find the primary site, possibly because the original cancer is small or has been destroyed by the immune system already. Prognosis for this rare malignant disease is very low – around 3–16...

To view this content, please register now for access

It's completely free