Algorithm outperforms humans in identifying cervical precancer

January 11, 2019

An automated artificial intelligence algorithm has been shown to accurately analyze digital images of a woman’s cervix and identify cervical precancer.

The AI approach, called automated visual evaluation, was developed by researchers at the National Cancer Institute and Global Good—a fund at Intellectual Ventures—whose results were published on Thursday in the Journal of the National Cancer Institute.

More than 60,000 cervical images from an NCI archive of photos—collected during a study conducted in Costa Rica from 1993 to 2000—were digitized and then used to train the AI algorithm to differentiate cervical conditions requiring medical treatment from those that did not.

Read More on Health Data Management