Google is using artificial intelligence to improve breast cancer identification. The company employed deep learning systems to look at thousands of slides of cancer cells from Radboud University. This system is meant to complement the work of humans, with the goal of increasing accurate diagnosis overall.
What we’ve trained is just a little sliver of software that helps with one part of a very complex series of tasks. There will hopefully be more and more of these tools that help doctors [who] have to go through an enormous amount of information all the time.
– Lily Peng, project manager, Google
Deep learning was found to match or exceed the performance of a pathologist. By one measure, the AI clearly outperformed a pathologist.
In fact, the prediction heatmaps produced by the algorithm had improved so much that the localization score (FROC) for the algorithm reached 89%, which significantly exceeded the score of 73% for a pathologist with no time constraint.