Dive Brief:
-
An artificial intelligence (AI) algorithm performed at least 91% as well as human physicians when tested against 21 dermatologists on their ability to identify malignant skin conditions and benign lesions, according to a new study published in Nature.
-
Researchers from Stanford University used more than 129,000 images and an image classification system developed by Google to develop the AI algorithm.
-
The technology can "potentially extend the reach of dermatologists outside of the clinic" and "provide low-cost universal access to vital diagnostic care," the researchers wrote.
Dive Insight:
While the technology has not reached the mainstream just yet, more and more evidence is shedding light on the potential for AI as an aid to doctors. However, fears of advanced robots in healthcare presenting the possibility of making doctors obsolete have emerged.
The researchers at Stanford University built the diagnostic AI system using a database of 129,450 images portraying more than 2,000 diseases. Compared with human dermatologists, who identified 95% of malignant melanomas and 76% of benign moles correctly in a test, the AI system identified 96% of malignant samples correctly and 90% of benign lesions.
This is not the first time researchers have produced results with AI. IBM, for instance, has delivered a taste of what the future has to offer with Watson, the supercomputer sifting through patient data to provide guidance and decision support that improves diagnoses and workflow. It should come as no surprise that the AI market is expected to grow to $6 billion in 2021 from $600 million in 2014.