Artificial intelligence that is as accurate as human specialists at identifying skin cancer has been developed by computer scientists and dermatologists.
The breakthrough was made by a team at Stanford University, who trained a deep-learning algorithm to diagnose skin cancer using a database of around 130,000 skin disease images.
“We realized it was feasible, not just to do something well, but as well as a human dermatologist,” said Sebastian Thrun, a professor at the Stanford Artificial Intelligence Laboratory.
“That’s when our thinking changed. That’s when we said: ‘Look, this is not just a class project for students, this is an opportunity to do something great for humanity.”
The research, published in the journal Nature, describes how the algorithm was able to match the performance of 21 board-certified dermatologists in diagnosing skin lesions—the most common and deadly form of skin cancer.
The team used a deep-learning algorithm developed by Google, which was already capable of identifying 1.28 million images from 1,000 object categories. They then created a dataset of skin cancer images for the algorithm to learn from.
“We made a very powerful machine learning algorithm that learns from data,” said Andre Esteva, co-lead author of the paper. “Instead of writing into computer code exactly what to look for, you let the algorithm figure it out.”
Similar technology for spotting skin cancer is already available as a smartphone app, so there is potential for the algorithm to be incorporated into consumer devices in the future.
By combining a smartphone’s sensors with the vast power of cloud computing, some app creators claim the technology could replace primary healthcare.
“I believe that GPs play a very vital role as the first point of contact for people when they become sick or worry about something,” Dick Uyttewaal, the CEO of SkinVision, told IBTimes UK in a 2015 interview.
“But there is an awful lot of work that they currently do where they don’t add value and that could be replaced by technology.”