Skip to main content

Deep learning algorithm matches dermatologists at recognizing skin cancer

Worried about a strange mole on your back? Why not let an algorithm look at it!

That’s the broad idea behind a recent project created by computer scientists at Stanford University, which applied the extraordinary machine vision powers of cutting-edge deep learning neural networks to the world of dermatology.

Using a database of close to 130,000 images of skin diseases, the team was able to create an artificial intelligence algorithm able diagnose skin lesions with a performance level matching trained experts.

“[We trained it to] classify images of skin conditions as benign or malignant, and found that it matches the performance of over 21 board-certified dermatologists at three key diagnostic tasks: identifying keratinocyte carcinomas (the most common human cancer), identifying melanoma (the deadliest skin cancer), and identifying melanoma when viewed using dermoscopy,” co-first author Andre Esteva told Digital Trends.

The neural network the researchers used was one originally designed by Google and trained to recognize 1.28 million images, with the somewhat frivolous purpose of distinguishing cats from dogs.

“We saw that it was demonstrating superhuman performance at distinguishing between 200 different types of dog,” co-first author Brett Kuprel told us. “We thought we could apply this to something more useful, such as skin cancer diagnosis.”

Prior to the project, neither Esteva or Kuprel had any background in dermatology, which means the algorithm they created was able to achieve expert-level performance without benefitting from any specially encoded domain-specific knowledge.

However, if the algorithm was to be used by trained physicians, they could take advantage of a so-called “salience map,” revealing how important each pixel in an image was in the AI’s prediction process. In other words, rather than replacing dermatologists, this could prove to be a useful tool in their arsenal — the equivalent of a smart X-ray that offers its own interpretation about what it sees.

For now, though, that’s jumping way ahead. “There are definitely regulatory rules to get the FDA to approve it,” Kuprel said. “That would be important before any application could be deployed.” Beyond this, though, the investigators aren’t saying what is next.

“We are still deliberating on next steps and cannot yet comment,” Esteva said.

Editors' Recommendations

COVID-19 lockdown in China may slow Mac Studio shipments
Apple's Mac Studio desktop.

Anyone ordering Apple’s recently unveiled Mac Studio desktop should carefully be eyeing the shipping date as a COVID-19-related lockdown in the Chinese city of Shenzhen threatens to leave customers waiting longer than originally expected to receive their new machines.

Announced on Sunday, March 13, the lockdown in one of the world's busiest tech hubs will force the city's suppliers and manufacturing facilities to close for at least a week. Apple supplier Foxconn has already announced the temporary closure of several Shenzhen manufacturing sites, one of which produces iPhones.

Read more
Circular confirms its $259 smart ring is coming to the U.S.
best wearables of ces 2022 circular ring

The Circular smart ring is finally going to be available for pre-order on Sunday, February 27, via the Circular website and will cost $259. The wearable tech will be available for presale in European countries (France, Germany, the U.K., and Italy,) the United States, Australia, Hong Kong, and Singapore. Pre-orders will go live at 1:30 p.m. ET on Sunday, February 27. Those who pre-order the smart ring should expect delivery between April and June 2022, according to a Circular press release.

Circular doesn't clarify what ring sizes will be available when presales go live, however, the company has said that seven sizes for both men and women will be available. Digital Trends has reached out for clarification on the available sizes, and will update this article when we hear back. The Circular smart ring also comes in four different colors that can be switched out with replaceable outer shells: Black, rose gold, silver, and gold.

Read more
FDA clears first-ever smartphone app for insulin delivery
Someone checking their insulin levels.

People with diabetes will have an easier time getting insulin into their bodies with just one tap on their phone. The Food and Drug Administration has cleared the first-ever smartphone app developed by Tandem Diabetes Care to program insulin dosage for its t:slim X2 insulin pump users.

Tandem Diabetes Care, a leading insulin delivery and diabetes technology company, based in California, made the historic announcement on Wednesday, saying that its app is the first app for both Android and iOS capable of delivering insulin from a smartphone. Insulin delivery previously had to be conducted from the pump alone.

Read more