Smartphones are coming more and more in handy—they can reveal low red blood cell counts, sleep apnea, and even HIV. Now, skin cancer can be added to the list.
Back in 2011, an iPhone app offered risk assessments on suspect moles. Today, a new research project at Stanford University is promising to bring things up to a professional grade of diagnosis, through a deep learning algorithm that can detect potential cancers with the same accuracy as dermatologists in early tests.
Early detection of skin cancer is critical to survival rates. The sooner you catch it, the better off you are, specifically for melanoma. If detected in its early stages, the five-year survival rate is 97 percent. When detected in its later stages, the survival rate is just 14 percent. Unfortunately, not everybody has access or money to drop by the doctor’s office and get their skin oddities checked out as soon as they appear.
Looking to improve the odds in these cases, computer scientists from Stanford University set out to build an artificially intelligent algorithm that uses deep learning to detect early-stage skin cancers.
Deep learning is when, instead of developing a system preloaded with answers, the system has the ability to work out the problem on its own. In this instance, the researchers built a database of almost 130,000 images of skin lesions representing more than 2,000 different diseases. They then took a Google-developed algorithm designed to distinguish cats from dogs, and adapted it to their skin cancer problem by feeding it each image as raw pixels and an accompanying disease label.
In total, the team then had 21 trained dermatologists diagnose cancerous and non-cancerous lesions from over 370 images. To test their work, the researchers tasked the AI with identifying the most common skin cancers, and then separately identifying the deadliest of skin cancers: malignant melanomas. In both tasks, the algorithm’s performance was on par with the experts, proving that the AI could classify skin cancers on a comparable level to trained dermatologists.
Getting this kind of technology into as many hands as possible involves only one avenue: the smartphone with its ever-improving camera and various sensors. While the algorithm was created and exists on a computer today, the team believes it could be easily adapted for the smartphone.
There are still months of testing to be done before potential launch, but soon it may be possible to pull out your phone to test any suspicious skin markings, and get answers immediately.
