This selfie could save your life.

Researchers at the University of Washington have developed a smartphone app that could diagnose pancreatic cancer by snapping a user’s self portrait.

The BiliScreen works by analyzing the amount of yellow discoloration in the whites of the eyes, which is a sign of bilirubin buildup in the blood, a.k.a. jaundice, before it can be seen by the naked eye.

Jaundice, often seen as a discoloration of the skin in infants, is a tell-tale symptom of pancreatic cancer – which has a five-year survival rate of just 9% because it often goes undetected until it reaches an advanced stage – as well as a sign of diseases like hepatitis and liver damage from alcoholism. And while different skin colors can make jaundice difficult to spot, changes to the whites of the eye (or the sclera) can be seen consistently across all races and ethnicities.

So doctors, computer scientists and electrical engineers, building on previous work out of the UW’s Ubiquitous Computing Lab, created the BiliScreen. The user puts on paper glasses printed with color squares to help calibrate color, and a box accessory over the that, which blocks out ambient lighting. Then he snaps a selfie of his eyes with his smartphone camera, and the app runs the color of his sclera into a color descriptor, which then translates the how white his whites are into an estimate of his bilirubin levels.

The new diagnostic tool correctly identified “cases of concern” in just under 90% of the 70 subjects in its initial clinical study, compared to the blood test currently used by doctors to measure bilirubin levels.

“This relatively small initial study shows the technology has promise,” wrote co-author Dr. Jim Taylor, a professor in the UW Medicine Department of Pediatrics, whose father died of pancreatic cancer.

“Pancreatic cancer is a terrible disease with no effective screening right now,” he added. “Our goal is to have more people who are unfortunate enough to get pancreatic cancer to be fortunate enough to catch it in time to have surgery that gives them a better chance of survival.”

The research team needs to conduct more testing on a wider range of people at risk, and it is also working toward making the BiliScreen easier to handle by nixing the need for the box and the glasses. But researchers hope their device will be an accessible, non-invasive diagnostic tool for people without health insurance or with limited access to medical care. It could also be applied to monitor patients with pancreatic cancer who need to monitor their bilirubin levels; now they can just snap a selfie instead of having blood drawn.

BiliScreen is a new smartphone app that can screen for pancreatic cancer by having users snap a selfie. Dennis Wise/University of Washington

“The problem with pancreatic cancer is that by the time you’re symptomatic, it’s frequently too late,” wrote lead author Alex Mariakakis, a doctoral student at the Paul G. Allen School of Computer Science & Engineering. “The hope is that if people can do this simple test once a month — in the privacy of their own homes — some might catch the disease early enough to undergo treatment that could save their lives.”

More and more mobile health (mHealth) apps are turning mobile phones into handy medical clinics. The mHealth market is expected to hit $26 billion this year, and there are more than 97,000 apps related to health and fitness.

BiliScreen is a new smartphone app that can screen for pancreatic cancer by having users snap a selfie. (Dennis Wise/University of Washington)

Doctors from Harvard Medical School launched the Buoy Health application earlier this year, which asks about 20 questions, the way a doctor or nurse would when you walk into a clinic, before suggesting a potential diagnosis (out of 1,600) and the steps to take next. It won’t say that you have cancer or heart disease, however – it will instead suggest seeing a doctor to get a biopsy or imaging scan.

And Google reportedly acquired Senosis Health recently, a startup that also plans to monitor health stats or detect diseases by using phone features – for example, tapping your microphone to listen to your breathing and determine if you have a lung condition like asthma.