Better screening tools for all cancer would only be a good thing, but one type in particular where it could have a huge impact is pancreatic cancer. The symptoms of this disease don't often reveal themselves until it is well progressed, and it carries a five-year survival rate of just nine percent. Scientists have now developed what could prove an exciting new diagnostics tool, in the form of a smartphone app that scans the white part of the eye for one of the disease's early tell-tale signs.

Because it is so difficult to detect, sufferers of pancreatic cancer are often diagnosed well after it has already spread. This means surgical removal of the tumor, the only potentially curative treatment, is not possible. One of the early symptoms of pancreatic cancer is jaundice, characterized by yellowing of the skin and eyes as a result of a substance called bilirubin in the blood.

But the trouble with bilirubin buildup, other than the fact it can be indicative of a number of diseases, is that it can only be picked up by a blood test that physicians won't administer unless there is already cause for concern. We have seen exciting progress towards a urine test for pancreatic cancer recently, but this new technology under development at the University of Washington (UW) would require no bodily fluids at all.

Dubbed BiliScreen, it uses a smartphone camera along with computer vision algorithms and machine learning to detect elevated bilirubin levels in the white part of the eye, known as the sclera. In adults, the sclera is more sensitive than the skin to increased bilirubin levels, and it the UW team's hope that BiliScreen can detect those changes before they become visible to the naked eye, when it is already too late.

"The eyes are a really interesting gateway into the body — tears can tell you how much glucose you have, sclera can tell you how much bilirubin is in your blood," said senior author of the study Shwetak Patel. "Our question was: Could we capture some of these changes that might lead to earlier detection with a selfie?"

It put the technology to the test in a clinical trial involving 70 people. The BiliScreen smartphone app was used in combination with either a special box made to control light exposure or paper glasses to calibrate color. The team found it was able to correctly detect causes for concern 89.7 percent of the time compared to the conventional blood test.

It does this by using the camera and flash to capture an image of the eye, and then computer vision software to isolate the sclera. The app software then assesses the wavelengths of light that are reflected and absorbed by the sclera, calculating color information and then uses machine learning algorithms to correlate that with bilirubin levels.

"This relatively small initial study shows the technology has promise," says co-author Dr. Jim Taylor, a professor in the UW Medicine Department of Pediatrics. "Pancreatic cancer is a terrible disease with no effective screening right now. 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."

From here, the team is looking to build on its exciting early results by testing the app on larger groups of people at risk of jaundice and related conditions. They will also look to make the technology more convenient to use, by possibly removing the need for the box and glasses.

The video below provides an overview of the technology, while the team is preparing to present its research next month at the Ubicomp 2017, the Association for Computing Machinery's International Joint Conference on Pervasive and Ubiquitous Computing.

View gallery - 3 images