Today, people who are suffering from pancreatic cancer typically have an extremely low survival rate and this is primarily because the disease goes undiagnosed both in its early and intermediate stages. The thing is, there are no apparent symptoms at the early stage of the disease and the tools for detecting early pancreatic tumors before they metastasize are yet to be developed — this is bad news for patients. It’s also important to note that over 80 percent of patients seek medical attention at a stage where the tumors cannot be removed completely.

Well, the good news is, we’re living in a world where great innovations are introduced from time to time with the aim of making life better for us. Speaking of innovations, a team of researchers at the University of Washington have created a novel smartphone app which has the capability to perform noninvasive screening for pancreatic cancer based on image analysis — it also does the same for other diseases. The app gives individuals the chance to identify signs of jaundice by taking a quick selfie.

For the most part, the app is known as BiliScreen and it works with a smartphone camera, machine learning tools and computer vision algorithms to identify yellowing of skin and eyes which come as a result of high bilubrin levels — this is typically undetectable at minimally elevated levels. The app is designed to detect early stage of jaundice in the whites of the eyes so patients can know next step of action. In other words, BiliScreen acts as an early screen for pancreatic cancer, hepatitis, Osteogenesis imperfecta and other diseases.

Alex Mariakakis, a doctoral student at the Paul G. Allen School of Computer Science & Engineering says, “The problem with pancreatic cancer is that by the time you’re symptomatic, it’s frequently too late,” he also adds, “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.”

The app actually works by analyzing images of a person’s eye with the help of computer vision system that examines color metrics from the sclera, and eventually use artificial intelligence to correlate the information with bilubrin levels. It’s also great to know that BiliScreen has been tested in up to 70 individuals and it has been delivering accurate results 90% of the time — this shows that this technology is incredibly promising and way better than traditional forms of jaundice detection.

Future tests will still be conducted to learn more about the utility of the app in a larger group of people at risk of pancreatic cancer, hepatitis and other related diseases. The researchers are also looking to improve the usability of the app so people can use it without additional accessories like the box and glasses.