New App uses Selfie for early Detection of Pancreatic Cancer

Researchers at the University of Washington (UW) have created and mobile app that can detect various diseases such as pancreatic cancer while clicking a selfie. Pancreatic cancer is very difficult to detect due to very few noninvasive screening tools and revealing symptoms. One of the early symptoms of this disease is jaundice, in which eyes and skin get yellow due to accumulation of bilirubin in the blood. Detection through bilirubin levels is difficult for bare eyes to actualize.

Targeting Sclera for Detection of Bilirubin Level

A doctoral student at the Paul G. Allen School of Computer Science & Engineering at UW and the lead author of the study explained that the major problem with pancreatic cancer is that it is very late when you are symptomatic. This gives patients very less time to treat themselves. He shares that the motive of creating this app is to let the patients detect their disease in early stages and provide them more time to treat themselves. This app could save many lives.

This app uses machine-learning tools, computer vision algorithms, and a smartphone camera to detect increasing levels of bilirubin in a patient’s sclera. In a test, 70 people used this app and the results shows that this app was 89.7% accurate. Ph.D., Professor of computer science and engineering and electrical engineering at UW, and senior study investigator Shwetak Patel noted that studying sclera can tell the level of bilirubin in blood. She was motivated to use these changes in sclera due to changes in bilirubin for early detection through easy means.