A new smartphone application (app) ‘BiliScreen’ developed by a research team from the University of Washington (UW) in the US has been designed to use selfies for the early screening of pancreatic cancer.
The app identifies increased bilirubin levels in the white part of the eye called sclera by using a smartphone camera, computer vision algorithms, and machine learning tools.
Go deeper with GlobalData
Enhanced bilirubin levels in the blood are known to cause yellow discolouration of the skin and eyes (jaundice), which is considered an early symptom of pancreatic cancer.
For validation in various lighting conditions, the app was tested using paper glasses that were printed with coloured squares to calibrate colour and a 3D-printed box that could block ambient lighting.
According to a clinical study of 70 people, the app correctly detected cases of concern 89.7% of the time, when compared to the existing blood test.
See Also:
The app is based on the UW’s other smartphone app called BiliCam, which is developed to spot newborn jaundice by analysing an image of baby's skin.
How well do you really know your competitors?
Access the most comprehensive Company Profiles on the market, powered by GlobalData. Save hours of research. Gain competitive edge.
Thank you!
Your download email will arrive shortly
Not ready to buy yet? Download a free sample
We are confident about the unique quality of our Company Profiles. However, we want you to make the most beneficial decision for your business, so we offer a free sample that you can download by submitting the below form
By GlobalDataDesigned as an easy-to-use and non-invasive tool, the BiliScreen app intends to assist pancreatic cancer patients who need periodic bilirubin monitoring.
The UW team developed an accompanying computer vision system for automatic isolation of the eye’s sclera when BiliScreen collects pictures of the eye as a person takes a selfie.
Based on the wavelengths of light being reflected and absorbed in the sclera, BiliScreen calculates colour information and uses machine learning algorithms to correlate it with bilirubin levels.
University of Washington Pediatrics Medicine department professor Dr Jim Taylor said: “This relatively small initial study shows the technology has promise.
“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 UW team intends to test the app on a wider population who are at jaundice risk and will work on improvements such as eliminating the need for accessories for the app.
Image: BiliScreen is a new smartphone app that can screen for pancreatic cancer by having users snap a selfie. Photo: courtesy of Dennis Wise/University of Washington.
