Neonatal jaundice, a yellowing of the skin as a result of hyperbilirubinemia, is an almost ubiquitous condition in newborn infants. In very rare circumstances, if an infant with signficant hyperbilirubinemia is undetected, kernicterus, a devastating and permanent neurologic condition, can develop. Still rare, but much more frequently, a neonate with significant jaundiced is not identified until the bilirubin level is ver high, necessitating complicated and expensive care to prevent bilirubin encephalopathy. Thus, the identification of infants with moderate hyperbilirubinemia, at levels that are easily treatable is a central focus of neonatal care in the US. Unfortunately, bilirubin levels typically peak in neonates around 96 hours of life, well after most infants are discharged from the newborn nursery. There is currently a lack of accurate, inexpensive and widely available methodologies to screen discharged infants for jaundice, leaving a notable void in the overall system of care designed to prevent severe hyperbilirubinemia. Our group of academic pediatricians and computer science and electrical engineering faculty has developed a non-invasive technology to measure bilirubin in newborn infants. The technology is based on the analysis of digital images of newborn skin that are obtained with a smartphone app that we have developed. A color calibration card is placed on the newborn's skin to control for different lighting, and other aspects of the image are standardized. The images are sent electronically to a central computer server for an analysis. The calculated bilirubin level is then downloaded to the smartphone app. In preliminary studies on a small number of infants, our technology, called the Bilicam app system, has accurately predicted bilirubin levels. The proposed study is designed to rigorously assess the ability of the Bilicam app system to accurately predict serum bilirubin levels in a racially, ethnically and geographically diverse sample of 250 newborn infants who are 3-5 days old, when bilirubin levels are peaking. The accuracy of the Bilicam app system in predicting serum bilirubin levels will be compared to published data on the accuracy of transcutaneous bilirubin meters, another non-invasive methodology for assessing newborn jaundice that is in widespread clinical use. The final goal of the project is to assess and improve the user experience with the smartphone app. A 2-year study period is planned.

Public Health Relevance

The lack of methodologies to accurately assess the severity of neonatal jaundice, as a result of hyperbilirubinemia, after discharge from the newborn nursery is a notable void in the overall system of care designed to prevent kernicterus, a devastating neurologic condition. The goal of this study is to assess the effectiveness of a novel technology, based on analysis of digital images obtaining using a smartphone app, in predicting serum bilirubin levels in newborn infants in outpatient settings.

National Institute of Health (NIH)
Eunice Kennedy Shriver National Institute of Child Health & Human Development (NICHD)
Exploratory/Developmental Grants (R21)
Project #
Application #
Study Section
Pregnancy and Neonatology Study Section (PN)
Program Officer
Higgins, Rosemary
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of Washington
Schools of Medicine
United States
Zip Code
Taylor, James A; Stout, James W; de Greef, Lilian et al. (2017) Use of a Smartphone App to Assess Neonatal Jaundice. Pediatrics 140: