A study of the accuracy of a technology for assessing jaundice in outpatient neonates based on analysis of digital images demonstrated that this smartphone application (app) may be useful for screening newborns for jaundice.
A study of the accuracy of a technology for assessing jaundice in outpatient neonates based on analysis of digital images demonstrated that this smartphone application (app) may be useful for screening newborns for jaundice. Investigators obtained paired BiliCam (the name of the app) images and total serum bilirubin (TSB) levels in a diverse sample of newborns (aged younger than 7 days) at 7 sites across the United States. Using specialized software, they extracted data on color values in the images, then used specialized techniques to develop models for predicting an estimated bilirubin level for each of the 530 newborns.
Overall, the correlation between the estimated bilirubin levels and actual TSB levels was 91%, with the highest correlation among white neonates (92%) and the lowest among Asian American newborns (88%). The sensitivities of BiliCam in identifying newborns with high TSB levels were 84.6% according to 1 decision rule and 100% according to another. Using these same decision rules, specificities were 75.1% and 76.4%. Investigators noted that BiliCam might best be used as a screening device to determine which neonates require a blood draw to determine a TSB level (Taylor JA, et al. Pediatrics. 2017;140:e20170312).
BiliCam is not ready to replace total serum bilirubin in assessment of neonatal jaundice. However, like transcutaneous bilirubin measurement, this app may have a role in screening with the added advantages of low cost and ability to gather data from anywhere with Internet access.