The smart phone study: assessing the reliability and accuracy of neonatal jaundice measurement using smart phone application

Abstract

Background: Current gold standard for bilirubin estimation in newborns is invasive while transcutaneous bilirubinometer which is a noninvasive option is costly. Smart phone applications can be used as an alternative to measure bilirubin. We did the study to evaluate “Biliscan” medical application for screening of neonatal jaundice.Methods: During the first seven days, neonates with suspected jaundice were taken blood samples for serum bilirubin. Within two hours photographs were taken using Biliscan app by placing a colour calibration card over the chest. Bilirubin estimate obtained after colour balancing, feature extraction and machine learning regression were compared with blood values.Results: There is a good correlation (0.6) between Biliscan bilirubin estimation and serum bilirubin levels (p <0.0001) in the present study on 35 neonates. Biliscan thorax values correlated better than abdomen values (0.6 versus 0.551) with serum values. Hence chest is the preferred area for Biliscan measurement. Conclusions: Biliscan app appears to be a good cheap option to screen for jaundice in newborns noninvasively.

Neonatal wearable device for colorimetry-based real-time detection of jaundice with simultaneous sensing of vitals

“…Actually, only a smartphone with a camera and a dedicated app were required. However, the accuracy of this method has been reported to vary widely, with R = 0.48 to 0.91 ( 21 , 22 , 43 , 44 ). The measurements were affected by the surrounding environment.…”

Bilirubin estimates from smartphone images of newborn infants’ skin correlated highly to serum bilirubin levels

“…Studies published in the last few years have described novel ways of assessing neonatal jaundice by using digital images or smartphones, by adopting slightly different approaches. [11][12][13][14][15][16] Based on previous research on the bio-optical properties of newborn skin, we have developed a physics-based system that estimates bilirubin from digital images. 17 The purpose of this study was to evaluate the accuracy of this new method.…”

“…Mobile phone health (mHealth) is of interest, because even areas with scarce economic resources can use it 10 . Studies published in the last few years have described novel ways of assessing neonatal jaundice by using digital images or smartphones, by adopting slightly different approaches 11‐16 …”

“…Rong et al conducted a Chinese study on BiliScan machine-learning software (Shenzen Beishen Healthcare Technology Co), which provided bilirubin estimates from smartphone images, with a correlation of r = .79 14. Swarna et al tested the BiliScan application on35 Indian infants and found that the correlation was substantially lower at r = .6 16. The difference between these two studies might have been because the software had not been trained on Indian infants.…”

The Need for the Establishment of Biomedical Engineering (BME) as an Academic and Professional Discipline in the Philippines — A Quantitative Argument

“…The data acquired from Biliscan were compared with measures of total serum bilirubin (TSB), which is the current gold standard, as reference. It was then found that there exists a good correlation of ~0.6 between Biliscan estimates and TSB levels [73]. Additionally, Pelham et al designed a portable, accurate, and low-cost device called BiliSpec that is capable of quantifying TSB levels from small drops of whole blood applied to a lateral flow card.…”