Jaundice affects about 60 percent of newborn babies, and while it usually clears up on its own, it can lead to hearing loss or brain damage. It's therefore important to detect and treat the condition as soon as possible, which is exactly what a new camera-based system is designed to do.
Babies with jaundice have visibly yellow-ish skin, due to excessive bloodstream levels of an orange-yellow pigment known as bilirubin. In cases where those levels are getting dangerously high, blue light exposure is used to break down the bilirubin in the infant's skin.
Previously, we've seen optical devices – either hand-held by a caregiver or adhered to a baby's forehead – that are able to detect jaundice by analyzing an infant's skin color. The former is only used when a caregiver is present, however, while the latter may irritate the baby's skin.
Seeking a full-time, less invasive alternative, scientists from the University of South Australia and Iraq's Middle Technical University developed a system wherein an ordinary 1080p webcam is constantly pointing at the infant from a distance of 30 cm (11.8 in). A linked Arduino Uno microprocessor is used to continuously analyze the camera's output, via a custom program.
If it's determined that the baby's skin is becoming too yellow, the Arduino automatically triggers an adjacent blue LED light to illuminate, plus it wirelessly sends an SMS alert to a designated caregiver – the whole process takes just one second. The caregiver can subsequently come and check the infant for themselves, allowing the phototherapy treatment to continue if it's deemed necessary.
The system has already proven accurate at assessing bilirubin levels in 20 newborns that were already known to have jaundice, plus it was able to differentiate between healthy and jaundiced infants in 16 supplied photos. Further testing is now being planned.
"[Other] methods trialed have been unreliable, costly, inefficient and in some cases caused infections and allergies where sensors needed skin contact," says the lead scientist, U South Australia's Prof. Javaan Chahl. "Our system overcomes these obstacles by immediately detecting jaundice based on a novel digital representation of color which allows high diagnostic accuracy at a relatively low cost. It could be widely used in hospitals worldwide and medical centers where laboratory facilities and trained medical staff are not available."
The research is described in a paper that was recently published in the journal Designs.
Source: University of South Australia
