The University of South Australia has published a study where an AI-powered computer vision system that has the same accuracy as an ECG machine was used to remotely monitor premature babies' vital signs. The platform can also detect their faces while lying in hospital beds.

WHAT IT'S ABOUT

According to a press statement, the university's researchers developed the "baby detector" software using a dataset of videos of premature babies in the neonatal intensive care unit (NICU) to reliably detect their skin tone and faces.

In the study, whose findings were published in the Journal of Imaging, seven infants in NICU at the Flinders Medical Centre in Adelaide, South Australia were filmed using high-resolution cameras at close range. Their vital physiological data such as heart and respiratory rates were also extracted using advanced signal processing techniques that detect subtle colour changes in heartbeats and body movements not visible to the human eye.

It was demonstrated that the AI software can "reliably" detect a premature baby's face and skin when covered by tubes, clothing or undergoing phototherapy for jaundice. The researchers claimed that the vital sign readings "matched those of an ECG", even outperforming it in some cases. 

WHY IT MATTERS

The use of AI to detect babies' faces is regarded as a "significant breakthrough" for remote patient monitoring, according to UniSA neonatal critical care specialist Kim Gibson. "In the NICU setting, it is very challenging to record clear videos of premature babies. There are many obstructions and the lighting can also vary, so getting accurate results can be difficult. However, the detection model has performed beyond our expectations," she said. 

The study, according to UniSA, is part of an ongoing university project to replace contact-based electrical sensors with non-contact video cameras to avoid skin tearing and potential infections on babies' skins. Gibson says monitoring premature babies' health is traditionally done using adhesive electrodes placed on their skin which could be "problematic". 

THE LARGER TREND

While AI has been used to detect adult human faces, the researchers claimed that their study presents the first application of the technology in detecting babies' faces. 

Last year, the same UniSA researchers developed a technology now employed in commercial products sold by Canadian aerospace IT provider Draganfly to measure and screen adults' vital signs for COVID-19 symptoms.