(Australian Associated Press)
A computer vision system that can automatically detect a tiny baby’s face in a hospital bed and remotely monitor its vital signs has been developed by researchers in Adelaide.
The University of South Australia says the system is the first time artificial intelligence software has been used to reliably detect a premature baby’s face and skin when covered in tubes, clothing, and undergoing phototherapy.
Engineering researchers and a neonatal critical care specialist remotely monitored heart and respiratory rates of seven infants in the Neonatal Intensive Care Unit at Adelaide’s Flinders Medical Centre.
“Babies in neonatal intensive care can be extra difficult for computers to recognise because their faces and bodies are obscured by tubes and other medical equipment,” lead researcher Javaan Chahl said.
“Many premature babies are being treated with phototherapy for jaundice, so they are under bright blue lights, which also makes it challenging for computer vision systems.”
Dubbed the “baby detector” and using a digital camera, the system was developed from a dataset of videos of babies in intensive care to reliably detect skin tone and faces.
Vital sign readings matched those of an electrocardiogram and in some cases appeared to outperform the conventional electrodes, endorsing the value of non-contact monitoring of pre-term babies in intensive care.
The study is part of an ongoing University of South Australia project to replace contact-based electrical sensors with video cameras, avoiding skin tearing and potential infections that adhesive pads can cause to fragile skin.
Infants were filmed with high-resolution cameras at close range and vital physiological data was extracted using processing techniques that can detect subtle colour changes from heartbeats and body movements not visible to the human eye.
Neonatal critical care specialist Kim Gibson said using neural networks to detect the faces of babies was a significant breakthrough for non-contact monitoring.
“In the intensive care setting it is very challenging to record clear videos of premature babies,” she said.
“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.”
The university said worldwide more than 10 per cent of babies were born prematurely and needed their vital signs monitored continuously.
Professor Chahl said non-contact monitoring was also particularly relevant given the COVID-19 pandemic and the need for physical distancing.