Every year in Canada, approximately 15,000 newborns are admitted to neonatal intensive care units (NICUs). Whether premature births or full-term babies with pressing medical conditions, they are among the most vulnerable patients in the country’s health-care system.
Because these newborns are in a precarious state, they must be monitored extremely closely, with sensors tracking vital signs such as heart rate, respiratory rate and blood oxygen level so nurses and doctors can take immediate action at the first indication of trouble. Sometimes, despite their fragility, babies have to be transferred from one location to another, such as when they require a specialized type of treatment.
To help ensure that babies are monitored as effectively as possible in the NICU, and that newborn transport is safe and smooth, Carleton University engineering researchers Jim Green and Rob Langlois are collaborating with the Children’s Hospital of Eastern Ontario (CHEO) on a pair of cutting-edge projects.
Ultimately, their work is about upholding the health of the tiniest humans by allowing clinicians to focus on providing world-class care.
A New Approach to Monitoring Vital Signs
The conventional way to monitor a baby’s vital signs in the NICU is via wired sensors attached to the fingertip, chest and other body parts. All those wires can make it difficult for parents to hold their newborns, even though skin-on-skin contact provides important benefits
Moreover, for various reasons, including patient movement disrupting sensors, roughly half of the alarms they trigger are false. This makes the NICU a loud and busy place, which is stressful for both families and staff who have to respond to all the beeping and ringing. (A baby’s vitals will often temporarily deviate from “normal,” so the alarm should sound, but it should be ignored if the baby’s condition improves on its own.)
In partnership with CHEO, Systems and Computer Engineering researcher Jim Green and his team of a dozen graduate students have developed a new approach: a pressure-sensitive mat that goes under the infant and an advanced imaging camera positioned above the baby’s isolette or crib.
The mat, which resembles a simple black bedsheet, detects time-varying contact pressure. This data, when processed using machine learning, can estimate respiration rate and interpret the meaning of certain movements. The camera can analyze colour changes on the baby’s face and skin and estimate heart rate and other vitals.
“The mat is so sensitive that as the baby breathes, even though it’s just a slight bounce on the mattress from the chest rising and falling, we can extract the respiration signal from all the noise and get an estimate of respiration rate,” says Green.
“These non-contact sensors enhance the capabilities of traditional wired sensors by adding layers of context, even when a baby is covered in a blanket. This information could change how clinicians interpret alarms and care for their tiny patients.”
This approach could also be extended to home care. “Imagine a scenario where a premature baby has been discharged but still requires monitoring,” says Green. “Instead of being tethered to wired sensors, the baby can be comfortably nestled in a crib equipped with non-contact sensors, creating an opportunity for parents to scoop them up without being tangled with wires.”
Minimizing Vibration and Noise During Transit
When a newborn needs to be moved between health-care facilities, specialized pediatric transport teams follow strict protocols. In Ontario, they use the Neonate Patient Transport System, which features an isolette augmented with a ventilator, medical air, infusion pumps, monitors, a defibrillator and several other devices.
But even though the baby is secured within a five-point harness and the 400-pound unit is latched onto a stretcher, the vibration and noise in an ambulance or airborne helicopter can pose health risks.
Mechanical and Aerospace Engineering researcher Rob Langlois and his team at Carleton’s Applied Dynamics Laboratory do experiments on all sorts of vehicles in motion, from cars and buses to fire trucks and ships.
“We’ve got lots of experience with the dynamics of objects and human bodies in moving environments,” he says.
“Whether it’s a helicopter secured to the deck of a rolling ship or a stretcher secured in an ambulance, they share a lot of the same issues.”
Working with CHEO, Toronto’s SickKids, the National Research Council of Canada and other partners, Langlois is exploring ways to reduce exposure to vibration and noise for newborns in transit.
To date, they have conducted a range of lab and field tests, shaking, rotating and vibrating medical equipment made for transporting infants — or, in some cases, shaking an entire ambulance with a stretcher, medical equipment and patient manikin inside to simulate driving along various road types.
By looking at variables such as how patient transport equipment is designed and secured in vehicles, they are aiming to reduce the vibrations and sound experienced by patients and to develop novel approaches for real-time monitoring.
“This work will improve infant safety during medical transportation,” says Langlois, “and find ways to mitigate noise and vibration exposure and improve health outcomes.”
Although this project is ongoing, the researchers already have a deeper understanding of how vibration transmits through the patient transport systems and how effective several approaches could be toward mitigating vibration exposure.
Jim Green and fellow Carleton Systems and Computer Engineering researcher Adrian Chan are also involved in this research, and though it’s different than the NICU monitoring work, the overarching goal is the same: giving sensitive babies the healthiest start possible.
