24 november 2021 – 09:05
How infrared cameras, smart algorithms and an intelligent pacifier can help premature babies.
Apnea, a sudden interruption of breathing, affects adults and children alike. It especially affects premature babies, whose respiratory system is not yet fully developed. In collaboration with Máxima MC and Philips, researcher Ilde Lorato of Eindhoven University of Technology has developed a new method to monitor apnea in children without causing unnecessary damage. The new approach uses infrared cameras, image processing software and even a smart pacifier. Lorato will defend her PhD research on Friday 19 November.
Apnea, a disorder in which breathing is interrupted for 10 seconds or more, is a common problem in adults when they sleep. But apnea is also common in children. Almost all premature babies younger than thirty weeks or weighing less than one kilogram suffer from it.
“Apnea in premature babies can cause serious problems, especially if it lasts longer than 20 seconds,” says Ilde Lorato, researcher in the department Electronic Systems from the Department of Electrical Engineering.
Peter Andriessen, neonatologist in Máxima MC (MMC), adds: “Apnea in premature babies is associated, among other things, with cerebral palsy or poor neurological development later in life. It is therefore very important that we can reliably monitor these infants, so that we can intervene immediately if necessary.”
Apnea in premature babies can be treated by tactile stimulation (massage, patting, pressing, stroking, etc.), or by giving the babies extra oxygen or caffeine.
Experiments in the hospital
Premature babies are nursed in so-called neonatal intensive care units (NICUs), where breathing is monitored with the same electrodes that also measure the activity of the heart. “However, this solution isn’t ideal,” Lorato says, “because electrodes don’t detect apnea very well. “
“In addition, the skin of premature babies is very sensitive. It can be easily irritated or even damaged by the use of adhesive electrodes. We have therefore looked for alternatives that cause less discomfort for these vulnerable infants, such as cameras.”
However, the use of cameras in a clinical context is not without problems, especially if you also want to monitor the patients at night, when there is insufficient light. In addition, babies tend to move a lot, leading to unwanted noise in the data. And then, of course, there’s the issue of privacy and cost.
“It was therefore very important to test our solution in a hospital. Only then can you see if it really works in a complex environment,” says Lorato. And that is exactly what the Italian researcher has done.
Thermal image
Thanks to the good cooperation between TU/e, Philips and Máxima MC within e/MTIC, she was able to carry out practical experiments in the Neonatal Medium Care Unit of MMC. There, she used three low-resolution infrared cameras to track fifteen babies, each nursed in an open bed for three hours.
This image shows how smart algorithms can visualize the movement and flow of breath (combined or separately) in an infrared image (left). The signals are then translated into a respiratory rate (RR). This makes it possible to detect potential apnea. The two frequencies derived from the video match the reference frequency (right). (click on image for YouTube video)
“We chose thermal imagers because they also work in complete darkness. Infrared cameras also show both the breathing movement and the flow. Their low resolution also means they are cheaper and privacy is better guaranteed,” says Lorato. “We then used the infrared images to develop a number of new image processing algorithms that can potentially detect apnea.”
“Our algorithms are able to automatically identify pixels that contain respiration data. They even work without first determining the relevant parts of the body or face, which is a challenge with small babies in a hospital environment and with low resolution images. With some adjustments, they can also be used when the babies are moving.”
smart pacifier
During her research at the hospital, Lorato made an interesting discovery. She saw that the pacifier that infants suck on causes movements that resemble breathing. This, of course, distorted her test results. That’s why she came up with the idea to use a smart pacifier that registers the sucking movements. They can be so easily distinguished from the breath.
Lorato and MMC are pleased with the results, because they show that you can monitor the breathing of premature babies with a number of cheap infrared cameras. Hopefully that will lead to the removal of the annoying electrodes in some babies. Lorato’s research has already resulted in two patent applications (one for the pacifier and one for the algorithm). Still, there is still a long way to go before these types of cameras become a common sight in neonatal intensive care units.
Lorato: “The infrared cameras cannot see through the incubator wall, so if you want to monitor babies in an incubator, you have to place the cameras in the incubator. Furthermore, our videos do not contain images of real apnea. We had to simulate these by editing the available video material. So there is still some work to be done.”
The follow-up study will take place within the UMOSAproject by Sander Stuijk, the senior lecturer who supervised Lorato’s PhD project. This project is also looking at other applications of thermal imaging, for example for sleep apnea in adults. In Máxima MC, a new PhD student will PICASSOproject to follow up on the development of less intrusive monitoring techniques for premature babies.
Ilde Lorato defends her thesis Video Respiration Monitoring: Towards Remote Apnea Detection in the Clinici on 19 November at the TU/e. She was supervised by Gerard de Haan, Sander Stuijk and Carola van Pul.
This research was conducted within the Eindhoven MedTech Innovation Center (e/MTIC) and the Center for Care & Cure Technology Eindhoven (C3Te). Both centers strive for innovation in healthcare in close collaboration between TU/e, hospitals and the business community.
Bron: YOU / e
—
Editorial Medicalfacts/ Janine Budding
I studied physiotherapy and health care business administration. I am also a registered Independent client supporter and informal care broker. I have a lot of experience in various positions in healthcare, the social domain and the medical, pharmaceutical industry, nationally and internationally. And have broad medical knowledge of most specialties in healthcare. And of the health care laws from which health care is regulated and financed. Every year I attend most of the leading medical conferences in Europe and America to keep my knowledge up-to-date and to keep up with the latest developments and innovations. I am currently studying applied psychology.
My messages on this weblog do not reflect the strategy, policy or direction of an employer, nor are they the work of or for a client or employer.
—-
