- University of Sheffield engineers are developing pioneering new technology that will transform how doctors can assess sleep disorders in children
- Technology is harnessing the power of the Internet of Things (IoT) to enable doctors to study sleep disorders remotely and without the need for a child to wear lots of sensors
- The collaboration with Sheffield Children’s Hospital and the University if Oxford is set to save time and money for the NHS while also saving children from going through distressing sleep disorder assessments
The new technology, being developed by a team of researchers led by Professor Lyudmila Mihaylova from the University of Sheffield’s Department of Automatic Control and Systems Engineering in collaboration with Sheffield Children’s Hospital and the University of Oxford, is harnessing the power of the Internet, artificial intelligence and machine learning to enable doctors to study sleep disorders remotely and without the need for children to wear lots of sensors.
More than 11,000 children are affected by sleep disorders every year. To diagnose their cause and identify potential treatments, children are usually required to stay in hospital overnight and wear a range of specialist sensors while they sleep.
These sleep studies enable doctors to study the child’s sleep pattern, but they can often be an uncomfortable and distressing experience.
Sleep studies can require repeated overnight stays in hospital, which can put additional strain on bed capacity.
Also the sensors currently used produce a huge amount of data that requires a considerable amount of time and resource for clinicians to analyse.
The new technology being developed by the Sheffield engineers, in a collaboration brought together by Pitch-In and funded by the Connecting Capability Fund, which is part of Research England, is utilising the potential of the Internet of Things (IoT) - a concept in which multiple devices are connected and can communicate with each other via the Internet.
Using one small wearable device that can be placed on the end of a child’s finger, the system can monitor their sleep pattern, automatically detect any sleep disorders and send the data to a clinician - meaning the sleep assessment can be easily done while the child is at home, sleeping in their usual setting and without the need to wear multiple sensors.
Through artificial intelligence and machine learning, the system can also automatically analyse the data from sleep assessments - freeing up more time for clinicians.
“Given the increasing demands that hospitals are faced with, there is a real need to reduce the human involvement in data processing.
“Currently a clinician has to spend long hours analysing the data that is gathered through sleep assessments for each patient. Also, the sleep assessment process can be uncomfortable and inconvenient for patients and their families, so we want to develop a technology which helps address both of these problems.
“This technology can be used on both children and adults who are suffering from sleep disorders. The project is a fantastic opportunity for us to lay the foundations of the hospitals of the future.”
Professor Lyudmila Mihaylova
Professor of Signal Processing and Control at the University of Sheffield
Dr Mahnaz Arvaneh, Lecturer in Automatic Control and Systems Engineering at the University of Sheffield, said: “This is a multidisciplinary project that requires tight collaborations between data scientists, system engineers, as well as clinicians, medical doctors and psychologists.
“In this project, we also try to address the barrier of acceptance of IoT by clinicians and patients through co-design and co-creation approach, where the stakeholders get familiar with this technology and discuss their concerns and expectations, including data privacy.
“Although this technology is for sleep disorders, it can be expanded as a diagnostic tool for neurological diseases such as dementia and Alzheimer's. As we monitor the breathing patterns, our technology could also be used for Covid-19 as a diagnostic and monitoring tool. In addition, the possibility of having this technology at home, means it can provide a service in a safer and more convenient environment, especially for example during a pandemic.”
Professor Heather Elphick, Consultant in Paediatric Respiratory Medicine at Sheffield Children’s Hospital, said: “The most complex type of sleep study is called a polysomnography and that involves over 20 different sensors being attached to the child’s body - to look at leg movements, breathing parameters, oxygen and carbon dioxide levels, cardiac signals and we monitor the electrical signals from the brain to identify the stages of sleep. Although it’s not painful, it can be uncomfortable and distressing for the child as it usually takes around an hour to put all of these sensors on.
“This technology will help us to save admissions to hospital, thereby saving money for the NHS, but also save staff time in terms of the time it takes to set up the sleep studies and analyse them manually.”
Navin Cooray from the University of Oxford, said: “The collaboration between our two universities has enabled the exploration and the development of machine learning techniques to utilise wearable devices to assist clinicians in making diagnostic decisions.
“At the University of Oxford we focus on a specific sleep disorder and their association with Parkinson’s disease. While our applications in sleep medicine differ, the core principles, techniques, and desired outcomes remain the same. Our collective research works towards providing clinicians with robust and effective analysis to support their diagnostic decisions in a timely and highly accurate manner.”