In a new study, researchers from the University of Copenhagen’s Department of Computer Science have collaborated with the Danish Center for Sleep Medicine at the Danish hospital Rigshospitalet to develop an artificial intelligence (AI) algorithm that can improve diagnoses, treatments, and overall understanding of sleep disorders.
Thousands of Danes suffer from a range of sleep disorders, including insomnia, sleep apnea, and narcolepsy. Furthermore, it is estimated that up to 200,000 Danes have undiagnosed sleep apnea.
Dr. Mathias Perslev, lead author of the study remarked that “the algorithm is extraordinarily precise. We completed various tests in which its performance rivaled that of the best doctors in the field, worldwide.”
Admittance to a sleep clinic is usually the first step in a sleep disorder examination today. Here, a person’s night sleep is monitored using various measuring instruments. The 7-8 hours of measures from the patient’s overnight sleep are then reviewed by a sleep disorder specialist.
The doctor manually splits these 7-8 hours of sleep into 30-second intervals, which must all be classified as different sleep phases such as REM (rapid eye movement) sleep, light sleep, deep sleep, and so on. It’s a time-consuming task that the algorithm can complete in a matter of seconds.
Dr. Poul Jennum, Prof of Neurophysiology and Head of the Danish Center for Sleep Medicine added that “this project has allowed us to prove that these measurements can be very safely made using machine learning which has great significance. By saving many hours of work, many more patients can be assessed and diagnosed effectively.”
More than 4,000 polysomnography examinations, also known as PSG or sleep studies, are conducted annually on patients with sleep apnea and other more difficult sleeping problems in the Capital Region of Denmark alone.
It takes 1.5-3 hours for a doctor to analyze a PSG study. By implementing the new algorithm, between 6,000 and 12,000 medical hours might be freed up in the Capital Region of Denmark alone.
“We have collected sleep data from across continents, sleep clinics, and patient groups. The fact that the algorithm works well under such diverse conditions is a breakthrough,” said Dr. Perslev.
They hope that, in the future, the algorithm will aid doctors and researchers all over the world in learning more about sleep problems.
Dr. Perslev stated that “just a few measurements taken by common clinical instruments are required for this algorithm. So, the use of this software could be particularly relevant in developing countries where one may not have access to the latest equipment or an expert.”
The researchers are currently working with Danish doctors to get the software and algorithm certified for clinical use.
