Researchers in Denmark applied an AI algorithm to analyze the 7-8 hours of data from a patient’s overnight sleep.

Sleep disorders range from sleep apnea to insomnia, narcolepsy and many more. For the typical diagnostic procedure (called a polysomnography test—or PSG or sleep study)  the patient is admitted into a sleep clinic overnight. Various measuring instruments monitor the patient’s sleep patterns and their data is then reviewed by a sleep specialist. It could take the doctor as long as three hours to analyze the readings. However, an AI algorithm can perform the same task in seconds. (Futurity)

Researchers at the University of Copenhagen trained an AI algorithm on data from 20,000 sleep nights from the U.S. and various European countries.

As a result the algorithm could divide a night’s data of sleep into 30-second intervals, and categorize them into different sleep phases, such as REM (rapid eye movement) sleep, light sleep, deep sleep, etc. – all of which the doctor did manually.

Very satisfying results on sleep disorders

“The algorithm is extraordinarily precise,” says Mathias Perslev, a PhD in the computer science department at the University of Copenhagen and lead author of the study. “We completed various tests in which its performance rivaled that of the best doctors in the field, worldwide.”

“This project has allowed us to prove that these measurements can be very safely made using machine learning—which has great significance,” says Poul Jennum, professor of neurophysiology and head of the Danish Center for Sleep Medicine. “By saving many hours of work, many more patients can be assessed and diagnosed effectively.”

Open source

The sleep analysis software is freely available at sleep.ai.ku.dk and can be used by anyone, anywhere.

“Just a few measurements taken by common clinical instruments are required for this algorithm,” Perslev says.

“So, use of this software could be particularly relevant in developing countries where one may not have access to the latest equipment or an expert.”

Related Story:   AI May Provide An Objective Second Opinion On Parkinson’s