New tech could detect sleep disorders

By Pol Allingham via SWNS

Machine learning technology can predict sleep disorders such as insomnia, according to a new study.

Insomnia, sleep apnea, restless leg syndrome and narcolepsy are all examples of sleep disorders, instances of which have almost doubled in the UK since 2012, according to the NHS.

U.S. experts are documenting the same steep rise, leading researchers to develop a machine-learning model that could predict sleep disorder diagnoses with “strong accuracy.”

According to a new machine, the greatest predictors of a sleep disorder are depression, weight, age, and weight circumference.

Researchers warned sleep conditions are “significant” risk factors for diabetes, heart disease, obesity and depression.

Publicly available medical data was analyzed using the machine-learning model XGBoost for the study published in PLOS One.

The team from Northwestern Feinberg University School of Medicine, and Virginia Commonwealth University School of Medicine, both in the U.S., hope machines like XGBoost could effectively screen patients for sleep disorder risk without relying on a doctor’s bias or judgment.

Researcher Samuel Huang said: “What sets this study on the risk factors for insomnia apart from others is seeing not only that depressive symptoms, age, caffeine use, history of congestive heart failure, chest pain, coronary artery disease, liver disease, and 57 other variables are associated with insomnia, but also visualizing the contribution of each in a very predictive model.”

They analyzed 684 variables from data for 7,929 patients, including responses to questionnaires on demographic, diet, exercise and mental health.

Physical and laboratory information was assessed too.

A third of the group, 2,032 patients, had a diagnosed sleep disorder. Of the 684 variables, 64 were useful in deducing participants’ risk.