Researchers develop AI tool to predict episodes of mood disorder using sleep data

Researchers have developed an AI-based tool that can predict episodes of mood disorder in patients using only sleep-wake data recorded by wearable devices such as smartwatches. People suffering from mood disorders, including bipolar disorder, experience long periods of sadness, depression, joy or mania. Mood disorders are closely linked with sleep-wake rhythms with disturbances potentially triggering a mood episode.

The team of researchers, including from the Institute for Basic Science in South Korea, said the growing popularity of wearable devices has made collecting health data much easier.

"By developing a model that predicts mood episodes based solely on sleep-wake pattern data, we have reduced the cost of data collection and significantly improved clinical applicability."


"This study offers new possibilities for cost-effective diagnosis and treatment of mood disorder patients," lead researcher Kim Jae Kyoung said.

For the study, published in the journal 'npj Digital Medicine', the researchers analysed 429 days' worth of data from 168 mood disorder patients. Thirty-six sleep-wake, or circadian, rhythms were extracted, which were used for training machine learning algorithms.

A form of artificial intelligence (AI), a machine learning algorithm learns to detect patterns in the data it is trained on to make predictions about the future.

The AI model that the team developed was thus able to predict depressive, manic, and hypomanic episodes with an accuracy of 80 per cent, 98 per cent and 95 per cent, respectively.

"Using mathematical modelling to longitudinal data from 168 patients, we derived 36 sleep and circadian rhythm features. These features enabled accurate next-day predictions for depressive, manic, and hypomanic episodes," the authors wrote.

The researchers found that daily changes in circadian rhythms are a key predictor of mood episodes.

Specifically, delayed circadian rhythms, falling asleep and waking up later in the day, increase the risk of depressive episodes. On the other hand, advanced circadian rhythms - sleeping and waking up earlier in the day, increase the risk of manic episodes, the researchers said.

"Notably, daily circadian phase shifts were the most significant predictors: delays linked to depressive episodes and advances to manic episodes," the authors of the study added.