(Image source from: Canva.com)

Researchers have developed an artificial intelligence-based tool that can predict mood disorder episodes 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, happiness, or mania. Mood disorders are closely related to the sleep-wake rhythm, the disruption of which can lead to mood swings. The growing popularity of wearable devices has made health data collection much easier, said a group of researchers from the South Korea Institute of Basic Science and others. “Developing a model that predicts mood episodes based solely on sleep-wake pattern data has reduced the cost of data collection and significantly improved clinical utility,” said lead researcher Kim Jae-kyung. For the study, published in the journal NPJ Digital Medicine, researchers analyzed 429 days of data from 168 patients with mood disorders. The sleep-wake or circadian rhythm of 36 people was extracted and used to train the machine learning algorithm.

Machine learning algorithms, a type of artificial intelligence (AI), learn to recognize patterns in data that are trained to predict the future. Therefore, the AI model developed by the team was able to predict depressive, manic and hypomanic episodes with an accuracy of 80, 98 and 95%, respectively. “Using mathematical modeling on longitudinal data from 168 patients, we obtained 36 sleep markers and circadian rhythms,” the authors write. Researchers have found that daily changes in circadian rhythms are important predictors of mood swings. In particular, delays in circadian rhythms, falling asleep and waking up late in the day increase the risk of depressive episodes. On the other hand, researchers say, the more advanced your circadian rhythm, i.e., the more likely you are to have a manic episode. h. They go to bed and get up earlier. “In particular, diurnal changes in circadian phase were the most important predictors,” the study authors added; Latency was associated with depressive episodes and the development of manic episodes.”