If we had a breakthrough machine-learning tool to assist us discover our body's own clock, we could make healthier sleep selections.
Researchers in Surrey and Groningen looked into whether blood metabolites and machine learning techniques could use our circadian rhythms to tell what time of day it was. The results were written up in the Proceedings of the National Academy of Sciences. Dim light melatonin onset (DLMO), which is the time when our natural melatonin cycle starts, is now the standard for circadian timing. Debra Skene, a professor at the University of Surrey and one of the study's co-authors, said, "After taking two blood samples from our participants, our method was able to predict the DLMO of individuals as well as or better than older, more invasive estimation methods."
Blood samples were taken from 12 men and 12 women every so often. Before going to the university's clinical research facility, all of the participants had been healthy for seven days, hadn't smoked, and kept a regular sleep routine. A targeted metabolomics method was used to study the circadian rhythms of more than 130 chemicals.
Professor Skene said, "We are excited about our new way of predicting DLMO, but we are also being careful." It is easier and needs less data than other methods. Our plan might help people with sleep problems caused by their circadian clocks and people who are healing from injuries, but more study is needed."Smart devices and wearables give helpful information about sleep patterns," the authors write. "However, our research paves the way for truly personalized sleep and meal plans that are in sync with our biology and have the potential to improve health and lower the risk of serious illness linked to poor sleep and eating at the wrong times."
Roelof Hut, a coauthor from the University of Groningen, said, "Our results could help to develop an affordable way to estimate our own circadian rhythms, which will improve the timing of behaviors, diagnostic sampling, and treatment."
