Physiological Dynamics of the Sleep Onset Process: Conventional and Novel Analyses Using SHHS Data
2022 Focused Projects Grant for Junior Investigators
Yan Ma, MD, MPH
Brigham and Women’s Hospital and Harvard Medical School
Key Project Outcome
This research project explored new ways to identify insomnia using brief recordings of brain and heart activity during wakefulness before sleep. Traditionally, diagnosis and assessment insomnia rely on patient self-reports and questionnaires, which can miss key physiological signs of the disorder. Our study found that differences in brainwave patterns and heart rate dynamics can be detected within the first few minutes of lying in bed before sleep, allowing us to identify insomnia objectively and conveniently. These findings suggest that early pre-sleep stress, a known contributor to insomnia, may be measured in an objective way.
Using artificial intelligence, we tested the accuracy of predicting insomnia from heart rate data gathered just after lights out, achieving a relatively high level of accuracy, which indicates that insomnia can potentially be monitored over time using wearable technology, creating new opportunities for real-time, personalized sleep assessments.
By establishing physiological markers for insomnia, our findings pave the way for more effective and accessible tools to help individuals manage and understand their sleep health. This approach could lead to better outcomes for people with insomnia by enabling earlier diagnosis and more targeted treatments that address pre-sleep stress.
Abstracts
SLEEP
Pre-sleep wakefulness in patients with insomnia: heart rate variability during sleep onset as a physiological biomarker
Detecting Insomnia and Predicting Difficulty Falling Asleep: Machine Learning of Heart Rate Data During Sleep Onset