Can we predict sleep health based on brain features? A large-scale machine learning study
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Objectives
Normal sleep is crucial for brain health. Recent studies have reported robust associations between sleep disturbance and various brain structural and functional traits. However, the complex interplay between sleep health and macro-scale brain organization remains inconclusive. In this study, we aimed to uncover the links between brain imaging features and diverse sleep health-related characteristics by means of Machine Learning (ML).
Methods
We used 28,088 participants from the UK Biobank to calculate 4677 structural and functional neuroimaging markers. Then, we employed them to predict self-reported insomnia symptoms, sleep duration, easiness getting up in the morning, chronotype, daily nap, daytime sleepiness, and snoring. We built seven different linear and nonlinear ML models for each sleep health-related characteristic to assess their predictability.
Results
We performed an extensive ML analysis that involved more than 100,000 hours of computing. We observed relatively low performance in predicting all sleep health-related characteristics (e.g., balanced accuracy ranging between 0.50-0.59). Across all models, the best performance achieved was 0.59, using a Linear SVM to predict easiness getting up in the morning.
Conclusions
The low capability of multimodal neuroimaging markers in predicting sleep health-related characteristics, even under extensive ML optimization in a large population sample suggests a complex relationship between sleep health and brain organization.
