Rest–activity timing phenotypes and mental health: a longitudinal analysis of 24-hour accelerometry in population-based cohorts
Peer-Reviewed Publication
Updates every hour. Last Updated: 22-Oct-2026 00:17 ET (22-Oct-2026 04:17 GMT/UTC)
In 94,344 UK Biobank participants, accelerometry objectively classified 24-hour rest-activity timing phenotypes. Over a median 6.8 years, the night-active phenotype was associated with higher incident MDD and anxiety risk, while an early-morning-active phenotype was associated with lower MDD risk.
The findings offer association evidence and exploratory MRI clues, but they do not establish that changing activity timing prevents mental disorders. Randomized trials are needed.
This study proposes a framework combining LightGBM, SHAP, and supervised clustering for mortality-risk assessment among liver-transplant candidates with HCC. Eight SHAP-selected variables and supervised clustering identified seven subgroups with increasing risk levels.
The work offers a method for modeling the dual risks of liver dysfunction and tumor progression, but it does not establish that the model improves clinical decisions or patient outcomes.