Deep learning for depression prediction in older adults: A retrospective cohort study from CHARLS (2011–2020) with independent cohort validation in CLHLS (2008–2018) | Synapse
March 3, 2026
Deep learning for depression prediction in older adults: A retrospective cohort study from CHARLS (2011–2020) with independent cohort validation in CLHLS (2008–2018)
Key Points
Depression prediction accuracy improves with deep learning algorithms, suggesting a novel tool for early intervention.
The model achieved a sensitivity of 85% and specificity of 90% in identifying depression cases within a retrospective cohort.
Retrospective cohort analysis used data from CHARLS (2011–2020) and validated results with an independent cohort from CLHLS (2008–2018).
The findings highlight the promising role of advanced algorithms in mental health assessment, warranting further exploration.