PREDICTING DEPRESSION RISK IN EARLY ADOLESCENCE VIA MULTIMODAL BRAIN IMAGING

Area of Science:

• Neuroscience
• Psychiatry
• Machine Learning

Background:

• Depression is a prevalent psychiatric disorder in adolescents, with early intervention crucial.
• Identifying at-risk youth before adolescence is vital for effective prevention strategies.
• Family history of depression is a significant risk factor.

Approach:

• Utilized machine learning on multimodal neuroimaging data from the Adolescent Brain Cognitive Development (ABCD) Study (N=2658).
• Applied rigorous leave-one-site-out cross-validation for robust prediction model assessment.
• Investigated prediction accuracy in a high-risk cohort with parental depression history (N=625).

Key Points:

• All neuroimaging features significantly predicted depression risk beyond chance.
• Resting-state functional connectomes demonstrated superior predictive performance in the high-risk group.
• Functional brain connectivity outperformed structural MRI and task-based fMRI in predicting depression onset.

Conclusions:

• Brain functional connectivity effectively predicts depression risk in early adolescence.
• Connectome interactions capture greater individual variability in psychopathology than single-region measures.
• This approach highlights the potential for neuroimaging in early depression risk assessment.