Contributed Talk Sessions | Poster Sessions | All Posters | Search Papers

Poster Session C: Friday, August 15, 2:00 – 5:00 pm, de Brug & E‑Hall

Longitudinal Multimodal Data Fusion Reveals Brain–Symptom Change Patterns in Depression

Effat Salehi Far¹, Tilo Kircher, Udo Dannlowski, Lea Teutenberg, Kira Flinkenflügel, Eva Mennigen¹; ¹Technische Universität Dresden

Presenter: Effat Salehi Far

Understanding how brain structure and psychopathology co-evolve over time is central to unravel individual variability in affective disorders. We applied multiset canonical correlation analysis followed by joint independent component analysis (mCCA + jICA) to longitudinal data from 105 participants (48 healthy controls, 57 with major depressive disorder, MDD), using cortical surface area (SA), cortical thickness (CT), and symptom checklist-90-revised (SCL-90) symptom scores as input modalities. Five joint components were extracted. Two components revealed significant brain–symptom change associations involving depression-dominant SCL-90 profiles. Regions showing progressive reduction in SA and CT overlapped with those reported by ENIGMA studies in adolescent and adult MDD, respectively. Correlations between structural and symptom component loadings were stronger in the MDD subgroup than in controls, highlighting diagnosis-specific change trajectories. Our results support multimodal fusion as a promising approach to identify clinically meaningful brain–symptom change patterns and better understand brain–behavior coupling in affective disorders.

Topic Area: Methods & Computational Tools

Extended Abstract: Full Text PDF