Ecologically-Valid Emotion Signatures Enhance Mood Disorder Diagnostics.

Area of Science:

• Neuroscience
• Psychiatry
• Medical Imaging

Background:

• Mood disorders (MDD, BD) are prevalent, causing emotional dysregulation and functional impairment.
• Existing neuroimaging research lacks reliable neurofunctional markers for these conditions.

Purpose of the Study:

• To introduce Divergent Emotional Functional Networks (DEFN) as a novel neurofunctional marker for mood disorders.
• To validate DEFN using functional magnetic resonance imaging (fMRI) in naturalistic contexts.

Main Methods:

• Utilized naturalistic emotion induction, dynamic functional connectivity (dFC), and machine learning.
• Trained and validated DEFN models on healthy individuals and clinical cohorts (MDD, BD).
• Employed static functional connectivity (sFC) and 10-fold cross-validation for model assessment.

Main Results:

• DEFN identified emotion-specific patterns in healthy individuals with 83.99% accuracy.
• DEFN-based models achieved higher classification accuracy for MDD (70.33%) and BD (75.18%) compared to baseline models.
• DEFN demonstrated high reproducibility across age and sex.

Conclusions:

• DEFN represents a promising, reproducible, and clinically relevant neural marker for diagnosing mood disorders.
• This approach offers potential for more effective and timely interventions in psychiatry.
• DEFN enhances diagnostic capabilities beyond traditional methods.