Distinct Connectivity Signatures of Emotions Enhance Precision of Network Biomarkers in Mood Disorders

Abstract

Mood disorders, including Major Depressive (MDD) and Bipolar (BD) Disorder, are highly prevalent and debilitating conditions that contribute significantly to the global disease burden. These disorders are characterized by persistent emotional dysregulations, such as pervasive sadness and anhedonia, resulting in substantial functional impairments. Although neuroimaging studies have identified differences in brain activity and connectivity between individuals with MDD (MDDs) or BD (BDs) and healthy controls (HCs), reliable and reproducible neurofunctional markers for clinical diagnosis and treatment remain elusive. This study seeks to address this gap by introducing a novel approach that utilizes Divergent Emotional Functional Networks (DEFN), derived from functional magnetic resonance imaging (fMRI) during dynamic emotional processing in naturalistic contexts. Using a combination of naturalistic induction of sustained emotional experience with dynamic functional connectivity (dFC) and machine learning techniques, we decoded emotion-specific patterns of happiness and sadness in healthy individuals. Based on the dynamic functional connectivity signatures, we identify the DEFN and applied it to large clinical mood disorder datasets, including MDD (n=63) and BD patients (n=61). The model with DEFN demonstrated significant improvements in classification accuracy compared to conventional baseline models, achieving up to 10.75% and 9.92% performance increases in MDD and BD datasets, respectively. Additionally, DEFN were found to be highly reproducible across age, gender and models from emotion dataset, supporting the robustness of this model in distinguishing mood disorders from healthy controls. In conclusion, the DEFN approach presents a promising, reproducible, and clinically relevant neural marker for diagnosing and understanding emotional dysfunction in mood disorders, offering potential for more effective and timely interventions.