Network-based alterations in task-induced functional connectivity in bipolar disorder: An functional near-infrared...

Area of Science:

• Neuroscience
• Cognitive Science
• Medical Imaging

Background:

• Functional near-infrared spectroscopy (fNIRS) is increasingly utilized for cognitive assessment and bipolar disorder (BD) diagnosis.
• Research on functional connectivity in BD is limited, especially concerning task-induced changes.

Purpose of the Study:

• To differentiate patients with bipolar disorder (BD) from healthy controls (HCs) using fNIRS-derived functional connectivity.
• To investigate task-induced functional connectivity patterns during cognitive tasks.

Main Methods:

• fNIRS data analyzed from 50 BD patients and 50 HCs during resting state and two cognitive tasks (verbal fluency test [VFT] and Stroop color-word test).
• Functional connectivity quantified using coherence and correlation; network characteristics (local/global efficiency, clustering coefficient, closeness centrality) extracted.
• Group differences assessed using t-tests with false discovery rate correction.

Main Results:

• BD patients exhibited significantly altered coherence values and lower global clustering coefficients compared to HCs.
• A k-nearest neighbor (KNN) classifier utilizing coherence-based network metrics during the VFT achieved high classification performance (accuracy: 0.818).

Conclusions:

• Task-induced functional connectivity, specifically coherence metrics during the VFT, effectively distinguishes BD patients from HCs.
• Network-based approaches during cognitive tasks are valuable for characterizing functional connectivity alterations in BD.