Multimodal EEG-fNIRS classification as a clinical tool for bipolar disorder diagnosis

Area of Science:

• Neuroscience
• Psychiatry
• Biomedical Engineering

Background:

• Bipolar disorder (BD) diagnosis is challenging due to overlapping symptoms and subjective assessments.
• Emotional dysregulation and cognitive dysfunction are key features of BD.
• Differentiating BD subtypes requires objective diagnostic markers.

Purpose of the Study:

• To investigate a multimodal approach using EEG and fNIRS for BD diagnosis and subtype differentiation.
• To identify neural and vascular patterns associated with emotional dysregulation in BD.
• To assess the feasibility of a simplified, portable EEG-fNIRS system for BD assessment.

Main Methods:

• An emotional visual task was used to assess cognitive function and emotional interference.
• Whole-head electroencephalography (EEG) and frontal functional near-infrared spectroscopy (fNIRS) were employed.
• Data from bipolar disorder patients and healthy controls were analyzed for group classification.

Main Results:

• Behavioral analysis showed significant performance differences between BD patients and controls.
• Integrating EEG and fNIRS improved classification accuracy compared to EEG alone.
• Analysis using frontal regions with fNIRS integration demonstrated robust classification, supporting a simplified system.

Conclusions:

• EEG and fNIRS provide complementary insights into neural and vascular markers of BD.
• Multimodal neuroimaging significantly enhances the accuracy of BD diagnosis and subtype differentiation.
• The findings support the development of portable, multimodal diagnostic tools for bipolar disorder.