Differential diagnosis of bipolar II disorder and major depressive disorder: Integrating multimodal approaches to overcome clinical challenges

Table 1 Key biomarkers in differentiating bipolar II disorder from major depressive disorder

Biomarker/category	MDD	BD-II	Ref.
Inflammatory cytokines: IL-6, TNF-α	Lower or normal compared to BD	Generally elevated compared to MDD	[19]
Neurotransmitter metabolites: Serotonin, norepinephrine	Levels relatively stable or altered without clear episodic pattern	Fluctuations associated with mood episodes	[20]
Neutrophil-to-lymphocyte ratio	Lower than BD and close to healthy controls	Significantly elevated; correlated with frequency and severity of depressive episodes	[21,22]
Anti-inflammatory cytokine: IL-10	Consistently reduced; linked to impaired suppression of inflammation	Marked variability: Lower during depressive episodes, higher or normalized during hypomanic episodes	[23,24]
Inflammatory cell counts	Leukocyte and NLR within normal or slightly elevated range	Leukocyte and NLR significantly elevated	[25]
Oxidative stress markers: ROS, antioxidant enzymes (SOD, CAT)	ROS low/normal; antioxidant enzyme activity preserved or slightly elevated	ROS elevated; antioxidant enzyme activity reduced	[26]
Cortisol levels	Less pronounced fluctuation; moderate increase under stress	Greater fluctuation; significantly elevated during depressive episodes; distinct pattern from other depression types	[27-29]

MDD: Major depressive disorder; BD-II: Bipolar II disorder; ROS: Reactive oxygen species; NLR: Neutrophil-to-lymphocyte ratio; TNF-α: Tumor necrosis factor-α; IL-6: Interleukin-6; IL-10: Interleukin-10; SOD: Superoxide dismutase; CAT: Catalase; BD: Bipolar disorder.

Table 2 Comparative electroencephalography features in bipolar II disorder vs major depressive disorder

Feature category	BD-II characteristics	MDD characteristics	Clinical implication	Ref.
Power Spectrum (Alpha Band)	Resting-state alpha power close to healthy controls; no significant reduction	Significant alpha power reduction in centro-parietal regions; negatively correlated with symptom severity	Alpha power reduction may indicate MDD	[43,45]
Power Spectrum (Theta Band)	Reduced spontaneous phase variability in frontal theta	Reduced spontaneous phase variability in frontal theta	Theta reduction is a shared feature	[43]
Phase Dynamics (Alpha Band)	No significant alpha phase delay; stable phase variability	Pronounced alpha phase-cycle delay; reduced phase variability	Alpha phase delay may be MDD-specific	[43,49]
Mood-State Dependency	EEG patterns close to healthy controls during euthymic state; altered patterns during mood episodes	Abnormal EEG patterns even at rest	Reflects state-related differences between disorders	[43,45]
Machine Learning Classification	EEG features can be classified by SVM/KNN with approximately 93% accuracy	Same as BD-II	Enhances diagnostic accuracy	[50]

KNN: K-Nearest Neighbors; MDD: Major depressive disorder; EEG: Electroencephalography; SVM: Support vector machines; BD-II: Bipolar II disorder.

Table 3 Characteristics and application scenarios of different machine learning algorithms in bipolar II disorder vs major depressive disorder differential diagnosis

Algorithm type	Advantages	Limitations	Application scenarios	Example applications	Ref.
Support Vector Machine	Performs well on small, high-dimensional datasets; clear decision boundaries	Sensitive to parameter tuning; high computational cost for large datasets	Gene expression analysis; language pattern recognition	Distinguishing emotional text features between BD-II and MDD	[51,54,55]
Random Forest	Strong robustness; handles nonlinear relationships; provides feature importance ranking	May overfit small datasets	Multi-modal feature integration; questionnaire + imaging data	Gene combination screening; questionnaire-based classification	[52,53]
Deep Learning (CNN, RNN)	Automatically extracts complex features; suitable for image and text data	Requires large datasets and high computational power; low interpretability	Facial expression analysis; social media data	Micro-expression recognition; emotion classification	[9,52,55,56]
Linear Discriminant Analysis	Simple and efficient; high interpretability	Limited by linear assumptions; unsuitable for highly nonlinear data	Emotional vocabulary frequency analysis	Vocabulary-based emotion classification	[54,55]
Polygenic Risk Score	Integrates genetic information; enables personalized risk prediction	Depends on large-scale genomic data	Early risk prediction; pediatric cohort studies	Predicting future risk of BD development	[10]
Hybrid Models	Combines strengths of multiple algorithms; improved performance	Complex implementation; requires coordination between models	High-dimensional, multi-modal data analysis	SVM + RF combined diagnosis	[55,57,58]

MDD: Major depressive disorder; SVM: Support vector machines; BD-II: Bipolar II disorder; BD: Bipolar disorder; RF: Random forest.