Example: Connectivity Analysis (EEGLAB)¶

This page explains the connectivity_analysis_pipeline_eeglab.signalJourney.json example file, which documents spectral connectivity analysis using EEGLAB and MATLAB functions.

Pipeline Overview¶

This EEGLAB pipeline demonstrates connectivity analysis for examining relationships between EEG channels: - Load cleaned data from ICA decomposition pipeline
- Extract epochs for connectivity analysis - Compute power spectral density using EEGLAB's spectopo - Calculate coherence matrix using MATLAB Signal Processing Toolbox - Generate connectivity report with network visualizations

Pipeline Flowchart¶

flowchart TD
    A[Load Cleaned Data
pop_loadset] --> B[Extract Epochs
pop_epoch]
    B --> C[Compute PSD
spectopo]
    C --> D[Calculate Coherence
mscohere]
    D --> E[Generate Network Plot
topoplot_connect]
    E --> F[Save Results
save]

    %% Input file
    G["📁 sub-01_task-rest_desc-cleaned_eeg.set
From: ICA Decomposition Pipeline"] --> A

    %% Inline data
    B --> V1["📊 Epoch Windows
[-1.0, 2.0] s"]
    C --> V2["📊 Frequency Range
Alpha: 8-12 Hz"]
    D --> V3["📊 Coherence Method
MATLAB mscohere"]

    %% Final outputs
    F --> H["💾 sub-01_task-rest_desc-connectivity_eeg.mat
Connectivity results"]
    E --> I["💾 sub-01_task-rest_desc-network_plot.fig
Network visualization"]
    E --> J["💾 sub-01_task-rest_desc-connectivity_matrix.png
Matrix plot"]

    %% Quality metrics
    C --> Q1["📈 Frequency bins: 256
Spectral resolution: 0.25 Hz"]
    D --> Q2["📈 Significant pairs: 312/2016
Max coherence: 0.78"]

    %% Styling
    classDef processStep fill:#e1f5fe,stroke:#01579b,stroke-width:2px
    classDef inputFile fill:#fff3e0,stroke:#e65100,stroke-width:2px
    classDef outputFile fill:#e8f5e8,stroke:#1b5e20,stroke-width:2px
    classDef inlineData fill:#f3e5f5,stroke:#4a148c,stroke-width:1px
    classDef qualityMetric fill:#f9f9f9,stroke:#666,stroke-width:1px

    class A,B,C,D,E,F processStep
    class G inputFile
    class H,I,J outputFile
    class V1,V2,V3 inlineData
    class Q1,Q2 qualityMetric

Key EEGLAB Features Demonstrated¶

EEGLAB Connectivity Functions¶

• pop_epoch: Extract event-related epochs from continuous data
• spectopo: Power spectral density computation with multitaper method
• mscohere: MATLAB coherence calculation between channel pairs
• topoplot_connect: Network visualization on scalp topography
• MATLAB integration: Leveraging Signal Processing Toolbox functions

EEGLAB-Specific Parameters¶

• Epoch extraction: EEGLAB event-based epoching with GUI support
• Spectral analysis: Integrated with EEGLAB channel structure
• Visualization: EEGLAB topographic plotting with connectivity overlays
• File formats: MATLAB .mat files for connectivity matrices

Example JSON Structure¶

The EEGLAB coherence computation demonstrates MATLAB integration:

{
  "stepId": "4",
  "name": "Calculate Coherence Matrix",
  "description": "Compute magnitude-squared coherence between all channel pairs using MATLAB mscohere.",
  "software": {
    "name": "MATLAB",
    "version": "R2023a",
    "functionCall": "for i=1:64; for j=1:64; [coh,f] = mscohere(epochs(i,:), epochs(j,:), window, noverlap, nfft, srate); end; end"
  },
  "parameters": {
    "window": "hann(128)",
    "noverlap": 64,
    "nfft": 256,
    "srate": 500,
    "freq_range": [8, 12]
  }
}

EEGLAB Dataset Integration¶

EEGLAB connectivity analysis leverages the EEG structure:

{
  "stepId": "3",
  "name": "Compute Power Spectral Density",
  "software": {
    "name": "EEGLAB",
    "version": "2023.1",
    "functionCall": "[spectra,freqs] = spectopo(EEG.data, 0, EEG.srate, 'freqrange', [1 40], 'electrodes', 'on')"
  },
  "parameters": {
    "freqrange": [1, 40],
    "electrodes": "on",
    "overlap": 50,
    "nfft": 256,
    "winsize": 128
  }
}

EEGLAB Connectivity Features¶

MATLAB Signal Processing Integration¶

• mscohere function: Magnitude-squared coherence with Welch's method
• Cross-spectral density: Full spectral analysis capabilities
• Windowing options: Hann, Hamming, Bartlett windows
• Frequency resolution: Configurable FFT parameters

EEGLAB Visualization Tools¶

• topoplot integration: Connectivity overlaid on channel locations
• Channel location support: 3D electrode positions for accurate plotting
• Network graphs: Node-edge representations of connectivity
• Matrix visualization: Heatmaps and connectivity matrices

Quality Control Features¶

• Spectral validation: Power spectral density verification
• Coherence thresholds: Statistical significance testing
• Channel quality: Bad channel identification and exclusion
• Epoch rejection: Artifact-contaminated epoch removal

EEGLAB vs MNE-Python Comparison¶

EEGLAB-Specific Workflow¶

MATLAB Ecosystem Integration¶

EEGLAB connectivity analysis benefits from: 1. Signal Processing Toolbox: Professional-grade spectral analysis functions 2. Parallel Computing: Multi-core coherence computation 3. Visualization Tools: Advanced plotting and 3D visualization 4. Statistical Toolbox: Comprehensive statistical testing capabilities

EEG Structure Preservation¶

• Channel information: Electrode locations automatically used
• Event markers: Epoch extraction based on EEG.event structure
• Sampling rate: Automatically extracted from EEG.srate
• Data history: Processing steps recorded in EEG.history

Interactive Analysis¶

• GUI integration: Pop-up functions for parameter selection
• Visual feedback: Real-time plotting during analysis
• Manual adjustment: Interactive parameter tuning
• Batch processing: Automated analysis across multiple datasets

Usage Notes¶

This example demonstrates: - EEGLAB connectivity workflows using MATLAB integration - Spectral analysis integration with EEGLAB functions - Multi-format outputs for matrices and visualizations - Quality control with spectral validation - MATLAB ecosystem leveraging Signal Processing Toolbox

The pipeline showcases EEGLAB's connectivity analysis capabilities through MATLAB integration while maintaining comprehensive parameter documentation for reproducible network analysis. The combination of EEGLAB's EEG-specific tools with MATLAB's signal processing functions provides a powerful framework for connectivity research.