Example: Basic EEG Preprocessing Pipeline (MNE-Python)¶

This page explains the basic_preprocessing_pipeline_mne.signalJourney.json example file, which documents a standard EEG preprocessing workflow using MNE-Python.

Pipeline Overview¶

The MNE-Python basic preprocessing pipeline demonstrates fundamental EEG preprocessing steps using MNE-Python functions:

• Load raw data from FIF format
• Apply band-pass filtering (1-40 Hz) using raw.filter
• Apply notch filtering (60 Hz) using raw.notch_filter
• Set average reference using raw.set_eeg_reference
• Interpolate bad channels using raw.interpolate_bads

Pipeline Flowchart¶

flowchart TD
    A[Load Raw Data
mne.io.read_raw_fif] --> B[Apply Band-pass Filter
raw.filter
1-40 Hz]
    B --> C[Apply Notch Filter
raw.notch_filter
60 Hz]
    C --> D[Set Average Reference
raw.set_eeg_reference]
    D --> E[Interpolate Bad Channels
raw.interpolate_bads]

    %% Input file
    F["📁 sub-01_task-rest_raw.fif
Raw EEG data"] --> A

    %% Final output
    E --> G["💾 sub-01_task-rest_desc-preproc_eeg.fif
Preprocessed data"]

    %% Quality metrics
    D --> Q1["📈 Projection added: true"]
    E --> Q2["📈 Channels interpolated: [EEG 053, EEG 021]"]

    %% 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 qualityMetric fill:#f9f9f9,stroke:#666,stroke-width:1px

    class A,B,C,D,E processStep
    class F inputFile
    class G outputFile
    class Q1,Q2 qualityMetric

Key MNE-Python Features Demonstrated¶

MNE-Python Function Calls¶

• mne.io.read_raw_fif: Load FIF format files with preloading
• raw.filter: FIR filtering with linear phase response
• raw.notch_filter: Notch filtering for line noise removal
• raw.set_eeg_reference: EEG referencing with projection
• raw.interpolate_bads: Spherical spline interpolation

MNE-Python-Specific Parameters¶

• Filter specifications: FIR design with 'firwin' method
• Reference handling: Projection-based average referencing
• Channel interpolation: Accurate mode with automatic bad channel reset

Example JSON Structure¶

The signalJourney file documents each processing step with:

{
  "stepId": "2",
  "name": "Apply Band-pass Filter",
  "description": "Apply a FIR band-pass filter (1-40 Hz).",
  "software": {
    "name": "MNE-Python",
    "version": "1.6.1",
    "functionCall": "raw.filter(l_freq=1.0, h_freq=40.0, fir_design='firwin')"
  },
  "parameters": {
    "l_freq": 1.0,
    "h_freq": 40.0,
    "method": "fir",
    "fir_design": "firwin",
    "phase": "zero"
  }
}

Quality Control Integration¶

Each step includes quality metrics specific to MNE-Python processing: - Projection status for referencing - Interpolated channel tracking - BIDS-compatible file naming

MNE-Python vs EEGLAB Comparison¶

Usage Notes¶

This example demonstrates: - MNE-Python workflow patterns with object-oriented design - Parameter documentation for reproducible processing - BIDS compatibility with standardized naming conventions - Quality metrics relevant to MNE-Python processing

The pipeline serves as a foundation for more complex analysis workflows including ICA decomposition, time-frequency analysis, and source localization.