EegFun

Appearance

Plot Artifacts ​

This demo demonstrates methods for detecting and visualizing artifacts in EEG data.

What are EEG Artifacts? ​

Artifacts are unwanted signals from non-neural sources:

Physiological:

  • Eye movements/blinks: Large amplitude deflections in frontal channels

  • Muscle activity (EMG): High-frequency noise from jaw, face, or neck

  • Cardiac signals (ECG): Heart electrical activity contamination

External:

  • Electrode issues: Poor contact, bridging, movement

  • Line noise: 50/60 Hz environmental interference

  • Movement: Low-frequency drifts and transients

Detection Methods ​

Threshold-based:

  • Absolute amplitude criteria (e.g., ±100 μV)

  • Simple and interpretable

  • Quick identification of extreme values

Statistical:

  • Z-score outlier detection

  • Variance, range, and kurtosis metrics

  • Flags unusual distributions

Visual inspection:

  • Grid-based epoch review

  • Manual validation of automatic detection

  • Refinement of rejection criteria

Visualization Approaches ​

This demo shows multiple complementary views:

Channel-wise:

  • Which electrodes are most affected

  • Spatial patterns of artifact contamination

  • EOG correlation for eye artifact identification

Epoch-wise:

  • Trial-by-trial rejection statistics

  • Temporal distribution of artifacts

  • Condition-specific artifact rates

Rejection summaries:

  • Percentage of data flagged per condition

  • Channel quality metrics

  • Decision support for repair vs. rejection

Artifact Management Strategy ​

Repair (Interpolation):

  • Use for isolated bad channels in good epochs

  • Maintains trial count

  • Cannot fix widespread contamination

Rejection:

  • Use for severely contaminated epochs

  • Removes entire trials

  • Reduces but cleans the dataset

Hybrid approach (recommended):

  1. Repair isolated channel issues

  2. Reject epochs with widespread artifacts

Workflow Summary ​

This demo shows:

  1. Loading data: Continuous or epoched EEG

  2. Detection: Applying threshold and statistical criteria

  3. Visualization: Plotting detected artifacts across channels and epochs

  4. Assessment: Evaluating artifact patterns to guide cleaning strategy

  5. Reporting: Quantifying data attrition for methods documentation

Critical Principle ​

Garbage in, garbage out: No computational method can rescue poor quality data. The most effective strategy is careful data collection with proper electrode preparation and participant instruction. Artifact management is a secondary cleaning strategy, not a substitute for quality acquisition.

Code Examples ​

Show Code
julia
# Demo: Artifact Detection Visualization
# Shows plots for visualizing artifact detection and rejection.

using EegFun
# Note: EegFun.example_path() resolves bundled example data paths.
# When using your own data, simply pass the file path directly, e.g.:
# dat = EegFun.read_raw_data("/path/to/your/data.bdf")

# read raw data
dat = EegFun.read_raw_data(EegFun.example_path("data/bdf/example1.bdf"));
dat = EegFun.create_eegfun_data(dat)

# minimal preprocessing
EegFun.rereference!(dat, :avg)
EegFun.highpass_filter!(dat, 1)

# extract epochs
epoch_cfg = EegFun.EpochCondition(name = "ExampleEpoch1", trigger_sequences = [[1]])
epochs = EegFun.extract_epochs(dat, epoch_cfg, (-2, 4))

# Artifacts
artifacts = EegFun.detect_bad_epochs_automatic(epochs)

# Plot artifacts
EegFun.plot_artifact_detection(epochs, artifacts)

# Or with custom criteria
artifacts = EegFun.detect_bad_epochs_automatic(epochs, abs_criterion = 100, z_criterion = 0)
EegFun.plot_artifact_detection(epochs, artifacts)

artifacts = EegFun.detect_bad_epochs_automatic(epochs, abs_criterion = 0, z_criterion = 3)
EegFun.plot_artifact_detection(epochs, artifacts)

# extract multiple epochs
epoch_cfg = [
    EegFun.EpochCondition(name = "ExampleEpoch1", trigger_sequences = [[1]]),
    EegFun.EpochCondition(name = "ExampleEpoch2", trigger_sequences = [[2]]),
]
epochs = EegFun.extract_epochs(dat, epoch_cfg, (-2, 4))

# Artifacts
artifacts = EegFun.detect_bad_epochs_automatic(epochs)

# Plot artifacts
EegFun.plot_artifact_detection(epochs, artifacts)
EegFun.plot_artifact_detection(epochs[1], artifacts[1])


# Interactive artifact detection
EegFun.detect_bad_epochs_interactive(epochs, dims = (4, 4))

# Or combine with automatic artifact detection
artifacts = EegFun.detect_bad_epochs_automatic(epochs)
EegFun.detect_bad_epochs_interactive(epochs, artifact_info = artifacts, dims = (4, 4))

# Or combine with automatic artifact detection
EegFun.detect_bad_epochs_interactive(epochs, artifact_info = artifacts, dims = (4, 4), colormap = :GnBu)

See Also ​