Plot Channel Summary

This demo shows how to compute and visualize channel summary statistics for quality control and data characterization.

What is Channel Summary?

Channel summary computes aggregate statistics across time for each channel:

• Descriptive statistics: Mean, median, min, max, standard deviation

• Distribution metrics: Variance, kurtosis, skewness

• Range metrics: Absolute range, interquartile range

These statistics help identify problematic channels, assess data quality, and characterize recording conditions.

Available Metrics

The channel_summary() function computes:

Metric	Description
mean	Average amplitude
median	Median amplitude
std	Standard deviation
var	Variance
zvar	Z-scored variance
min	Minimum value
max	Maximum value
range	Max - Min
kurtosis	Distribution tail weight

Visualization Options

Single Metric:

plot_channel_summary(cs, :range)

Shows one statistic across all channels as a bar plot

Multiple Metrics:

plot_channel_summary(cs, [:min, :max, :std, :range])

Creates multiple subplots for comparison

Epoched Data:

plot_channel_summary(cs, :range, average_over = :epoch)

Averages statistics across epochs before plotting

Common Use Cases

1. Quality Control:

• High variance/range: Noisy or artifactual channels

• Extreme min/max: Electrode saturation or poor contact

• High z-variance: Outlier channels requiring interpolation

2. Preprocessing Validation:

• After filtering: Verify reduced variance in target frequencies

• After baseline correction: Check mean values near zero

• After rereferencing: Assess reference choice effectiveness

3. Hardware Issues:

• Systematic patterns: Poor grounding, bridging between neighboring channels

• Single channel anomalies: Loose connection, bad electrode

Interpretation

Healthy channels typically show:

• Moderate, similar variance across scalp

• No extreme outliers in min/max

• Range appropriate for recording gain settings

Problematic patterns:

Very high variance (>2-3× median):

• Likely artifact (muscle, movement, poor contact)

• Consider interpolation or exclusion

Very low variance (<0.5× median):

• May indicate flat/dead channel

• Check hardware connections

Extreme values (saturated at amplifier limits):

• Indicates clipping/saturation

• Cannot be recovered, must exclude

Workflow Summary

This demo demonstrates:

1. Load and preprocess continuous data

2. Compute summaries with flexible channel/sample selection

3. Plot single or multiple metrics for visualization

4. Apply to epoched data with epoch averaging

The channel selection examples show how to:

• Select specific channels by name

• Use pattern matching (e.g., midline channels ending in "z")

• Apply custom sample selections

Code Examples

Show Code

# Demo: Channel Summary Plots
# Shows visualization of channel-wise summary statistics.

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);

# Some minimal preprocessing (average reference and highpass filter)
EegFun.rereference!(dat, :avg)
EegFun.highpass_filter!(dat, 1)

# basic channel summary statistics
cs = EegFun.channel_summary(dat)
cs = EegFun.channel_summary(dat, channel_selection = EegFun.channels([:Fp1, :Fp2]))
cs = EegFun.channel_summary(dat, channel_selection = EegFun.channels([:Fp1, :Fp2]), sample_selection = x -> x.sample .< 2000)
cs = EegFun.channel_summary(dat, channel_selection = x -> endswith.(string.(x), "z")) # all midline channels 
cs = EegFun.channel_summary(dat, channel_selection = x -> .!(endswith.(string.(x), "z"))) # all non-midline channels 

# Plotting Channel Summaries
EegFun.plot_channel_summary(cs, :range)
EegFun.plot_channel_summary(cs, :min)
EegFun.plot_channel_summary(cs, :min, bar_color = :red)
EegFun.plot_channel_summary(cs, [:min, :max, :std, :range, :var, :zvar])

#################################
# Epoched DataFrameEeg
#################################
# some epoched data
epoch_cfg = EegFun.EpochCondition(name = "ExampleEpoch1", trigger_sequences = [[1]])
epochs = EegFun.extract_epochs(dat, epoch_cfg, (-2, 4))
cs = EegFun.channel_summary(epochs)

EegFun.plot_channel_summary(cs, :range, average_over = :epoch)
EegFun.plot_channel_summary(cs, [:min, :max, :std, :range, :var, :zvar], average_over = :epoch)

# some epoched data
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))
cs = EegFun.channel_summary(epochs)

EegFun.plot_channel_summary(cs, :range, average_over = :epoch)
EegFun.plot_channel_summary(cs, [:min, :max, :std, :range, :var, :zvar], average_over = :epoch)

See Also

• API Reference