Skip to content

Signal Statistics

What are the basic properties of my signal?

Features in the cobrabox.signalstats domain. Access them as cb.<Feature> (canonical), or as cb.signalstats.<Feature> / cb.feature.<Feature>.

Each feature has two forms: a class for building pipelines, and a one-shot function for a single call.

AmplitudeVariation

Compute amplitude variation over the time dimension.

cb.amplitude_variation(data, ...)  # one-shot
cb.AmplitudeVariation(...).apply(data)  # composable, for pipelines

Tags: variability, standard-deviation, anesthesia, eeg, io:scalar-per-channel

Autocorrelation

Compute normalized autocorrelation at a given lag along a required dimension.

cb.autocorrelation(data, ...)  # one-shot
cb.Autocorrelation(...).apply(data)  # composable, for pipelines

Tags: temporal-dependence, lag, stationarity, eeg, fmri, io:scalar-per-channel

EpileptogenicityIndex

Compute the Epileptogenicity Index (EI) per channel (Bartolomei et al., 2008).

cb.epileptogenicity_index(data, ...)  # one-shot
cb.EpileptogenicityIndex(...).apply(data)  # composable, for pipelines

Tags: epilepsy, seizure-onset-zone, energy-ratio, beta-gamma, page-hinkley, intracranial-eeg, eeg, io:scalar-per-channel

LineLength

Compute line length over the time dimension.

cb.line_length(data, ...)  # one-shot
cb.LineLength(...).apply(data)  # composable, for pipelines

Tags: variability, signal-complexity, epilepsy, seizure-detection, eeg, io:scalar-per-channel

Max

Compute the maximum value across a dimension.

cb.max(data, ...)  # one-shot
cb.Max(...).apply(data)  # composable, for pipelines

Tags: reduction, io:scalar

Mean

Compute the mean value across a dimension.

cb.mean(data, ...)  # one-shot
cb.Mean(...).apply(data)  # composable, for pipelines

Tags: reduction, io:scalar

Min

Compute the minimum value across a dimension.

cb.min(data, ...)  # one-shot
cb.Min(...).apply(data)  # composable, for pipelines

Tags: reduction, io:scalar

SpikeCount

Calculate spikes in the input data using the IQR method.

cb.spike_count(data, ...)  # one-shot
cb.SpikeCount(...).apply(data)  # composable, for pipelines

Tags: outlier-detection, artifact, epilepsy, spike-detection, eeg, io:scalar-per-channel