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.
Tags: variability, signal-complexity, epilepsy, seizure-detection, eeg, io:scalar-per-channel
Max¶
Compute the maximum value across a dimension.
Mean¶
Compute the mean value across a dimension.
Min¶
Compute the minimum value across a dimension.
SpikeCount¶
Calculate spikes in the input data using the IQR method.
Tags: outlier-detection, artifact, epilepsy, spike-detection, eeg, io:scalar-per-channel