Parameters

The MEAlytics package contains a wide range of parameters that can be used to alter the analysis pipeline. Described below are all the parameters, and how they affect the analysis pipeline.

The default parameter values are stored as a dictionary in the library, and can be accessed as follows:

from CureQ.mea import get_default_parameters
print(get_default_parameters())

Default parameters are automatically loaded into the MEAlytics GUI

Bandpass filter

Low cutoff

Low cutoff frequency for the Butterworth bandpass filter.
Default value: 200 Hz

High cutoff

High cutoff frequency for the Butterworth bandpass filter.
Default value: 3500 Hz

Order

The filter order used for the Butterworth bandpass filter.
Default value: 2

Threshold

Threshold portion

Portion of the data that is used to calculate the threshold value. Higher values can give a more accurate estimate of the background noise level, but takes longer to compute.
Default value: 0.1

Standard deviation multiplier

Value that is multiplied with the standard deviation of a portion of the data to create a threshold that determines whether this portion of the data contains purely noise or potential spikes.
Default value: 5

RMS multiplier

Value that is multiplied with the root mean square (RMS) of the background noise to determine the threshold for the spike detection.
Default value: 5

Spike detection

Refractory period

The refractory period of the spike. The time in which only one spike can be detected. Value should be given in seconds.
Default value: 0.001

Spike validation

Spike validation method

The spike validation method used to determine whether a threshold crossing should be considered a neuronal signal or discarded.
Possible options: Noisebased, none.
Default value: Noisebased

Exit time

The time the signal must drop/rise below a certain amplitude before/after it has reached its absolute peak. Value should be given in seconds.
Default value: 0.001

Drop amplitude

Multiplied with the surrounding background noise of the spike to determine the amplitude the signal must drop.
Default value: 5

Max drop

Value multiplied with the threshold value of the electrode to determine the maximum amplitude the signal can be required to drop.
Default value: 2

Burst detection

Minimal amount of spikes

The smallest amount of spikes that can form a single channel burst.
Default value: 5

Default interval threshold

The default inter-spike interval (ISI) threshold used for burst detection. Value should be given in milliseconds.
Default value: 100

Max interval threshold

The maximum ISI threshold that can be used for burst detection. Values should be given in milliseconds.
Default value: 1000

KDE bandwidth

The bandwidth of the Kernel Density Estimate (KDE) used for determining the burst detection parameters.
Default value: 1

Network burst detection

Min channels

The minimal amount of channels that should be participating in a network burst to consider it as one. Value should be given as a percentage.
Default value: 0.5

Thresholding method

The method used to automatically calculate the threshold used for determining high-activity periods.
Possible values: Yen, Otsu, Li, Isodata, Mean, Minimum, Triangle. See scikit-image filters.
Default value: Yen

KDE bandwidth

The bandwidth used for the kernel density estimate representing spike activity.
Default value: 0.05