Multichannel Systems LTP-Director Manual - Manual (Page 77)

73
8 Signal Detection
8.1 Signal Detection
All input signals are filtered by a Savitzky Golay filter. The Savitzky Golay method essentially
performs a local polynomial regression on a series of values to determine the smoothed value
for each point. The purpose is to smooth a noisy signal without greatly distorting the signal.
Please set the parameters of the Savitzky Golay filter in main menu "Settings", "Filter".
To analyze the data of each sweep it is necessary to detect the evoked response of the slice.
In order to separate the signal from the stimulation artifact, the detection is performed within
a user adjustable region of interest (ROI).
The detection algorithm finds the signal via detection of local minima and maxima and applies
a three point pattern to detect population spikes (PS) from the soma of a neuron, and excitatory
post synaptic spikes (EPSP) from the dendrite of a neuron. The detected maxima and minima are
shown as dots in the signal trace. It is important for the signal detection to work properly with
the ROI. The region of interest has to be set left and right of the response signal, otherwise the
three point pattern can not be used.
The picture shows a typical signal of a somatic evoked response. The ROI settings are correct
for detecting the pop spike (PS) via three point pattern.
In this example the ROI settings are not correct for spike detection via three point pattern.
The right ROI cursor is moved too much left, and the third maximum peak of the dendritic
EPSP is not detected anymore.