Skip to content

Home News Data Software Models Articles Search Help
    You are not logged in Log in Feedback
You are in: home » Members » szleski's Home » Inverse CSD

Created by
Document Document
Last modified:
Sign in
User name



Inverse CSD

Home page of Szymon Łęski

Inverse Current Source Density
in three dimensions

Szymon Łęski, Daniel K. Wójcik, Joanna Tereszczuk,
Daniel A. Świejkowski, Ewa Kublik, Andrzej Wróbel

Neuroinformatics 5 (2007), 207-222
Preprint: pdf. Poster: pdf.
Matlab scripts: zip.
Experimental data: raw (over 150 MB) zip, preprocessed (~15 MB) zip.
(It is sufficient to download the preprocessed data only.)

CSDensity is a method for analysing local field potentials in neural tissue. It is used to calculate the distribution of source density from the potential data. The "inverse" method was proposed in 2006 by Klas H. Pettersen et al. We developed a three-dimensional generalization of this method.

Here is a link to a sample movie presenting the current-source density reconstructed from potentials measured on a 3-D array in the rat forebrain. The black lines are approximate boundaries of histological structures in brain. Blue signals are sources and red signals are sinks. Time goes from -10 to 20 miliseconds. The rat's whiskers are stimulated at t=0, the activation of different areas in the thalamus can be observed from t=2 ms.

Some more movies in 3D from rat sow1: t = 19ms after stimulus, whole movie. Here is a sample slide from the last movie:

You need appropriate (MPEG-4) codecs to play this movie. Under Windows you may use FFDShow.

Volume CSD data

Volume data: zip

Plain text files containing volume CSD data. The file name indicates time after stimulation and type of coordinates used (relative or absolute).

Format: each line consists of four numbers: x y z f, separated by tabs. The position x, y, z is given in milimeters (every 0.1mm), f is current source density in [nA/mm3]. Absolute coordinates are standard atlas coordinates (x: distance from the midline, y: distance from Bregma, z: depth). Relative coordinates use the same axes orientation but are shifted so that the most posterior, closest to the midline, top point is (0,0,0).  Data from each of the files form a 22 x 29 x 43 array  (this corresponds to 2.1 x 2.8 x 4.2 mm).

See also  the Neuroinformatics article, Figures 9 and 10. Note that in the article the orientation of y axis is different.

web stats