Difference between revisions of "AnyWave:Plugin EI"
(add how to cite section in EI plugin) |
|||
Line 13: | Line 13: | ||
<br /> | <br /> | ||
<big>[[AnyWave:Plugins_MCR|See the tutorial section about MATLAB plugins to see how to install the required packages.]]</big> | <big>[[AnyWave:Plugins_MCR|See the tutorial section about MATLAB plugins to see how to install the required packages.]]</big> | ||
− | |||
− | |||
− | |||
− | |||
− | |||
== What does it do? == | == What does it do? == |
Revision as of 15:04, 1 March 2021
Contents
Epileptogenicity Index
How to cite
To cite the core method of the epileptogenicity index, please use:
Fabrice Bartolomei, Patrick Chauvel, Fabrice Wendling, Epileptogenicity of brain structures in human temporal lobe epilepsy: a quantified study from intracerebral EEG, Brain, Volume 131, Issue 7, July 2008, Pages 1818–1830, https://doi.org/10.1093/brain/awn111
To cite the AnyWave plugin, please use:
Roehri, N., Pizzo, F., Lagarde, S., Lambert, I., Nica, A., McGonigal, A., Giusiano, B., Bartolomei, F. and Bénar, C.‐G. (2018), High‐frequency oscillations are not better biomarkers of epileptogenic tissues than spikes. Ann Neurol., 83: 84-97. doi:10.1002/ana.25124
IMPORTANT NOTE
This is a MATLAB Compiled plug-in which requires the MATLAB Runtime 2018b to be installed on the computer.
See the tutorial section about MATLAB plugins to see how to install the required packages.
What does it do?
The Epileptogenicity Index plugin is a MATLAB based algorithm which quantifies the degree of involment of brain structures according to their frequency content and their onset timing (Bartolomei et al., 2008). Even if it is a MATLAB based program, we provide a compiled version of the plugin so that MATLAB is not needed on your computer.
Tutorial
Before running AnyWave, make sure that the MATLAB runtime is installed (otherwise get it from here) and that the Epileptogenicity Index (EI) Plugin is correctly located in the AnyWave folder (MyDocuments\AnyWave\Plugins\MATLAB). The folder should contain at least these files files: desc.txt, EI.exe, and splash.png.
Launch AnyWave and open the file containing a seizure. Use the marking tool to select a section of the file. Name it EI. Select the channels you want to study – Ctrl+Left Click – they should turn red, otherwise the process will be applied to all the channels. Then launch the process by clicking: Processes:MATLAB:Epileptogenicity Index.
The Graphical User Interface (GUI) opens. It is divided into 6 sections: the signal window 1, the Energy Ratio (ER) Parameters panel 2, the Detection Parameters panel 3, the Epileptogenicity Index Panel 4, the Display Parameters panel 5, and the Export panel 6.
The signal window 1 displays the traces in red lines, the ER maps and the Un curves.
The Energy Ratio Parameters panel (2) control the first step of the EI computation. The ER corresponds either to the energy ratio of the high frequencies over the low frequencies (default) or the ratio of the high frequencies over the whole frequencies. This can be set using the radio button on this panel. The frequency bands are tuned using the editable boxes. Min Low and Min High correspond to the lower limit [in Hz] of the low and high frequency bands, respectively and Max Low and Max High to the upper limit [in Hz] of the low and high frequency bands, respectively. The Window Size corresponds to the time-duration [in number of samples] over which the ER is calculated. The Step is the amount of time [in number of sample] the previous window is shifted.
The Detection Parameters panel (3) handles the Page-Hinkley parameters. The algorithm is described in Bartolomei et al., 2008. The Bias controls the slope of the Un curve. It has to be high enough for the Un curve to be decreasing. The Threshold corresponds to the value for which the Un curve has to increase to be marked as a significant change in the frequency structure. The Decay controls how the energy has to be weighted during the EI computation step.
The Epileptogenicity Index panel (4) handles the last step. Here the Window Size box correspond to the duration of the time window [in sample] in which the ER is summed. When all the parameters are set, you can press the Compute button. A wait bar will pop up. At the end, the results will be plotted in the signal window 1 . The cyan crosses represent the start of the variation (possibly the seizure) in the channel and the green circles represent the moment when the Un curve reaches the threshold. You can manually adjust a detection by clinking firstly on cursor tool, then on the channel at the moment you want the detection to be and finally on User Marking. For multiple channel adjustment, you simply have to press Shift while clicking.
On the Display Parameters panel (5), you can check or uncheck elements to be displayed. Once your choice is made, click on the plot button. The red and cyan lines referred to the Signal and the Un respectively. The Gain slider enables you to adjust the gain of the ER map. You can change the number of channel you want to display by changing the number in the Num box. Several colormaps are available in the Colormap pop-up menu. To display the EI, push the Display EI button. Another window will pop up and display a bar graph representing the normalized EI and the EI before time weighting.
You have several possibility to export the results (6). Click on the To AnyWave button to send the markers to AnyWave; on the To Marker File button to save the result in an AnyWave marker file, and finally on the To Excel button to generate an Excel file containig the results and the figures of the EI graph and the ER map (if wanted). For the Mac and Linux version, an .csv file will be created instead.
Moreover, you can zoom in or out of a zone using these tools and move the window using this one