Curry Instructions

EGI Netstation data:  Use File->Save a Copy As (with calibrated checked) which saves the data as 4 byte floating point numbers in the usual .RAW format used by segraw2mat.  These files are not pure binary so that Curry can not read them directly.  There is a new Matlab function, segraw2curry, which reads these RAW binary formatted files and writes them out as pure 4 byte floating point numbers which can be read into Curry.  It also writes to the display and to the PROCESS.LOG file the number of channels, number of samples, sample rate and number of segments which will be needed by Curry to read the file.

For example if the Netstation RAW file is binary_saveas_floating.raw then start Matlab and navigate to the directory with this file and type:

segraw2curry('binary_saveas_floating.raw');

at the prompt. The following lines are displayed and the output file is binary_saveas_floating.dat

Function segraw2curry() called with the following arguments:
Arg #1: string 'binary_saveas_floating.raw'
Current directory is /d8b/larry/CurryBinary
Number of segments: 18
Number of channels: 129
Number of samples: 600
Sample rate: 500
Writing data to file: binary_saveas_floating.dat

 

This file is now ready for Curry.  Curry recognizes BESA electrode position files (actually pair of files: .ela and .eps)  so I've put copies of these in N:\EGI_electrode_files\ForCurry egi_129.ela (labels) and egi_129.eps (digitized positions).  Start Curry and open database. Choose Edit->Insert New Subject. In Performances in Examinations, choose Edit->Choose and Insert New Performance and select the binary data file, binary_saveas_floating.dat.  Click the down arrow to copy this to Performances In Study.  Repeat these steps for the BESA electrode file, egi_129.eps (Curry will automatically use egi_129.ela).  The database window should now look similar to:

Now click the Use-> button, hide the database window, and click on Data to bring up the Data-Parameters window.  Several selections need to be made in this window in order for Curry to correctly input the data.  Change Format to Binary: Float,  choose Samples, Byte order should be Motorola (since written by Apple?), skip number of channels for now, set Number of Samples to 600, Number of Epochs/Trials to 18, Sample Time to 2, and Interstim. Interval needs to be set greater than the 1198 ms duration of each trial so that Curry doesn't treat the data a continuous.  For the BESA electrode position file set Length Unit for Coordinates to cm, Column for x: right; Column for y: Front and Column for z: Up.  In the scrolling list of sensors deselect the first three which are landmarks (Nasion, left and right preauricular points) which will lower the number of channels to 129.  Next select these landmarks in the similar list of landmarks at the bottom of the window.  The Data-Parameters window should look similar to that shown below. 

Choose Montage->Use and click Yes on the dialogs which pop up about fitting the electrodes.  The data display window should now open:

Clicking on the Meas. box will display the individual trials on the top and the average on the bottom.  For a 3D display choose 3D View to bring up the following window (which maps the Averaged Data; haven't figured out how to display individual trials):

Selecting Scale on the right will bring up the following window for selecting the time point for the contour mapping or to play movies: