Back to the main page.

Bug 1403 - unit ERP?

Reported 2012-03-31 21:43:00 +0200
Modified 2012-08-23 10:33:47 +0200
Product: FieldTrip
Component: documentation
Version: unspecified
Hardware: PC
Operating System: Windows
Importance: P3 normal
Assigned to: Robert Oostenveld
Depends on:
See also:

Lilla Magyari - 2012-03-31 21:43:57 +0200

Hi all, I am analyzing EEG data with FieldTrip and I have compared the grandaverage ERP waves in FT to waves I got with Brain Vision Analyzer. I got different scales (or units). The units of the BVA are in microV. I am wondering what is the unit of the ERP in FieldTrip (if not microV). And it would be nice to document this. (I could not find anything about this in the tutorial). Lilla

Robert Oostenveld - 2012-04-01 09:32:47 +0200

the units after timelockanalysis should be identical as the units in which the data was read (ft_read_data or ft_preprocessing). For BVA I had expected that to be uV. There is ongoing work to get the units represented in the data. The ft_read_header already returns hdr.units (and hdr.chantype) but for most data those are still unknown. Can you save the ERP from BVA in a file, and can that be read in fieldtrip? If so, then you could specify the uV value according to BVA (by looking in the BVA figures) and we can check in FT that we get the same value. So if possible, please make an ERP in BVA and write that to file for testing.

Lilla Magyari - 2012-04-01 20:02:57 +0200

(In reply to comment #1) Maybe the problem is than that I use a script that Marcel wrote for reading in BVA data. I will try to read it in also with the appropriate functions.

Lilla Magyari - 2012-04-10 19:28:21 +0200

Created attachment 250 ch31 plotted with FieldTrip

Lilla Magyari - 2012-04-10 19:29:09 +0200

Created attachment 251 ch31 plotted with BVA y-axis: -50 50 uV

Lilla Magyari - 2012-04-10 19:36:44 +0200

(In reply to comment #1) I tried ft_read_header but cfg.chanunit says: "unknown". You can find the data exported from BVA in /home/language/lilmag/bug1403_erpunit. I read the data in: cfg=[]; cfg.headerfile = 'PP1_SEM_MATCH_Average_AUT2.hdr'; data = ft_preprocessing(cfg,data); Then, plotted: cfg=[]; cfg.baseline = [-0.2 0]; cfg.layout = '61chan_MPI.lay'; cfg.interactive = 'yes'; ft_multiplotER(cfg, data); The range of the y-axis on the BVA image is between - 50 and 50 uV, while the y-axis in FieldTrip is around min. -35. thanks for help! Lilla

Lilla Magyari - 2012-04-10 19:37:44 +0200

Created attachment 252 layout file

Robert Oostenveld - 2012-04-11 09:39:53 +0200

(In reply to comment #6) I copied the data and the layout file to /home/common/matlab/fieldtrip/data/test/bug1403

Robert Oostenveld - 2012-04-11 09:49:05 +0200

(In reply to comment #7) I have added fieldtrip/test/test_bug1403, which contains the following cd /home/common/matlab/fieldtrip/data/test/bug1403 cfg=[]; cfg.dataset = 'LauraPP1_SEM_MATCH_Average_AUT2.vhdr'; data = ft_preprocessing(cfg); cfg=[]; cfg.layout = '61chan_MPI.lay'; cfg.interactive = 'yes'; figure ft_multiplotER(cfg, data); figure plot(data.time{1}, data.trial{1}) legend(data.label) It produces a figure with positive deflections at 450ms with a max positive amplitude of 35. I presume this is uV. Can you show a BVA figure?

Lilla Magyari - 2012-04-11 13:23:14 +0200

the pdf attached is the BVA picture about channel 31 (Sorry, earlier the description of the file was wrong). The y axis there has a min of -50 and max 50. Unfortunately, you can't see the numbers of the y axis on the pdf. So, I attached also a .ppt with ch31 from BVA.

Lilla Magyari - 2012-04-11 13:24:01 +0200

Created attachment 255 ch31 plotted with BVA (ppt file)

Robert Oostenveld - 2012-04-11 17:00:46 +0200

the length of the vertical bar is 50 uV in total. If you draw in the ppt a zero line, and have one of these bars stick out to the top and one to the bottom you get a better idea of the peak amplitude. See attached. To me it seems all ok.

Robert Oostenveld - 2012-04-11 17:01:27 +0200

Created attachment 256 powerpoint edit

Lilla Magyari - 2012-04-12 12:09:42 +0200

(In reply to comment #12) ok, I am sorry then! I understood the BVA y-axis that it runs from -50 to 50 but I should understand then that it runs from -25 to 25. Thanks for the help!

Robert Oostenveld - 2012-04-12 12:17:33 +0200

(In reply to comment #13) so it can be closed

Robert Oostenveld - 2012-08-23 10:33:47 +0200

closed multiple bugs that have been resolved for some time