Back to the main page.

Bug 1269 - documentation: "pitfalls/procedures of source analysis; define 'virtual sensor' "

Reported 2012-01-19 19:37:00 +0100
Modified 2012-01-31 12:14:35 +0100
Product: FieldTrip
Component: documentation
Version: unspecified
Hardware: All
Operating System: All
Importance: P3 enhancement
Assigned to: Johanna
Depends on:
See also:

Johanna - 2012-01-19 19:37:36 +0100

Below I have copied and pasted from an email to the discussion list that I believe was never answered. Perhaps some of what she asks is too broad, but perhaps a few specific action points for the wiki could be made. 1) define virtual sensor, and what output from ft_sourceanalysis counts as a 'virtual sensor' 2) show more examples of things going wrong in source analysis, and how/why they happen and thus how to avoid. other suggestions? or "won't fix"? ************************************************* from Belluscio, Beth (NIH/NINDS) [E] date 7 October 2011 17:40 subject [FieldTrip] Using "virtual sensors" for data analysis Dear Fieldtrippers- In my process of learning about different methods of analyzing MEG data, I have found the different Fieldtrip tutorials extremely useful. One method that is frequently quoted in the literature is using “virtual sensors” to determine, for example, the frequency response associated with an event within a specific brain region. In trying to learn about the processing steps necessary for this method, it has been difficult for me to translate the methods quoted in the literature into specific steps within Fieldtrip. I wonder what people think about trying to construct a tutorial that would act as a guide to the procedures (and pitfalls) associated with analyzing data within source space. Beth. Beth Belluscio MD-PhD Clinical Fellow Human Motor Control Section NINDS, NIH 301-402-3495

Beth Belluscio - 2012-01-20 17:49:00 +0100

I had in mind a tutorial that would be equivalent to the "Time-frequency analysis using Hanning window, multitapers, and wavelets," or the "Source reconstruction using minimum-norm estimate." It would include both an introduction with the theoretical basis of what constitutes a "virtual sensor" and a step-by-step description of how to create and use the virtual sensor. This technique is quoted so frequently in the literature, but never with enough detail to understand exactkt what is done. And it is challenging for some of us to translate the mathematic descriptions in methods papers to a usable script. If there is a reason not to pursue such a page, that's fine. I thought I would volunteer my input if there was interest in creating one. Also, Don Rojas mentioned to me that he was interested in putting this together.

Jan-Mathijs Schoffelen - 2012-01-27 08:40:42 +0100

assigned bug to get all bugs lined up before the grand bug binge