Fiducial points

Discussions
1

Hello,

Our group has been using BS for animal studies.
I realized that fiducial points are required to insert when MRI is loaded in the software.
I read the tutorials about the coordinate system, and it seems that fiducial points are used to create the SCS coordinate systems.
I wonder the fiducials are used any other purposes in the BS software except for the visualization purpose.
I wonder how critical having accurate fiducial points are.
I look forward to your response.
Thank you.

Regards,
Jihye Bae

2

Hi Jihye,

The points NAS/LPA/RPA have to be defined in the MRI because it defines the Brainstorm coordinate system. All the spatial information imported after (surfaces, electrode positions) are converted to this coordinates system.
To align correctly the position of the electrodes and the MRI, you have to mark the same points in the MRI Viewer and in the electrodes file.
Except for that, they can be anywhere you want, it’s not changing anything but the reference system. Just do not change them after you start importing data.
http://neuroimage.usc.edu/brainstorm/CoordinateSystems

The points AC/PC/IH are used for display purposes (for aligning the MRI slices in the 3D MRI view, not important at all), and for the subject co-registration (not relevant if you are not projecting the source values from one anatomy to another).
It doesn’t matter if those points are very approximate…

Francois

3

[QUOTE=Francois;7778]Just do not change them after you start importing data.[/QUOTE]

Hello François,
What happens if we modify them after having imported data and computed sources?
I realized after the source reconstruction that two subjects had bad RPA and LPA. Should I reset these subjects and start all over again ?

Best,
Louis

4

If you modify the fiducial points after the computation of the forward model, the surfaces are realigned in a new coordinate system, but not the source values.
You will still be able to visualize them but they would not be fixed. Therefore modifying the position of those points without recalculating the forward and inverse models is useless…

So yes, I would recommend you delete the entire subject (including all the surfaces and MRI) and start over, to make sure your results are correct.