BOLD signal dropout - Dura abnormalities?

Software Support
, ,
1

Hi all,

I’ve been noticing abnormalities in EPI BOLD rsfMRI in some of my old subjects, which appear as relatively large black holes resembling signal dropout. Upon further inspection of their structural scans, I noticed a bizarre dura abnormality (Here is one example: Fieldmap-EPI not registered in fMRIPrep, but also see below for four different subjects). My goal of analysis is to perform a graph-theoretical analysis, thus I require a connectome as complete as possible. The acquisition parameters are as follows: 2D acquisition on 3T Siemens Prisma, TR/TE = 2500/30 ms, Partial Fourier = 8/8, Flip angle = 80, PE-dir = i, voxel-size = 3.5 mm isotropic. Do you know what that possibly can be, and should I exclude these subjects due to the dropouts?

Best,
Amir

Example one
image
image

Example two
image
image

Example three
image
image

Example four
image
image

2

I’m not a neuro-radiologist but, given that you have some signal in the anatomical images, and a large blowout in a GRE-EPI indicates the presence of something with a very short T2* --probably blood.

Whether to exclude those subjects or not, it’s up to you, but you won’t be able to get any BOLD signal from those areas.