I am trying to identify two distinctive hippocampal layers of the CA1 region, stratum oriens and stratum pyramidale using the T1-weighted MRI. I would like to measure the ratio of the two layers in previously defined region of interest. I've identified the layers according to Hsu et al. (2007) (the paper is attached below).
My question is, whether T1-weighted imaging is sufficient for visualization of those layers? As you can see on the attached image, hippocampal layers are already distinguishable on the original MRIs (fig.a), but I've made them more visible by slightly adjusting the brightness (fig. b). The last figure (fig. c) is actually from the paper I've attached, but, as you may have noticed, it is a diffusion-weighted image, not T1-weighted. Do you think that T1 would do, or DWI would be mandatory in this case? Please, keep in mind that I am only interested in 5th and 6th layer.
Thank you very much in advance.
MRI acquisition characteristics:
- Bruker (7T)
- Repetition time: 1000 ms
- Echo time: 12 ms
- Imaging frequency: 300
- Echo train length: 1
Voxel size is 4.5 × 4.5 × 2 mm3 (CSI Voxel volume 0.8 µL). Histology will also be done, but for the purpose of individual cell quantification (cells count, diameter, perikaryon area, nucleus & nucleolus area etc.). I am aware that this whole layers quantification could be assessed hystologically, but I am curious if T1 MRI would also do.
Dear Valentin,
that was exactly what I did, I enhanced the contrast using a simple in-built extension for image editing (integrated in 3D Slicer), in order to make the strata more distinguishable. It seems that the results of the surface area assessment are in accordance with the results gained by IHC.
However, MRI results seem to be more precise, I would say, because you can measure the whole volume of the layer you interested in (throughout the whole hippocampus), whereas histological slices allow only for surface area measurement.
4.5 x 4.5 x 2mm3 are very coarse voxels. Or, do you mean 450microns x 450 microns x 2mm?
In human T1w data, anything beyond 1mm3 is too coarse to measure hippocampal layers. Obviously, the higher the voxel resolution, the more clearly the layers can be visualized, as long as your signal does not drop too much.
The latest version of FreeSurfer (version 6) relies of voxels being around 1mm3, and then it super-resolves the data to 0.33mm3 resolution, and then applies a hippocampal subfield template. If you search the FreeSurfer forum, you will see that there has been discussion on ways to extract hippocampal layers/subfields in rats from T1w data. Sometimes T2w data can help as well.
Dear Jerome, I apologize, I mistyped the numbers,
I've checked and it says 0.45 x 0.45 x 2 mm³.
I am working with 3D Slicer, I don't have any experience in FreeSurfer. Is FreeSurfer a freeware software?
FreeSurfer is free software and quite easy to install. Just follow the instructions on the wiki, or follow the instructions in my book (http://jeromemallershandbookofstructuralbrainmrianalysis.yolasite.com/).
- Badges
- Science topic
- Similar topics
- Gels