dHCP: Was intensity normalization not included in the dHCP pre-processing framework?

1

Hi everyone,

I am using both the HCP and dHCP dataset, and found a difference in the pre-processing framework between them. While HCP pre-processing framework included intensity normalization, the dHCP framework didn’t. I am curious about the reason for not including this step. Does anyone have any ideas?
Thank you very much in advance!

Kind Regards,
Huiqing

2

Hi Huiqing

We did not do intensity normalisation because we felt that different end users may have different requirements in this respect. When working with the data I personally do a FEAT-style grand mean intensity normalisation that is available in the dHCP fMRI pipeline:

I hope this helps.

Cheers, Sean

3

Hi Sean,

Thanks for your quick reply!
Another question here. Can I apply intensity normalization after doing temporal pass-filter? Is there any influence if I change the order of applying intensity normalization and temporal pass-filter?
Thank you,
Huiqing

4

Hi Huiqing

The intensity normalisation i use is just a scaling by a single global factor and we do it after temporal filtering. I think that is also the order on the FSL FEAT pipeline. I think the order would have a negligible effect, but I have never actually tested it. If you are concerned, you could run some empirical tests.

Cheers, Sean