FMRIprep multi-echo

Hi All,

I am using fMRIprep version 1.5.0rc1 to preprocess some multi-echo data and I noticed that fMRIprep automatically combines the echos. Is there a way to run the preprocessing steps without combining the echos?
Or should I use one of the earlier versions instead?

Thanks so much!


Hi @jngx47f,

If you’d like to process the echos individually, check out the --echo-idx flag !

That being said, if you’re using the individual echos to submit to e.g. tedana you might want to check out Combining fmriprep with ME-ICA from tedana

Hope that helps !


I will try that. Thank you so much!

I want to keep the echos separate to calculate realised volatility. I have a feeling I’m dealing with some particularly rough data.


I am also working with multi echo data, (new to multi echo data and fMRIPrep - loving fMRIPrep so far, so thank you!) and I was wondering how I can check that fMRIPrep is correctly combining the echos? The .html documentation does not make any indication of this step occurring. Is fMRIPrep calling on ME-ICA to combine and, if not, how is it combing them?

*If it does not automatically combing the multi echos, is there an available option for me to flag in order to combine them?

(*I’m running fMRIPrep-1-4-1 with singularity)

*clarification: we are collecting multi band, multi echo BOLD data

fMRIPrep uses tedana for optimal combination starting in v1.1.7, I believe (per poldracklab/fmriprep#1263). You could dig into the code for t2smap_workflow if you want.

On a semi-related note, there has been a lot of discussion among the tedana and fMRIPrep folks about disabling t2s_coreg (see here). You might want to avoid using --t2s-coreg when processing your data. At least for now.


Thank you so much for your response. And thanks for the note on --t2s-coreg, we had started looking into it so this was quite helpful!