Four-dimensional image registration

[rubenSaro]

@alexis-roche Hi, I am testing some slice timing correction approaches and a colleague point me to your paper.
I was wondering if it also works with multiband fMRI data, or the algorithm expects single-band data, considering that it is based on an ideal 4D model of the data (if I am not interpreting incorrectly).

Thanks,

Ruben

[alexis-roche]

Hi Ruben, I don't recall testing it with multiband data but, in principle, it should work even better (unless I'm overlooking something). Best,

…

On Thu, Oct 23, 2025 at 1:38 AM Ruben Sanchez-Romero < ***@***.***> wrote: *rubenSaro* created an issue (nipy/nireg#8) <#8> @alexis-roche <https://github.com/alexis-roche> Hi, I am testing some slice timing correction approaches and a colleague point me to your paper. I was wondering if it also works with multiband fMRI data, or the algorithm expects single-band data, considering that it is based on an ideal 4D model of the data (if I am not interpreting incorrectly). Thanks, Ruben — Reply to this email directly, view it on GitHub <#8>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AABZEOPDBCDACP6IY3MQGKT3ZAPQVAVCNFSM6AAAAACJ6OTSLKVHI2DSMVQWIX3LMV43ASLTON2WKOZTGU2DENZVG4ZDENA> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[rubenSaro]

Is this because there are less "effective" slice timing differences?

[alexis-roche]

Yes, that's the intuition - but please consider it as pure speculation.

…

On Thu, Oct 23, 2025 at 4:49 PM Ruben Sanchez-Romero < ***@***.***> wrote: *rubenSaro* left a comment (nipy/nireg#8) <#8 (comment)> Is this because there are less "effective" slice timing differences? — Reply to this email directly, view it on GitHub <#8 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AABZEOP3FXRJQGKWHGUVVQ33ZD2KNAVCNFSM6AAAAACJ6OTSLKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTIMZXG43DENBTGM> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[rubenSaro]

thanks, i will.

[alexis-roche]

Hi Ruben, Honestly, I haven’t worked on this code in a long time, so I can’t be certain of the details. From what I recall, the tool was intended to be used via nireg, and the nipy.algorithms.registration module was likely deprecated. I recommend choosing the more recent version by checking the Git history (git log). Cheers,

…

On Fri, Oct 31, 2025 at 2:00 AM Ruben Sanchez-Romero < ***@***.***> wrote: *rubenSaro* left a comment (nipy/nireg#8) <#8 (comment)> Hi Alexis, following the paper, I am testing the 4Dregistration from: from nipy.algorithms.registration import FmriRealign4d Is this the correct way, or it is better from nireg? I also have seen that the algorithm is also referred as "SpaceTimeRealign" in nipy. Any suggestion? Thanks again. — Reply to this email directly, view it on GitHub <#8 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AABZEOOLUVNTOCVZR3CLGXL32K7C3AVCNFSM6AAAAACJ6OTSLKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTINZRGAYDCMJXGU> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[rubenSaro]

Thanks Alexis.