sdm-help-list > t-map combination
Showing 1-10 of 10 posts
Display:
Results per page:
Jun 19, 2020  12:06 PM | Brandon Gunasekera
t-map combination
Hi,

I am running an image based meta-analysis.

For those researchers who kindly agreed to share their t-maps, we have received 2 t-maps each for each set of contrasts (one for drug>placebo and another drug

Therefore, within SDM, should we be first combining these maps before loading for meta-analysis or should we use them as separate?

Also in case we have to combine, what method would you suggest is most appropriate, and is the procedure different for .nii images as opposed to .hdr/img images?

I have examples which I am more than happy to email across.

Best wishes,
Brandon
Jun 19, 2020  05:06 PM | Lydia Fortea - Instituto de Investigaciones Biomédicas August Pi i Sunyer (IDIBAPS)
RE: t-map combination
Hi Brandon,

The best to do is to combine these t-maps before launching the meta-analysis . The procedure is exactly the same for both images, .nii and .hdr/.img; so you can choose whatever you prefer

Kind regards,
Lydia
Jun 19, 2020  05:06 PM | Brandon Gunasekera
RE: t-map combination
Hi Lydia,

Many thanks for your response. 

Would you suggest merging these maps using the SDM 'combine images' option, or through some other method?

Best wishes,
Brandon
Jun 20, 2020  09:06 AM | Lydia Fortea - Instituto de Investigaciones Biomédicas August Pi i Sunyer (IDIBAPS)
RE: t-map combination
Hi Brand, 

Yes, indeed. You should use the option "combine imagen" of the SDM.

Kind regards,
 Lydia
Jul 13, 2020  12:07 PM | Anna Fiorito
RE: t-map combination
Hi Brandon and Lydia, 

I am also running an image based meta-analysis. In my meta-analysis I also want to first combine images. However, I am wondering if the new version of SDM still allow combining images (in my understanding this option was not yet implemented). 

Thank you for clarifying.
Best wishes, 

Anna
Jul 14, 2020  07:07 AM | Lydia Fortea - Instituto de Investigaciones Biomédicas August Pi i Sunyer (IDIBAPS)
RE: t-map combination
Hi Anna, 

Your are right.. Unfortunately the new version of SDM has not available this option yet. We are working on it to make it possible as soon as we can.

My apologizes,

Lydia
Jul 14, 2020  01:07 PM | Anna Fiorito
RE: t-map combination
Hi Lydia,

Thank you for your answer. From the previous answer on the forum, this was also my understanding. However, since I really hope to combine contrasts, I just downloaded the old version of SDM (Sdm 4.142) and I tried to combine two contrasts of the same study/same subjects/same task (fear > neutral and angry > neutral, since my meta-analysis is on negative emotion > neutral contrast and some authors provide me with the contrast each emotion separately > neutral). Using 'Combine images' option I obtain the same map from Sdm 4.142 and from the latest version of SDM (Sdm-Psi 6.21).

I am now wondering if I can use the obtained maps to run the main meta-analysis and whether this means that 'Combine images' option actually works on the latest version of SDM.

Thank you for your help,
Anna
Jul 14, 2020  02:07 PM | Lydia Fortea - Instituto de Investigaciones Biomédicas August Pi i Sunyer (IDIBAPS)
RE: t-map combination
Hi Anna, 

We have to update the option for the new version. I'm sorry to say that wouldn't be a good approach to use the combined image obtained from SDM 4.142 for an analysis with SDM 6.21.

Kind regards
Lydia
Jul 20, 2020  12:07 PM | Brandon Gunasekera
RE: t-map combination
Hi Anna & Lydia,

Could you send across the link to download SDM 4.142

Best wishes,
Brandon
Jul 20, 2020  04:07 PM | Anna Fiorito
RE: t-map combination
Hi Brandon,

Here (https://www.sdmproject.com/old/) you will find the link to download the previous version of SDM (which is SDM 5.142, my apologies for the mistake).
However, as Lydia pointed out in this previous answer (https://www.nitrc.org/forum/message.php?...), you should consider the fact that SDM-PSI is much more robust than AES-SDM.

Best wishes,
Anna