Hi Zoë,

13 years is a lot. I would recommend to process everything cross sectionally and then look at a few cases with different age deltas (including the one with 13 years) using the longitudinal stream. If most of these work, you can run the rest through the longitudinal pipeline.
Also, most likely your scanner hardware and definitely the software changed during 13 years, so images will look rather different and it will be difficult to attribute reported measurements to scanner, motion or real effects. If you have several time points per subject you can maybe model a scanner/hardware (head coil, etc) effect in the statistics. Also you should try to check motion levels longitudinally and cross sectionally (if you have groups) via QC or fMRI/diffusion.

Best, Martin


On 05/10/2016 10:46 AM, Hawks, Zoe wrote:

Hi Martin,


Thank you so much for your reply! Reviewing our data, it looks like our age range is 6-23 rather than 6-18 as I initially specified. Within an individual, the largest delta is 13 years. Based on your experience, are these ranges likely to be too large for the longitudinal stream? Would you suggest we process longitudinally as a first-pass regardless? Thank you!


Best,


Zoë


From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of Martin Reuter <mreuter@nmr.mgh.harvard.edu>
Sent: Monday, May 9, 2016 2:15:27 PM
To: Freesurfer support list
Subject: Re: [Freesurfer] Pediatric longitudinal processing
 
Hi Zoe,

there are basically two approaches if the assumption of fixed head size is violated:
- use the longitudinal stream and see how far it get's you. Especially if the time delta is relatively small, this has good chances to work. It may require edits, so definitely check the results carefully.
- if surface initialization in the long is too far (that is, if the surfaces created in the base are too far away from their final position), there may not be much you can do with editing. If that happens, you may need to drop longitudinal processing and simply use the cross sectionally (independently) processed data. Since that data is the first step in the longitudinal pipeline, it is already there.

Best, Martin



On 05/09/2016 02:57 PM, Hawks, Zoe wrote:

I'm looking to use FS for longitudinal processing in a pediatric population ranging from 6-18 years. From what I've read, the longitudinal stream isn't appropriate because our data violate the assumption of fixed head size. Any advice on how to account for increases in head size would be appreciated. Thank you!



_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

-- 
Martin Reuter, PhD
Assistant Professor of Radiology, Harvard Medical School
Assistant Professor of Neurology, Harvard Medical School
A.A.Martinos Center for Biomedical Imaging
Massachusetts General Hospital
Research Affiliate, CSAIL, MIT
Phone: +1-617-724-5652
Web  : http://reuter.mit.edu 


_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

-- 
Martin Reuter, PhD
Assistant Professor of Radiology, Harvard Medical School
Assistant Professor of Neurology, Harvard Medical School
A.A.Martinos Center for Biomedical Imaging
Massachusetts General Hospital
Research Affiliate, CSAIL, MIT
Phone: +1-617-724-5652
Web  : http://reuter.mit.edu