External Email - Use Caution My opinion (the reviewers of your manuscript may have a different one): In general it is fine to use scans which are 10 years old. But in your case, the problem is that all control subjects were scanned in batch 1 (date: 10y ago/scanner 1/protocol 1/t2-availabilty=no/...) and all case subjects with batch 2 (date: now/scanner 2/protocol 2/t2-availabilty=yes/...).
If half of your scans would be 10 years old (including a random mixture of cases and controls) and the other half new with t2 (also including a random mixture of cases and controls), I guess I would not care *that* much. But this way, there is no way for the statistical analysis to differentiate between batch effect and group effect. Therefore, I doubt that you would be able to publish whatever results you get in a decent journal. Tim -- Dr. Tim Schäfer Postdoc Computational Neuroimaging Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy University Hospital Frankfurt, Goethe University Frankfurt am Main, Germany > On 02/15/2022 6:11 AM Eszter Boros <boros.eszter0...@gmail.com> wrote: > > > External Email - Use Caution > > Thank you. This is very helpful. > And just a follow-up question about this: > Can we use data that old? (And make a comparison to images collected this > year). > I mean is it possible to compare structural scans which were acquired such > a long time ago and with a different scanner using Freesurfer? (e.g., > measuring cortical thickness) > > All the best, > Eszter Boros > > Douglas N. Greve <dgr...@mgh.harvard.edu> ezt írta (időpont: 2022. febr. > 14., H, 10:48): > > > You will definitely get systematic differences between using T2 and not, > > so you could have a false positive (or negative) if you are comparing those > > two groups. On the other hand, you probably changed the T1 protocol a lot > > between now and 10 years ago, so you are likely to have bias either way. > > > > On 2/11/2022 12:53 PM, Eszter Boros wrote: > > > > External Email - Use Caution > > Dear FreeSurfer Team, > > > > We have collected T1 and T2 images in our study and running recon-all with > > T2 gives us much better results (it was really good to see it). > > > > Our aim is to compare these structural data (in Group A) to another > > patient group (Group B). The problem is that the data of Group B was > > collected 10 years ago and we did not have a T2 in that study. > > > > I am wondering what would be the best approach? Maybe we should not use T2 > > in Group A because we do not have a T2 in Group B? Or if we do the manual > > corrections correctly in Group B, this should not be a problem? > > I am using Freesufer 7.1.1 > > > > Thank you for your answer! I greatly appreciate it. > > > > All the best, > > Eszter Boros > > > > _______________________________________________ > > Freesurfer mailing > > listfreesur...@nmr.mgh.harvard.eduhttps://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer > > > > > > _______________________________________________ > > Freesurfer mailing list > > Freesurfer@nmr.mgh.harvard.edu > > https://secure-web.cisco.com/1wVcWXiNn37WyoR-pT_kwIeNB_UwOvN3uE9E1Dd3XP-QqzG1VfvbJCorre40tYedgIwMS32IlMTAt9BHUGqEv9IQPXXlqL3UU47mruB64O6IcdQ58GGunJ52Oz-phlyvJLdiur1H84UxeHSnyk-zTSLYxb5Edck9henxFtj2Unjk7CQolTzoGW5DRtiaFHzvXy85NS8AbPZfgsLlhz2078w8SXAPIYJsCjG2TNl2y7ngYLDybCK42c-SVg32jvk4kRBLTVXgAEkpd3N10M9LmTav07VlJa9AV_jzYFMotyLHw6W1FGlk8Y66gwKWhl6VnVJ3Z8Zc562IoZPE83g6gCw/https%3A%2F%2Fmail.nmr.mgh.harvard.edu%2Fmailman%2Flistinfo%2Ffreesurfer > _______________________________________________ > Freesurfer mailing list > Freesurfer@nmr.mgh.harvard.edu > https://secure-web.cisco.com/1E1Bw_s0DBBQ9GQu4ddAQDpfCiHBC_PImT8rnYQqf9rEV2YS-CO98N-uFjqCgFWMngfqfrJ3xJHrq5vhm9beH0GDqAnA4pANwT0CkKv-9h4yUOo961Wlb--pXWxzrHUuKutnIYGHALdcu2PQn6olWa-FdVu7YGcS39A2G5_W67vl8vflwCKV64dIZpGQNw9jNgQufDtzrsaxN2QjU1K5YxnO2spHnDFXpfHt40vCy_EWafE9__kL2gz3AQxIkjvLmuxcGflZzDJmzAXYR-TVTY9S9ixP6P16XXjTXLNtZqw584FKFm39xDj6guRpUfpuZ/https%3A%2F%2Fmail.nmr.mgh.harvard.edu%2Fmailman%2Flistinfo%2Ffreesurfer _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
