Today, I conduct the longitudinal analysis according to tutorials of freesurfer (https://surfer.nmr.mgh.harvard.edu/fswiki/LongitudinalProcessing). But, when I do recon-all -base <templateid> -tp <tp1id> -tp <tp2id> ... -all, it report the error:

---------------------------------------------------------------------------------------------------------------------------------------------------------------------------

mri_diff --notallow-pix /Users/wangkangcheng/Desktop/long_fs_test/long/s12002_MRI2/mri/rawavg.mgz /Users/wangkangcheng/Desktop/long_fs_test/long/s12002_MRI1/mri/rawavg.mgz \n
Volumes differ in dimension
v1dim 176 256 256 1
v2dim 176 184 200 1
\n*******************************************************************************
WARNING: Image geometries differ across time, maybe due to aquisition changes?
This can potentially bias a longitudinal study! Will continue in 10s.

-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

When I check the raw data and find that the structural data of s12002_MRI1 was the file which was converted by the command of dcm2nii and begin with "co". However, data of s12002_MRI2 was the file which begin with "o". But, when I used the file which begin with "co", they did not have same dimension. How could I solve the problem? (PS: have processed hundreds of data for subject of MRI1)