Hi,
It isn't possible on the basis of just these numbers to determine which is the best result.
It is true that the asymmetry that you are seeing here is unusually large, and that this might well indicate problems with the segmentation. However, the only way to really know is to look at your images and determine, by examining them, where
the hippocampal segmentations are good or not.
So have a good look at the images and then let us know if you are still not sure.
All the best,
Mark
Dear Expert,
We have scanned three subjects with different T1 sequences and performed sub cortical volume analysis using first_run_all script & fslstats.
Given below is the details of hippocampal volume for different T1 sequences.
Subject-1
fslinfo P1_T1.nii.gz
data_type FLOAT32 175x256x256 datatype 16 0.9x1x1
cal_max 0.0000 cal_min 0.0000 file_type NIFTI-1+
Hippocampal volume (output from terminal): Lt 3702 3701.971756 Rt 4090 4089.968796
In this case number of voxels and volumes are similar.
Subject-2; fslinfo P5_T1.nii.gz
data_type FLOAT32 301x240x240 datatype 16 0.599x1.04x1.04
cal_max 0.0000 cal_min 0.0000 file_type NIFTI-1+
Hipocampal volume: Lt 4583 2983.678203 Rt 3249 2115.201938
In this case number of voxels and volumes are different.
Subject-3a
fslinfo rb1.nii.gz
data_type INT16 dim 175x256x256 datatype 4 pixdim 0.99x1x1
cal_max 0.0000 cal_min 0.0000 file_type NIFTI-1+
Hippocampal volume: Lt 4732 4731.963898 Rt 4537 4536.965385
In this case unmber of voxels and volumes are similar.
Subject-3b;
fslinfo rb2x2.nii.gz
data_type FLOAT32 dim 176x128x128 datatype 16 pixdim 0.99x2x2
cal_max 0.0000 cal_min 0.0000 file_type NIFTI-1+
Hipocamal volume: Lt 755 3019.953918 Rt 862 3447.947388
In this case number of voxels and volumes are different.
Could expert explain me which sequence is the best for such kind of analysis particularly voxel based morphometric analysis?
Thanks in advance
Ramesh
