Dear Audrey,
On Tue, 5 May 2020 15:11:32 +0100, Audrey De Paepe <[log in to unmask]> wrote:
>Hello,
>
>I have a two-fold question pertaining to longitudinal gray matter volume results found using the CAT12 longitudinal processing pipeline (version CAT12.6 (r1450) | 2019-04-04). The study was carried out with patients of neurodegenerative disease over a span of approximately 18 months.
>
>To start, we have noticed that, for all patients, the Total Intracranial Volume (TIV) undergoes a slight change (positive or negative, but mostly negative) between Time 1 and Time 2. That is, in the majority of patients, the TIV is decreasing. What might explain this difference in TIV from Time 1 to Time 2? Could this just be a random error?
TIV should be usually independent from any atrophy. However, due to slight changes in image contrast and bias there might occur TIV changes. My experience is that this is mainly caused be a slightly different skull-stripping result which finally also affects segmentation.
You should additionally check that no changes in your MR sequence occurred between both time points or any other systematic change (e.g. coil change).
>
>Secondly, when we extract a region of interest (specifically, the anterior cingulate cortex) from the scans at Time 1 and the scans at Time 2, we see that approximately one third of the patients undergo an increase in volume in this region from Time 1 to Time 2. (In this analysis, we corrected for TIV using the average TIV from Time 1 and Time 2.) Can this change be interpreted literally as change in volume (mm^3)? Are you aware of any other cases where an increase in volume has been demonstrated in patients with neurodegenerative disease (whether as a protective factor, inflammation, etc.)?
First I would check whether your increase is still present if you don't correct for TIV. It's not strictly necessary to correct for TIV in longitudinal studies and if you notice any systematic TIV changes I would not recommend to correct for TIV in that case.
Best,
Christian
>
>Thank you very much in advance for your time.
>
>Best,
>
>Audrey
|