Marisa,

Because no controls were scanned at sites 2-4, its hard to assess whether the difference in these scans is due to scanner or due to the disease/disorder. You can either drop these subjects or collect a similar number of controls at the additional sites.

Best Regards,

Donald McLaren, PhD

On Mon, Aug 10, 2015 at 6:18 AM, Marisa Loitfelder <[log in to unmask]> wrote:

Dear experts,

We performed a multi-centre study and acquired T1-weighted scans from 180 controls and 180 patients. All controls and 78 patients were scanned at site 1, 24 patients in site 2, 48 in site 3, and 30 in site 4. We want to analyse grey matter volume in different networks, but find a bias for “centre” or "scanner", respectively. All scanners are 3T. What could be the cause and how can I correct for that? Any ideas are welcome.

Many Thanks,
Marisa