Keith S. Cover1, Ronald A. van Schijndel2,
Bob W. van Dijk3, Alberto Redolfi4, Dirk L. Knol5,
Giovanni B. Frisoni4, Frederik Barkhof2, Hugo Vrenken2,6
1Physics & Medical
Technology, VU University Medical Center, Amsterdam, North Holland,
Netherlands; 2Department of Radiology, VU University Medical
Center, Amsterdam, Netherlands; 3Department of Physics &
Medical Technology, VU University Medical Center, Amsterdam, Netherlands; 4Laboratory
of Epidemiology & Neuroimaging, IRCCS San Giovanni di Dio
Fatebenefratelli, Brescia, Italy; 5Department of Epidemiology
& Biostatistics, VU University Medical Center, Amsterdam, Netherlands; 6MS
Center Amsterdam & Alzheimer Center, VU University Medical Center,
Amsterdam, Netherlands
The back-to-back (BTB) reproducibility at both 3T and 1.5T of the Siena/FSL brain atrophy measurement algorithm was assessed. The BTB MPRAGEs at both 3T and 1.5T of 118 subjects, including MCI and AD patients, were acquired as part of the Alzheimer's Disease Neuroimaging Initiative (ADNI). A bootstrap simulation based on the Wilcoxon-Mann-Whitney test was used to estimate the group size required to detect a specified reduction in the disease progression. The 3T group sizes were consistently roughly 50% larger than those required by the 1.5T MPRAGEs. This preliminary result warrants further investigation into the underlying causes.