In light of artifact-induced high variability of activation in fMRI repeat
studies, we developed and tested a clinically useful plaster cast head hold
er (PCH) with improved immobilization, repositioning, and comfort. With PCH
, there were considerably lower levels of translational and rotational head
motion components compared to head fixation with conventional restraining
straps (CRS). Rotational components cannot be fully compensated by realignm
ent and lead to "false activations." In addition, task-correlated head moti
on, which highly increases, the risk of artifacts, was considerably reduced
with PCH, especially in a motion prone subject. Compared with PCH, head mo
tion was 133% larger with CRS in a highly cooperative subject. With a motio
n prone subject, head motion range was increased by 769% (PCH: 0.9 mm, CRS:
7.8 mm), which may indicate the usefulness of PCH for restless patients. I
n functional activation maps, PCH alone yielded fewer residual motion artif
acts than CRS + image registration. Subject tolerance of the head holder du
ring the long measurement times of up to 2.5 hr was good, and slice orienta
tion on different days confirmed the quality of repositioning. Hum. Brain M
apping 12:207-213, 2000. (C) 2000 Wiley-Liss, Inc.