Recent interest in hypothesis testing on functional imaging data has s
purred the development of several statistical techniques. The purpose
of this paper is to provide a method to reduce the computational inten
sity associated with randomization tests of positron emission tomograp
hy imaging data. We discuss the advantages and disadvantages of tradit
ional distributional hypothesis testing versus the advantages and disa
dvantages of randomization tests. A method for reducing the computatio
nal intensity of randomization uses a conjunction of updating and sequ
enching and results in Significantly reduced processing. The running t
imes of randomization methods are compared. (C) 1998 Academic Press.