Effective scatter correction techniques are required to account for er
rors due the high scatter fraction seen in positron volume imaging (PV
I). To be effective, the correction techniques must be accurate and pr
actical, but they also must not add excessively to the statistical noi
se in the image. We have investigated the noise added by three correct
ion methods: a convolution/subtraction method; a method that interpola
tes the scatter from the events outside the object; and a dual energy
window method with and without smoothing of the scatter estimate. The
methods were applied to data generated by Monte Carlo simulation to de
termine their effect on the variance of the corrected projections. The
convolution and interpolation methods did not add significantly to th
e variance. The dual energy window subtraction method without smoothin
g increased the variance by a factor of more than twelve, but this fac
tor was improved to 1.2 by smoothing the scatter estimate.