The derivation of detector efficiencies in PET from single photon blank scans

We developed a means of calculating detector efficiencies from blank transmission scan measurements that use a single photon source. The method is based on a new measurement model that shows how the detector efficiencies contribute to the recorded measurement. Two different approaches to calculate efficiencies are given, where one is based on the maximum likelihood principle. By way of example, blanks acquired from an EXACT3D scanner are processed using these methods. The calculated sets of efficiencies from the two methods are shown to be similar to each other. The variation in the blank caused by detector efficiency is shown to match very well the measurement model when these estimated efficiencies are used. The first application of our approach has been to perform a daily detector check, removing the necessity for a phantom.