On an information-theoretical assessment of PET system design

We treat a PET imaging system as a communication channel and employ the information theory to develop an analytic expression for characterizing its ability to transmit information about the source to the receiver. Based on our result, we propose a particular rate function as a useful measure for evaluating the performance of PET system. We find that this rate function possesses the desirable properties: it predicts that higher sensitivity and higher resolution PET systems perform better, better timing-resolution TOF-PET systems perform better, and TOF PET imaging is better than non-TOF PET imaging.