Tailoring PET time coincidence window using CT morphological information

In this paper we propose a general method to reduce the electronic time coincidence window in PET, based on a-priori morphological information obtained via a CT scan. A different time coincidence window is assigned to each line of response based on the CT p-map. A minimum and a maximum time-of-flight are computed based on the position of the patient in the field of view, and data are rejected if their time-of-flight is not compatible with the accepted range. This allows us to reduce the random fraction in the data, improving image quality and true data throughput. Experimental results based on phantom and patient studies show a random fraction decrease of about 70% and an image signal to noise improvement between 2% and 7%.