PT - JOURNAL ARTICLE AU - Karthikayan Balakrishnan AU - Parmeshwar Khurd TI - Monte Carlo study on the impact of timing resolution on PET image quality DP - 2009 May 01 TA - Journal of Nuclear Medicine PG - 1511--1511 VI - 50 IP - supplement 2 4099 - http://jnm.snmjournals.org/content/50/supplement_2/1511.short 4100 - http://jnm.snmjournals.org/content/50/supplement_2/1511.full SO - J Nucl Med2009 May 01; 50 AB - 1511 Objectives The goal of this study was to evaluate the impact of timing resolution on image quality in PET scanners using GATE Monte Carlo simulations of a cylindrical phantom. Changes in contrast-to-noise ratio (CNR) and background variance under various conditions were studied as a function of timing resolution. Methods GATE simulations modeling a Philips Gemini-TF PET scanner with near-perfect timing resolution were done for a cylindrical phantom (300 mm diameter, 98 mm long, 10.9 Bq/mm3 ) with a hot sphere (60 mm diameter, 72 Bq/mm3) insert. Gaussian noise was added to simulated TOF values corresponding to different timing resolution ranging upto 1200 ps. We investigated the effect of both true and scattered coincidences but random coincidences were filtered out from the data. The resulting list-mode files were reconstructed with the clinical Philips reconstruction protocol and the variation in the contrast-to-noise ratio, background variance and spatial contrast was studied. Results As expected, the CNR for trues only data is higher than that with trues and scatters. The spatial contrast was not affected by timing resolution whilethe background variance decreased monotonically with decreasing timingresolution. We observed that CNR increases with decreasing timing resolution and that lowering the timing resolution from 600 to 400 ps increases the CNR by a factor of 1.15 and lowering it from 400 to 200 ps changes CNR gain to 1.21. Conclusions The CNR was observed to increase monotonically with decreasing timingresolution, but less dramatically than simple theoretical predictions.