1: \begin{abstract}

2: The use of Time Of Flight (TOF) in Positron Emission Tomography (PET)

3: is expected to reduce noise on images, thanks to the additional information.

4: In clinical routine, a common reconstruction approach is the use of

5: maximum likelihood expectation maximization (MLEM) stopped after few

6: iterations. Empirically it was reported that, at matched number of

7: iterations, the introduction of TOF increases noise. In this work

8: we revise the theory describing the signal and noise convergence in

9: MLEM, and we adapt it to describe the TOF impact on early stopped

10: MLEM. We validated theoretical results using both computer simulations

11: and phantom measurements, performed on scanners with different coincidence

12: timing resolutions. This work provides theoretical support for the

13: empirically observed noise increase introduced by TOF. Conversely,

14: it shows that TOF not only improves signal convergence but also makes

15: it less dependent on the activity distribution in the field of view.

16: We then propose a strategy to determine stopping criteria for TOF-MLEM,

17: which reduces the number of iterations by a factor proportional to

18: the coincidence timing resolution. We prove that this criteria succeeds

19: in markedly reducing noise, while improving signal recovery robustness

20: as it provides a level of contrast recovery which is independent from

21: the object dimension and from the activity distribution of the background.

22: \end{abstract}