نبذة مختصرة : N° d'ordre : DU 1740, EDSPIC : 372 ; Positron emission tomographs (PET) are fitted with highly capable reading electronics, which owns qualities and drawbacks. Highly accused at first as an explanation of the poor sensitivity of these imagers, the present study points out that the dead-times shared along the chains contribute to only 16% in the data losses at a typical activity of 10 microCi/ml. The gross acquisition rates could thus be raised by 20% through a suppression of these saturations. Looking in details at the philosophy of the acquisition procedure, a property appears to circumscribe the sensitivity even more : the timing resolution. The latter conditions, to the first order, the rejection capabilities of randoms events, part of the scattered ones and hence noise which is finally ratioed to the true coincidences the signal is made up of. Minimizing the resolving time goes through the suppression of the unneeded actors along with the adoption of a well adapted time-stamping method (optimal filtering). In doing so, the intrinsic channel resolution appears to be possibly lowered by a factor 7, reducing to 350 ps. The bottom value of the coincidence window may be narrowed as a consequence, leading to an increase of the NECR by 50%. At this stage, a time of flight (TOF) algorithms can be implemented. As an opportunist, it promises a reduction of the noise variance by 430%, a gain that echos on the NECR figure. Finally merging all these ideas allows to expect an improvement close to an order of magnitude on the NECR, with the hope of routine exams shortened by the same amount. In this context, it appeared logical to imagine a new electronics acquisition synoptic dedicated to fully pixelated PET. The number of channels blows up by the way when compared to the existing, this statement being partially balanced by the decision to fully integrate the electronics. The measures of the energy and time are planned to be performed with a single channel, with a continuous and dead-timeless reading of the events. The ...
No Comments.