A modular data acquisition system for high resolution clinical PET scanners

In the last two decades, Positron Emission Tomography (PET) showed to be a key modality to interrogate biology for cancer and cardiac disorders, and to perform molecular imaging, a technology that permits individualized therapy of disease [Weissleder01]. PET's best characteristic is sensitivity: it is the most sensitive tech¬nique for medical molecular imaging, and the whole-body images it produces are unequalled by any other modality [Hoh97, Chae07]. Of course, it lacks anatomical reference, but this is a problem that can be easily overcome with X-ray (CT) (Computed Tomography) integration or coregistration [Lu07]. Moreover, even if PET uses relatively low statistics with respect to other modalities, which generally limits image resolution, it can produce excellent results by using dedicated detector geometries [Humm03]. Two remarkable examples for this kind of system specialization, that we will take as refer¬ence applications, are Positron Emission Mammography (PEM) and in-beam PET (ibPET) for dose delivery monitoring in hadrontherapy. However, the wide range of design possibilities has a counter effect: it re¬quires the development of specifically tailored acquisition systems, whose cost and performances could actually prevent the achievement of the theoretical advantages obtainable with a specialized detector assembly. This doctoral thesis aims at proposing an alternative technological archi¬tecture, able to achieve state of the art PET imaging performances by means of a compact, cost efficient acquisition platform, suitable for its adoption in both PEM and ibPET. Firstly, the state of the art and the controversies that prevent the broad use of dedicated PET in clinics will be explored. Special attention will be paid to the technological solutions and characteristics of previous PEM scanners. It will be also reviewed the current status of ibPET, with particular focus on the performances required to effectively support treatment planning in hadrontherapy. Secondly, a conceptual design solution will be discussed. Various alter-natives will be proposed and justified, with the aim of maximizing detection efficiency and minimize system cost. Through a series of intermediate pro¬totypes the various design choices are implemented and characterized. Thirdly, a final prototype of the acquisition system is designed and im¬plemented. This piece of hardware integrates and extends the solutions that have been validated through the previous systems. The research carried out during this thesis has allowed realizing a state of the art acquisition system that is suitable for specialized PET imaging in the clinical environment of the patient, and that will be used for further research in PEM and ibPET imaging