|
Descripción
|
|
|---|---|
| High-resolution (HR) time-of-flight (TOF) positron emission tomography (PET) enhances the performance of current PET systems to provide images of subtle molecular processes in the body. Thanks to its TOF feature, the quality of the images reconstructed in a HR TOF-PET system improves with respect to non-TOF-PET systems. Additionally, novel HR TOF-PET systems provide HR in three domains: time, energy and space. In as much as the time and energy information of the ?-events is precise, the noise level in the reconstructed image is reduced by accurate ?-events selection. Furthermore, fine spatial information leads to the detection of smaller tracer concentrations. In oncology, currently the main application of PET scanners, HR TOF-PET systems will be able to locate millimetric tumors and so greatly improve early cancer detection. HR TOF-PET has only been available recently thanks to the latest advances in the various instrumentation technologies that conform a PET detector. In this scenario, the EndoTOFPET-US project was created with the aim of developing an endoscopic HR TOF-PET scanner to provide early diagnostics for pancreatic and prostatic cancers. The present thesis focuses in the design and development of a high-performance data acquisition system (DAQ) system for its application in HR TOF-PET scanners, like the Endo-TOFPET-US. The HR properties in time, energy, and space, of HR TOF-PET scanners convey important implications in the design of the associated DAQ system: high channel integration, high data volume, and the requirement of efficient data processing. Therefore, the thesis proposes a DAQ architecture adapted to these characteristics, details the online processing techniques that have been implemented, and shows the results obtained making use of the resulting DAQ system. The outcome of this work is a flexible and high-performance HR TOF-PET DAQ system, which can be applied not only in the EndoTOFPET-US scanner but in many HR TOF-PET systems with different geometries built following a modular assembly scheme of DAQ elements and detector modules. For the control and readout of the detector form the DAQ server, a low-level communication protocol has been implemented that reduces the number of physical links, optimizes the data integrity, and accelerates the DAQ data processing. Additionally, two online processing techniques for the readout of the ?-events have been implemented according to the observed system needs: a front-end energy discrimination module and a time-sorting module. The energy filter reduces the system bandwidth needs and improves the final image quality by rejection of the detected ?-events that lead to an increment of the image noise level. On the other hand, the time-sorting module reduces the memory requirements in the digital electronics thanks to the implementation of a novel ?-events memory storage approach that takes into account the physics of the ?-photons detection process in a HR TOF-PET scenario. In the light of the results exposed in the present document, it is proven how the proposed DAQ is able to control and readout HR TOF-PET instrumentation while rejecting on-the-fly undesired data. Thus, it is expected that the contributions of this thesis in DAQ HR TOF-PET systems, when applied to the EndoTOFPET-US scanner, do their bit to help conducting the preclinical trials of the first endoscopic HR TOF-PET system. | |
|
Internacional
|
Si |
|
ISBN
|
|
|
Tipo de Tesis
|
Doctoral |
|
Calificación
|
Sobresaliente cum laude |
|
Fecha
|
28/09/2018 |