University of Thessaly (UTH)
Βελτιστοποίηση της λειτουργίας συστημάτων τομογραφίας εκπομπής ποζιτρονίων (PET)
Abstract
dc:descriptionIn the present dissertation, PET/CT (hybrid Positron Emission Tomography and Computed Tomography) systems were modeled using appropriate software. These models are then used to optimize their operation. The optimization of the operation was not only focused on the determination of the technical parameters of the detector and the geometry of the PET system, but also extended to the clinical context, this being the ultimate goal of the thesis. These optimizations are aimed at improving the quantitative results of the functional imaging of radiopharmaceutical uptaking tumors or assuring their repeatability. The anatomy chosen is the thorax and the type of lesion is the Solitary Pulmonary Nodule (SPN). The selection was based on various features and properties such as its small size, as well as the fact that its relative position in the thorax may be related to its prognosis. This practically means that the PET camera is required to operate in such a way, so that the quantitative or semi-quantitative results it produces are reproducible. The first part of the thesis focuses on building the simulation model and listing the corresponding results with emphasis on possible optimization schemes. Our results show that properly tuned Monte Carlo simulations predict accurately the count rates of the detectors, representing all the respective phenomena starting from the initial gamma emission to the allocation of the counts on the respective Lines of Response (LOR) in the sinogrammed raw data. A maximum deviation of 2% when comparing simulations and experiments validates the model accuracy for further use in the clinical context. On this basis, in the second part of the dissertation we use experimentally validated simulations, and the performance of the measurements of four different scanner models of a model of true clinical patient data with detected SPN is represented using the parameter of NECR (Noise Equivalent Count Rate). The response of this parameter is investigated with regards to its activity, its relative location in the thorax and the degree of obesity of patients. Significant fluctuation of the quantitative results is observed when the same patient is scanned in different detector configurations of scanner models, or even in the case of the same detectors where however the scans are taken at different time points and the patient characteristics (weight) have changed in-between. Non-optimal patient positioning eccentric to the isocenter is observed frequently and can potentially contribute further to result variability. Our results provide evidence of the ability for properly tuned simulations to be incorporated into daily clinical practice and provide additional information on the variability of response of PET camera measurements, especially in types of lesions where often borderline decisions are made regarding their characterization. Future work is the exploration of the possibility to use the method prescribed here in an extended clinical trial (e.g. on a country level).
Degree
thesis:*- Grantor dc:publisher
- University of Thessaly (UTH)
- Year dc:date
- 2020
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Theodorakis, Lampros
- Θεοδωράκης, Λάμπρος
Subjects
dc:subject × 8Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/46978