Dose management in CT is an increasingly important issue as the number of CT scans per capita continues to rise. One proposed approach for enhanced dose management is to allow the spatial pattern of x-ray fluence delivered to the patient to change dynamically as the x-ray tube rotates about the patient. The changes in incident fluence could be guided using a patient model and optimization method in order to deliver user-defined image quality criteria while minimizing dose. This approach is referred to as fluence field modulated CT (FFMCT). In this work, a framework and optimization method was developed for evaluating the dose and image quality benefits of FFMCT, both in simulated and experimental data. Modulated fluence profiles were optimized for different objects and image quality criteria using a simulated annealing algorithm. Analysis involved comparing predicted image quality maps and dose outcomes to those using conventional methods. Results indicated that image quality distributions using FFMCT agreed better with prescribed image qualities than conventional techniques allow. Dose reductions ranged depending on the task and object of interest. Simulation studies using a simulated anthropomorphic phantom of the chest suggest an average dose reduction of at least 20% compared to conventional techniques is possible, where local dose reductions may be greater than 60%. Across different imaging tasks and objects, integral dose reductions ranged from 20-50% when compared to a conventional bowtie filter. The results of this study suggest that given a suitable collimator approach, FFMCT could reap significant benefits in terms of reducing dose and optimizing image quality. Though the tradeoff between image quality and imaging dose may not be eliminated, it may be better managed using an FFMCT approach.
| Identifer | oai:union.ndltd.org:TORONTO/oai:tspace.library.utoronto.ca:1807/43470 |
| Date | 07 January 2014 |
| Creators | Bartolac, Steven J. |
| Contributors | Jaffray, David A. |
| Source Sets | University of Toronto |
| Language | en_ca |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.0104 seconds