A method is developed for pre-processing projection images for a SPECT brain imaging system. The projection images are recorded by modular gamma cameras that exhibit noisy response before processing. The image acquisition process is modeled so that the mean of the detected gamma-ray emissions is a linear transformation of the actual flux. Two models for detection are examined, one based on independent Poisson distributions and the other based on a multivariate distribution. The Expectation Maximization (EM) algorithm is used to invert the forward model to obtain a Maximum Likelihood estimate of the flux. Simulations using uniform, Gaussian and point flux patterns demonstrated that EM processing recovered improved estimates of these patterns. Processing measured images yielded improved estimates, but also revealed that both forward models are incomplete.
| Identifer | oai:union.ndltd.org:arizona.edu/oai:arizona.openrepository.com:10150/278351 |
| Date | January 1993 |
| Creators | Marcotte, Hope Ann, 1964- |
| Contributors | Barrett, Harrison H. |
| Publisher | The University of Arizona. |
| Source Sets | University of Arizona |
| Language | en_US |
| Detected Language | English |
| Type | text, Thesis-Reproduction (electronic) |
| Rights | Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author. |
Page generated in 0.0014 seconds