z-logo
Premium
Application of reconstruction‐based scatter compensation to single‐ and dual‐isotope SPECT imaging
Author(s) -
Kadrmas Dan Joseph
Publication year - 1998
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.598337
Subject(s) - compensation (psychology) , spect imaging , computer science , isotope , dual (grammatical number) , nuclear medicine , medical physics , physics , medicine , psychology , nuclear physics , art , psychoanalysis , literature
The presence of scatter in single‐photon emission computed tomography (SPECT) data can lead to degraded contrast and inaccurate quantitation in reconstructed images. A promising approach to compensate for these effects is to model the process of scatter during the image reconstruction process. Such reconstruction‐based scatter compensation (RBSC) has a number of advantages over subtraction‐based approaches, but difficulty in modeling the complex process of scatter has heretofore limited both the usefulness and evaluation of this approach. In this work, the utility of performing RBSC in single‐ and dual‐isotope SPECT imaging was investigated. Singular value decomposition techniques were used to study some fundamental issues regarding the reconstructibility and noise properties of scatter in SPECT. The results were used to determine energy window schemes and RBSC approaches that led to the best performance for Tc‐99m and Tl‐201 imaging. Model‐based compensation methods were also developed for use with simultaneously acquired Tc‐99m/Tl‐201 data. Finally, fast and practical implementations of RBSC were developed that can reconstruct images with fully 3D scatter compensation in clinically acceptable times. The conclusions of this work indicate that RBSC can be fast and practical, and may be the scatter compensation method of choice for both single‐ and dual‐isotope SPECT imaging.

This content is not available in your region!

Continue researching here.

Having issues? You can contact us here