Publications Repository - Helmholtz-Zentrum Dresden-Rossendorf

2 Publications

A Comparison of CT-Based Attenuation Correction Strategies for PET Data of Moving Structures

Richter, C.; Hoinkis, C.; Just, U.; Pönisch, F.; Woithe, J.; Enghardt, W.

Respiratory motion introduces image artefacts not only in standard 3D-CT but also in 3D-PET images due to two reasons: (a) Smearing of the activity concentration and (b) an incorrect attenuation correction. In 4D-PET the effect of smearing becomes negligible but the influence of incorrect attenuation correction remains important.
To investigate the quantitative influence of attenuation correction on both PET acquisition methods (3D- and 4D-PET), a comprehensive phantom study was performed using a respiratory motion mimicking phantom on a dedicated Siemens Biograph 16 PET/CT, which had the extended capability of acquiring 4D-PET and 4D-CT data. The used respiratory motion phantom is able to simulate typical lung tumor motion in two dimensions with two possible patterns of respiration. The 3D- and 4D PET data sets were corrected with different CT attenuation data, namely a standard 3D-CT (pitch 1.5), a slow 3D-CT (pitch 0.5), an ultraslow 3D-CT (pitch 0.15), and also an average CT and a maximum intensity projection, both calculated from a 4D-CT (pitch 0.1). Additionally, the 4D-PET was corrected phase-wise with the 4D-CT (phase-correlated attenuation correction). For that purpose the synchronization between 4D-PET and 4D-CT has been verified. The reconstructed PET images were analyzed concerning the reconstructed volume, the activity concentration and the full width half maximum (FWHM) of the activity distribution in the direction of the highest phantom movement. Additionally, the motion amplitude of the phantom was obtained from the 4D-PET data sets.
Our results suggest that attenuation correction of 3D-PET data should be performed with a slow CT. However, the 4D PET data should be reconstructed using phase-correlated attenuation correction with a 4D-CT.

  • Contribution to proceedings
    IEEE Dresden 2008, 19.-25.10.2008, Dresden, Germany
    IEEE Conference Report
  • Poster
    IEEE Dresden 2008, 19.-25.10.2008, Dresden, Germany

Permalink: https://www.hzdr.de/publications/Publ-11903
Publ.-Id: 11903